Some things have to be believed to be seen.
—Ralph Hodgson, author and poet1
Most worthwhile endeavors, be it mastering a new programming language, writing a book, or starting a successful micro-ISV, have at least two mandatory components: a great deal of work and the belief you can do it. In Chapters 1 through 6 I covered just about all the tasks you need to master in order to make your micro-ISV a success; in this chapter, I want to get at the other component—believing you can do it.
Now, I could have you chant "I can do it! I will do it! I am doing it!" a couple hundred times each morning. That works—except your co-workers might talk. Or I could send you off on a "Discover Your Inner Strength" seminar and charge you up that way. That would also work.
Being an old "just the facts" type of ex-reporter, I decided the best way to show you that you can build a successful micro-ISV was to go out and interview 25 micro-ISVers who are making money—sometimes a lot of money. Seeing is believing.
I conducted the following interviews in September 2005, so by the time you're reading this, things may have changed for each of the micro-ISVers who were kind enough to let me interview them via email. I've organized the interviews into three groups:
Most of these interviews started with the same set of questions and then spun out from there, so the juicy bits tend to be toward the end of each interview. Please excuse the length of this chapter; in some ways this is the most important chapter in this book. Each of these micro-ISVs has useful lessons to pass on, but here's the most important take-away: if they can do it, so can you!
__________
The first group of interviewees is on the first furlong of the micro-ISV race and is just starting to see, or will soon see, revenue. As you read these nine interviews, you'll see a couple of common themes emerge:
Of all these interviews, I have to confess to being most impressed with Ian Landsman's studied, careful, and meticulous approach. Since I conducted that interview in September 2005, Ian's product HelpSpot has gone live and by all accounts is doing very well.
ANDY BRICE, FOUNDER, ORYX DIGITAL
Micro-ISV: Oryx Digital
Web site: http://www.perfecttableplan.com
Blog: "Not yet"
Interviewee: Andy Brice, founder
What it sells: PerfectTablePlan, a Windows and Macintosh software application for planning seating arrangements for weddings, conferences, and other events.
Location: High Wycombe, England (about 30 minutes west of London)
Q. Can you tell me a bit about the software or Web services or products your micro-ISV sells?
A. I sell PerfectTablePlan, desktop software for doing table/seating plans for wedding receptions, corporate events, charity dinners, etc. Currently it is available for Windows, and I am just about to release a Mac version.
I got married last year, and I found doing the seating plan for the wedding reception a real nightmare. And we only had 60 guests. I looked around for software to help me, but I just couldn't find anything appropriate. There were a handful of products, but they were too expensive, [they were] too difficult to use, and I didn't think they were very well marketed. I realized there was a gap in the market, so when I was made redundant from my dot-com job, I took the plunge.
Particular interests of mine include usability, optimization, and cross-platform development, and PerfectTablePlan was interesting because I was able to combine all three. (It uses a genetic algorithm to automatically seat guests according to who you want them to sit next to.)
So far feedback from customers has been very positive. The main problem is that most people don't even realize that there is a better way to do a seating plan than using scraps of paper or cutting and pasting into a spreadsheet. I have to try [to] educate the market with a minimal marketing budget.
Q. Is revenue a) less than what you hoped, b) about what you predicted, c) exceeding expectations, or d) you're going to buy a small Pacific island later this year? Can you share your sales per month?
A. When I started, I had little idea how many I would sell and through what channels. But I divided my living costs by the expected margin for each sale, and I thought it was achievable. As a target I drew a graph of monthly revenue starting at £0 in month 1 and the equivalent of my salary as a full-time software engineer in month 12. Sales have increased linearly every month, and so far I have reached almost every monthly revenue target, with the help of some consultancy in the first few months.
However, I should stress that this revenue goes to the company, not to me.
I have paid myself very little so far for working very long hours and have mostly been living on savings. Thankfully, I have a very supportive wife (thanks, Claire!). If money is your primary motivation, I wouldn't recommend micro-ISV as a career path.
Q. What has been the biggest surprise in getting your micro-ISV started?
A. I've had a few surprises. It is a process of discovery, and that is part of what makes it interesting. I've learnt so much in the last eight months!
I had assumed that my Internet sales would be four in the United States for every one I made in the United Kingdom, but so far it has been the other way around. I think this is partly because many Americans (especially from the Midwest) don't do table plans for their weddings. They regard it as too formal. I guess I should have done more research, rather than making assumptions.
I thought the obvious sales channel for PerfectTablePlan was department stores with wedding gift lists. I managed to interest one but was flabbergasted by the deal that was suggested. I calculated that if they sold 1,000 units of PerfectTablePlan, I would owe them £1000. And this didn't even include my support, shipping, and printing costs. I felt that this was more of an insult to my intelligence than a business proposal.
Q. Do you have a marketing plan?
A. Yes. I thought quite a bit about how I was going to market PerfectTablePlan before I wrote the software. It's mostly in my head, rather than written down, though. It also evolves as I find out more about the market and what works and what doesn't.
I have quite a few ideas on how to make PerfectTablePlan more useful to customers, how to broaden the market, and related services I can offer. But I obviously don't want to say too much about future strategy.
Q. In what ways do you market PerfectTablePlan?
A. Mainly Google AdWords, search engines, and magazine ads.
AdWords is great because you can get really quick and detailed feedback on what works and what doesn't. Also, you can get it up and running in a few hours. However, it does require a lot of tweaking, and I still end up paying for click-throughs from people who are looking for seating plans for Yankee Stadium or Boeing 747s.
I found that it took months of waiting and quite a bit of tweaking of my Web pages to appear anywhere on the major search engines. But the effort was worth it, as I now get quite a lot of traffic that costs me nothing. I didn't pay anyone to do search engine optimization for me—all you need to know is out there on the Web.
Magazine ads probably don't make sense for a lot of micro-ISVs. However, I am selling to specific niches in the consumer market that are well served by magazines. Also, some of my customers might not think to surf the Web, and I need some way to reach them. Unfortunately, it's difficult to measure the success of print ads. I am trying to use a different URL for each ad so I can monitor the response.
PerfectTablePlan by Oryx Digital
Q. Given your experience so far, are you planning a second product?
A. Yes. I already have a basic prototype that I am using to help me with my work. Like PerfectTablePlan, it's a tool that solves a well-defined problem. However, I am unlikely to do much beyond developing it for my own use in the next few months—I am too busy with PerfectTablePlan.
ANDREY BUTOV, PRESIDENT, ANTAIR
Web site: http://www.antair.com
Blog: http://www.antair.com/andrey
Interviewee: Andrey Butov
What it sells: Chinchilla, a developer product.
Location: New York City, New York; United States
Q. Can you tell me a bit about the software or Web services or products your micro-ISV sells?
A. Antair's one and only product right now is Chinchilla. It is a desktop application geared toward .NET developers [that] allows them to extract class inheritance diagrams from within .dll and .exe .NET assemblies. There is an educational product in the works and a Mac/Linux port of Chinchilla for the Mono project.
Q. Is revenue a) less than what you hoped, b) about what you predicted, c) exceeding expectations, or d) you're going to buy a small Pacific island later this year? Can you share your sales per month?
A. Considering that Chinchilla was literally released last week and is only in the beta release program stage, I would say a for actual sales but b for expectations. I really don't know what to expect but am optimistic. I am looking into marketing options but expect later products to generate more revenue than Chinchilla has the potential for. No sales as of yet.
Q. What has been the biggest surprise in getting your micro-ISV started?
A. How much noncoding work there is. I had no idea how much effort it takes to take care of all the infrastructure and administrative work surrounding the launch of a business…even a one-man operation. From incorporation to bank accounts to marketing, I find myself going deeper and deeper into areas I was not trained to deal with.
Q. Why a chinchilla?
A. Chinchilla was chosen primarily for acoustic aesthetics. Antair is completely meaningless. I wanted the company name to stand on its own…meaning the word would have no meaning associated with it aside from the company. Chinchilla sounded well when spoken after Antair.
It doesn't hurt that the creatures are cute. I knew the resulting logo would come out nicely…and it did. I am much happier with Chinchilla's logo than with Antair's logo.
Q. What's your approach to marketing to developers?
A. I am warier of marketing than of any other noncoding aspect of the business. With all other administrative and business foundation matters, one can get advice from other start-ups, message boards, or people who own non-IT businesses. Marketing is the only thing that is not only product specific but also directly hits your pocketbook. If you make a mistake in marketing, you not only lose money in the attempt but you also lose time, which in a one-man operation is the most limited resource.
I intend on doing a lot of research, trying small, noncostly avenues of advertising and hoping that word of mouth spreads. From what I've read thus far, word of mouth brings more sales than any other form of advertising.
Q. Do you see a growing (little? lot?) Mac/Linux market via Mono?
A. I think as far as percentage of sales with respect to downloads/visits, the Mac market should be higher. The Apple community has a way of standing behind its developers. I find myself spending more on software for the Mac, even though my primary machines are Solaris and Windows. Visual Studio 2005 offers a similar feature to Chinchilla. Chinchilla is meant to be an affordable, stand-alone replacement. Since there is no Visual Studio on a Mac, it offers a more appropriate market. I think Mono could gain ground on the Mac platform, but the lack of a Windows Forms implementation might just prevent that. Gtk is just not satisfactory when you're writing .NET code. After many years in the IT industry, I am no longer naive enough to believe that there is a market for this in the Linux community. I don't think Mono will ever catch on in the UNIX derivative market.
Chinchilla by Antair
BORIS YANKOV, FOUNDER/OWNER/CEO, VIRTUOZA
Web site: http://www.virtuoza.com
Blog: http://www.borisyankov.com
Interviewee: Boris Yankov
What it sells: A task management application, a Windows start-up manager, and a time management application.
Location: Bulgaria
Q. Do you have one Web site or several?
A. My main Web site is http://www.virtuoza.com
. I had my products listed there, but now they all have their own Web sites. They are
http://www.get-in-control.com
http://www.smarttodo.com
http://www.overspy.com
Q. Can you tell me a bit about the software or Web services or products your micro-ISV sells?
A. I wish I could say I had some great strategy defined before creating my applications. It isn't so. InControl is a start-up manager. I created it back in 2001. Smart To-Do is a lightweight task management application. And OverSpy is report-remotely, computer-monitoring software.
I will not advise anyone to create several applications differing that much.
Q. Is revenue a) less than what you hoped, b) about what you predicted, c) exceeding expectations, or d) you're going to buy a small Pacific island later this year? Can you share your sales per month?
A. Before I started Virtuoza, I was doing some projects on Rent-a-Coder. [A] few of the jobs I did there made me more familiar with the current state of shareware. So I decided diving in myself [would] be the smart move—the way to getting rich.
When I started my micro-ISV, in July 2003, I hoped to start earning $1,000/month almost immediately. Far from it. What happened was that I began getting an average of a sale a month. It turned [out] that creating and marketing software is much harder than I anticipated in the beginning. As I learned later, most micro-ISVs started very slowly too, at just [a] few sold copies a month.
So, I needed to continue doing projects on Rent-a-Coder; my micro-ISV wasn't going to pay the checks for long. But I continued to invest time in developing software and learning how to market it better. Finally in March 2005 I began getting more serious. The sales went up from two to three a month to ten, then twenty, then thirty. I currently make about $1,000/month, but this is just a temporary position I am in. It is already a success for me, because now, I can devote my full time on my business, which will help me grow it much faster. I estimate sales of $2,000 to $3,000/month until the end of next year (2006). Or if I get lucky—much more.
Q. What has been the biggest surprise in getting your micro-ISV started?
A. The biggest surprise was that it would take me that long to start earning good money. Don't get me wrong—now I am quite happy. Even if $1,000/month doesn't seem [like] that much, think about one important aspect: I don't need [to] do almost anything in order to keep [this] money coming. I need to spend few hours a week answering customer emails, and I will keep [this] coming for a long time.
Have you considered joining the Association of Shareware Professionals? It helped me tremendously, and at $100 per year it is almost a no-brainer.
Q. Regarding ASP, how has it helped?
A. To the outsider, the ASP does not offer too much value. You may think that the private newsgroups, which are its biggest asset, are nothing too good. The difference comes from the people in it. There are about 1,500 members currently (and rising every year). Most of them have released a product and have real experience. A good portion is doing this full-time. The difference to the Business of Software forum on Joel's site and the public shareware.authors
newsgroup that you can find on Google is this: when a topic is discussed, most of the people giving opinions have already successfully implemented the things they advise. There are a lot of smart people in Joel's forum, but most of them are not actually a micro-ISV.
I can preach this for long time, but the bottom line is this: it brought much more value to me than the $100 I spent for the membership.
I blogged my initial impressions nine months ago here: http://borisyankov.blogspot.com/2004/12/should-you-join-asp-for-some-time-i.html
.
InControl by Virtuoza
Q. When you say, "Finally in March 2005 I began getting more serious," tell me more.
A. The first "serious" thing I did was to give each of my products its own Web site. For a year and a half my main Web site has been unchanged, and I learned a lot about better persuasive techniques. I incorporated them into the new sites (in other words, my previous Web site sucked big in terms of conversion ratios). I also released a version 2 of InControl. This was the time when my orders jumped immediately from the dismal one to two per month to fifteen and began rising.
Q. How do you market your various products?
A. Initially I embraced download sites. I found a long list of these Web sites and submitted to every one of them. Then I thought, "I am ready with my marketing." The problem with this approach is that this will bring you about 50 to 60 not very targeted visitors a day, which will hardly translate to more than a few sales a month. Then comes Search Engine Optimization (SEO). What I preach now is this: learn the basics of SEO so you don't mess things up. Then concentrate on creating a great site with great content.
Then comes Google AdWords. It is great. Conversions are higher than most of the alternatives you have, like advertising on download sites (a bad idea) and advertising on sites related in topic to your software (a better idea).
For my upcoming product, FusionDesk, I will be starting a blog. (FusionDesk will be an advanced task management application. It's somewhat of a competitor to MasterList Professional, but it has a very different philosophy than your app, and the market is huge, so I don't think it will be a problem if I give you more insider details about it.)
Q. Are your ongoing development efforts aimed at new products or new versions of your existing products?
A. I am concentrating on improving my current software. There are people who ask the question, should I develop lots of small apps, or should I concentrate on fewer apps (or even only one)? I strongly suggest keeping the number of products you have to a minimum. A well-done new version of your software can easily bring you two to three times the customers you had with your previous one.
See Joel Spolsky for [an] example. His FogBugz is doing great now; it is in its fourth version. He couldn't achieve this success with ten apps at version 1. He would've been stuck with ten not quite adequate products. He said that good software takes three versions and ten years to complete. And I agree with this.
BRIAN PLEXICO, FOUNDER, CLOUDSHACK AND MICROISV.COM
Micro-ISV: Cloudshack
Web site: http://microisv.com
Blog: http://microisv.com
Interviewee: Brian Plexico
What it sells: Specialized consumer database applications.
Location: Summerville, South Carolina (near Charleston, South Carolina); United States
Q. Can you tell me a bit about the software or Web services or products your micro-ISV sells?
A. My software was made to help people manage their collections of things such as baseball cards, watches, etc. From that I developed a niche product specifically for model airplane engine collectors.
I have also developed an app to help people keep track of their skeet shooting/sporting clays scores.
Q. Is revenue a) less than what you hoped, b) about what you predicted, c) exceeding expectations, or d) you're going to buy a small Pacific island later this year? Can you share your sales per month?
A. Originally, revenue was what I predicted. I sold my collection software on eBay exclusively for quite a while, but about a year after I started, competition from competing applications and increasing eBay fees brought revenue down to about breakeven. Currently, I am not actively promoting my software, but a few sales still trickle in from people who find the Web site via search engines.
My skeet software was dealt a blow when Google disallowed a lot of my AdWords because they were gun-related. I guess their version of "do no evil" includes anything gun-related even if it is for an established sport.
Q. What has been the biggest surprise in getting your micro-ISV started?
A. The amount of requests that come in for product enhancements and how-to questions. Even if you've designed something that is fairly simple to use and includes comprehensive documentation, people still ask a lot of how-to questions.
The other thing I've learned is that creating a niche product is good, but there needs to be a sufficient market to make it worthwhile. I've created some products that fill niches and had very little competition, but the overall market was limited.
Q. Are you developing any other products?
A. I have a few ideas that are bouncing around my head right now but nothing specific I'm working on. I am doing some research for a new product that may or may not come to fruition. I'm talking to some people who would be potential customers to see whether they would purchase the software if it were available.
OmniFile by cloudshack
Q. How/why did you start microISV.com?
A. I was reading Eric Sink's original micro-ISV article, and the idea just struck a chord with me since I had developed my first shareware apps not long before the article. I thought it would make a good site to have a place where the micro-ISV crowd could congregate, so I bought the domain name. On a Saturday, a week after I bought the domain name, I downloaded WordPress, made a quick HTML template, and made the first post, which incidentally was one year ago today if I'm not mistaken.
Q. What has been the response to microISV.com?
A. The response has been great! I only wish there was more interaction with comments and the forum. Having more interaction would allow me to take the site in a direction to benefit people even more.
Q. Any advice you'd give to someone starting a micro-ISV today?
A. My advice is to just go ahead and get a product out there. With my first app, the collection software, I spent a year "perfecting" it. I kept coming up with more and more features that I thought would be cool but are probably used only by the most hard-core users. With my shooting software, I conceived the product, coded it, and was selling it within a week.
Getting the software out there so quickly showed me that starting simple is best because you can add the features that are most requested later while starting to make money immediately. It was a pretty cool feeling to go from having an idea for a software app to selling it seven days later for $20 a sale.
GAVIN BOWMAN, DIRECTOR, V4 SOLUTIONS
Micro-ISV: V4 Solutions
Web site:http://www.oriador.com
and http://www.webhelperbrowser.com
Blog: http://www.webhelperbrowser.com/blog
and http://www.codesnipers.com
Interviewee: Gavin Bowman
What it sells: A staff-scheduling application and an enhanced Web browser.
Location: Cumbria, United Kingdom
Q. Can you tell me a bit about the software or Web services or products your micro-ISV sells?
A. My main product is Oriador Rota. It's a staff scheduling/Rota package. It's been on the market for about 18 months.
Web Helper Browser has been available for about six months, and it's a Web browser with various extras, including features for saving Web pages and fragments.
Q. Is revenue a) less than what you hoped, b) about what you predicted, c) exceeding expectations, or d) you're going to buy a small Pacific island later this year? Can you share your sales per month?
A. It has definitely been a, but I feel like it's approaching b for revised predictions. If the next year or two go well, it might even make c, but I think d is a long way off!
Q. What has been the biggest surprise in getting your micro-ISV started?
A. The biggest surprise has been how hard it is to get feedback. I think I thought everyone who used it would take the time to tell us what was lacking or what they'd like to see improved or changed. In retrospect I was very naive, but obviously I didn't know that at the time.
To quickly clarify, so that any "I" or "we" makes sense, there are two people in my company. For the most part I've been the sole developer on both products, but recently my partner took over our latest product so I can focus more clearly on Oriador.
Q. Do you find it easier/harder to market/support two products instead of one?
A. A lot harder. There's no natural connection between the two products, so there's no real crossover between support or marketing. They both need to make it on their own.
Until about a month ago I was developing both products too. It was way too much pressure and way too hard to focus.
Oriador Rota by V4 Solutions
Q. How do you and your partner make decisions, and how did you decide who gets what?
A. We're very different, but we've known each other a long time. When there's a conflict, the decision seems largely inevitable. It's usually obvious to one how important any idea or suggestion is to the other.
Deciding to divide the products between us was just something that had to happen. With both I was never going to get anywhere, and Neil had reached a place where he wanted to be more involved in the development side of the business. Once we made the decision, I would never have given up Oriador Rota. Fortunately, I think he wanted Web Helper Browser. I'd reached a point with it where I was fairly comfortable with what I'd achieved, so I could take a step back. He's still bursting with ideas to expand on it and take it forward. I've been working on Oriador for a lot longer, and I still have a lot of ideas. I really wouldn't have been able to let it go.
Q. How do you presently market Oriador Rota?
A. At the moment we do it all online, mostly the Web site combined with Google AdWords. The software is listed in some of the software directories, but we don't seem to get a lot of inquiries coming in that way. We tried a little direct mailing and a little cold calling, but we weren't very good at it. We've basically just tried to do as much as we can without overextending our budget.
Q. What has been the response to your blog, especially the micro-ISV mistake postings at CodeSnipers.com (http://codesnipers.com/?q=blog/12
)?
A. The response has been great. People seemed to really like the posts and agree with what I was saying. I was a bit worried when I started writing them. I had five mistakes in mind, but I wondered if #6 was going to be telling everyone about all the things I did wrong. [But] if it had turned out to be a mistake, it would have gone against all my previous experience. I've always found that, on a small scale anyway, people like it when you own up to mistakes and accept responsibility. I know I do too. If someone says, "Yeah, sorry, that was my fault," I feel like I can trust them. There's more honesty there than if they try to blame it on someone else. I wasn't sure if it would work like that on a larger scale.
I'd read plenty of advice on what to do and what not to do from successful people but not so much from people still trying to make it. I knew that there were people out there who would make the same mistakes, so I just wanted to tell them what I did, [tell them] how wrong it was, and suggest some alternatives. It had a kind of healing effect on me too, actually accepting some of these things and starting to think about trying to fix them seemed to give me renewed energy.
Overall, blogging has been everything I wanted it to be. I've been able to meet some great people, [I] had some great discussions, and I just generally feel more connected than I did before I started. It probably hasn't been directly responsible for a single software sale, but the change in my motivation and the way I feel about the future seem a lot more valuable.
GRAHAM ASHER, OWNER, CARTOGRAPHY
Micro-ISV: Cartography
Web site: http://www.cartotype.com
Blog: None
Interviewee: Graham Asher
What it sells: Portable mapping library.
Location: Berkhamsted, England (which is a small town about 25 miles northwest of London)
Q. Can you tell me a bit about the software or Web services or products your micro-ISV sells?
A. CartoType is a portable mapping library written in C++. It uses the client's data, via plug-in data accessor interfaces, to draw scalable maps for use in location-based software like navigation programs and city guides.
So far it has been ported to four different OSs: three operating systems for mobile devices (Symbian, Microsoft Pocket PC, and Palm OS) plus Windows.
There's a lot more about it on the Web site if you're interested, and I flatter myself that it's reasonably well-written.
Q. Is revenue a) less than what you hoped, b) about what you predicted, c) exceeding expectations, or d) you're going to buy a small Pacific island later this year? Can you share your sales per month?
A. Less than I hoped. I have sold just one license so far. However, I haven't exactly done much marketing, and I have managed to improve the product a lot lately, so things may change given some extra effort.
Q. What has been the biggest surprise in getting your micro-ISV started?
A. This is very corny, I know, but I've been surprised how easy the bug fixing has been. We used source control, defect tracking, and regression testing from day one, and I was lucky enough to find an excellent fellow programmer who I could trust to do the right thing, so we really haven't had any problems with hard-to-find bugs—well, one, in fact, but that turned out to be a stack overflow in the customer's code.
Oh, one other happy surprise: CartoType uses three open source components, and to my great joy they have been completely reliable and bug-free, as far as I can tell.
CartoType by Cartography
Q. Would you in general recommend using open source components to other micro-ISVs?
A. Yes, compared to commercial components and with the proviso that I have experience with relatively few open source components. The quality is high, the developers are dedicated and intelligent, and the support from the developers and the user community beats anything I have experienced with commercial components and in-house offerings at large companies I have worked for. These remarks apply mostly to FreeType, our most important open source component.
The three open source components that CartoType uses are FreeType, a font-rendering library; Expat, an XML interpreter; and Dlmalloc, a heap allocator.
I was very comfortable with FreeType before I started CartoType, because I had used it when I was working for Symbian. (At Symbian I implemented support for scalable fonts in the Symbian OS using FreeType.) I'd seen it develop from a rather messy piece of software to a more professional, understandable, and reliable component, and I'd been in correspondence with some of the FreeType developers (mainly the two founders, David Turner and Werner Lemberg), read their remarks on the FreeType email group, and grown to trust them. I also made a couple of very small contributions of my own to FreeType.
Expat was ported into CartoType by my colleague Lex Warners, who built a C++ wrapper for it and wrote unit tests. I didn't know anything about it before we started, but it has been rock solid.
Dlmalloc, Doug Lea's heap code, is the basis for Linux's malloc code and widely regarded as the best general-purpose heap allocator around. We use it on the Palm OS port, which has a very slow malloc implementation in its runtime library. The code looks very messy to me, and I can't follow it, but it works perfectly.
I also use two very important open source tools, the Subversion source control system and its Windows client TortoiseSvn. I used Perforce for many years and still use it in my work for a client I do consultancy for. A Perforce license costs several hundred dollars per seat. Subversion does more for me, more conveniently, and is entirely free. (Perforce may of course be more appropriate for large organizations; it may very well be more scalable than Subversion.)
Q. When working with open source components, any special precautions/efforts a developer should make than when working with a commercial component?
A. In a word, no. Open source components and commercial components require exactly the same care in choosing them, testing them, and interfacing them to your code. With commercial components you may not even get the source. But in fact the only commercial components I use in CartoType are the C runtime libraries for the various OSs, and not very much of them.
One obvious point: don't choose an open source project that was started yesterday, unless it's your own, and even then think twice. You need something mature, where the developers have demonstrated continuing commitment.
Q. How long have you been selling CartoType?
A. For just over a year. I sold the first license in January 2005.
Q. What are your current marketing plans?
A. I am going to continue to improve the product, particularly its visual appearance and speed. The appearance is very highly customizable and depends on style sheets written in XML, but users won't understand that or believe it unless I create more sample style sheets that demonstrate well-known mapping styles and display more screen shots on the Web site.
The Web site is very amateurish, although it provides plenty of information in a usable way. I'll have to decide whether to pay somebody to improve it or spend time on it myself.
I am going to find more contacts and potential sales prospects and ask my head of marketing (my wife) to call them and persuade them that CartoType can improve their products. I need to contact companies (like street atlas publishers) who possess cartographic data and are starting to move into location-based applications on mobile devices and need to get to market quickly. We would then point out that they can save a year at least of software development time by licensing CartoType.
IAN LANDSMAN, PRESIDENT, USERSCAPE
Web site: http://www.userscape.com
Blog: http://www.userscape.com/blog
Interviewee: Ian Landsman
What it sells: HelpSpot, a Web-based help-desk package.
Location: Lagrangeville, New York; United States
Q. Can you tell me a bit about the software or Web services or products your micro-ISV sells?
A. UserScape's first product, HelpSpot, is a Web-based help-desk software package with two major components.
Primarily, HelpSpot was designed to track and manage the workflow for support requests. Requests are created through a manual entry by support staff or automatically from emails or the customer portal form as outlined below. There are a variety of individual support staff workspace tools and global configuration options that allow organizations to effectively manage the request resolution process from initial creation to postresolution reporting.
The second component of HelpSpot is the customer service portal. Designed for organizations seeking to leverage the efficiencies of the self-service support model, the portal includes a robust forum system and knowledge books (similar to a knowledge base). For inquires that require interaction with support staff, there's also a request submission form and a means to check and update current requests.
Q. Is revenue a) less than what you hoped, b) about what you predicted, c) exceeding expectations, or d) you're going to buy a small Pacific island later this year? Can you share your sales per month?
A. HelpSpot is current in beta, so no revenues to report as of yet. I would be happy to share sales information post our mid-October release.
Q. What has been the biggest surprise in getting your micro-ISV started?
A. I'm fortunate to report that this process has largely gone as planned. With respect to budgets and timelines, I began planning and projecting about six months before kickoff, and both have been within my initial estimates.
The most unexpected aspect has been the success in blogging about the process. From the start blogging the process to share my experience starting a company with others was part of my plan, but I had no idea it would be such a huge asset to UserScape and the HelpSpot product. The direct dialogue with my readers has, very often, resulted in a perspective on specific features or overall design I wouldn't have considered without them. Of course, the sheer reach of the blogging community has allowed HelpSpot to capture the attention of over 200 companies, which are currently on our mailing list as well as the over 80 organizations participating in the beta.
Q. Can you describe in detail how you budgeted and planned HelpSpot?
A. Planning for HelpSpot started about a year before the beta launch.
Throughout my career I've worked on technical projects that involved end-user interaction and clearly saw a need for a robust tool to effectively manage incoming issues/requests. My wife is a techie herself so we sat down and talked through the details. After we established the base feature set, I went to work writing up a spec along with several specific user scenarios that served as a way to flesh out much of the workflow and primary screens.
Initially we planned for me to keep the "day job" and leave towards the end of development. However, it quickly became obvious that to finish HelpSpot in anything close to a reasonable amount of time I would have to either scale back our initial vision or leave my job sooner to devote myself full-time to the project. Scaling back the feature set simply wouldn't allow HelpSpot to be the product we envisioned and would result in a product that mirrored the help-desk software on the market, which wasn't our goal.
In agreement that maintaining the scope of initial development would require me to leave my "day job," my wife and I discussed the financial picture. A budget was created, assuming no revenue from HelpSpot, which would allow us to cover home/business expenses for a 12-month period using a combination of her income, our savings, and a short-term contract project I had just taken.
A big part of starting an ISV is bootstrapping what you need, but there are still unavoidable expenses. For us those were design of the Web site and logo, purchase of icon sets, lawyer fees to create the license agreements, an accountant, hardware/software to support development and testing, and a server. We budgeted $12,000 for these expenses and so far have come in at about that number.
HelpSpot by UserScape
Q. What budgeting and planning advice would you give to other micro-ISVs?
A. I do think budgeting is a key area where some Micro-ISVs go wrong. They aren't realistic. They only budget themselves to the point of release but don't properly plan for the postrelease period. This is critical because you want to give the product ample time to succeed in the market.
Get realistic by committing the numbers to paper! Make sure you get out Excel and really run the numbers on how much it costs you to live, what type of hardware and software you'll need, and how long you can tolerate going until your product catches hold.
While budgeting, also try to build in a "cushion" for those unknown/unexpected expenses. For example, my primary development machine had a serious hard drive crash, and it cost me about $1,000 to get things back up and running.
Q. How do you go from your topmost design to your coding?
A. I guess my style is sort of like an onion. I tend to think in terms of the database first, the core, so that's where I start. With HelpSpot, the core functionality is the creation/tracking of customer requests, so I put a lot of time into thinking about how to optimally store requests. With the database schema in place for requests, I created the framework for things like logging in, creating users, managing categories, and [doing] the other administrative aspects.
Once the foundation was in place, I "layered" on functionality starting with request management and gradually working out to the other self-service features. With the core functions done early in the development process, there's ample time to use that core functionality to make sure it works as well as it can, not just from a code perspective but from a user perspective. Does it really do what I need done? Does the UI work efficiently? Does this layout make the most sense?
In HelpSpot the request management screen is where most users will spend the majority of their time so getting that screen right was extremely important. By building it near the beginning of the process, there was a chance to get lots of feedback on it as well as do a great deal of testing with it.
Q. What developer tools have you found most useful (source code control, automated builds, testing, etc.)?
A. All the development for HelpSpot was done on a Mac, which has provided a great deal of efficiency and flexibility. I'm not a big fan of IDEs, so all the coding was done in BBEdit, a classic Mac text editor. It also integrates with Subversion, which is handy.
I'm a big Subversion fan; it's a great piece of software. It's just so easy to use, at least compared to the other source control systems I've used. I've managed to automate just about everything in the build and development process, which is key to reducing errors. To generate new builds all I need to do is run one shell script, and it gets the source from Subversion, runs it through the Zend Encoder (used to encode the PHP source files), [and] packs it all up in both tar.gz and zip formats, and it's ready to ship.
Another tool critical to my development process is VMware. It's just an amazing product. It allows me to have just one Windows machine in the office [that] can handle all the testing for dozens of platforms and database combinations. It also gives me the ability to easily replicate a customer's exact environment to debug tricky problems that have to do with specific PHP configurations or database setups.
Q. Can you tell me a bit about what you did before starting UserScape?
A. My programming career started at a small start-up called Active Learning Technologies. I started there as a project manager and moved into more of a programming role over time. After three years I went to Marist College where I was the assistant director of Academic Technology and eLearning (it's a mouth full!). In addition to helping run the eLearning program there, my office also had student programmers who did various Web programming projects for the school and community. As the "Web" person there, I also did a lot of internal consulting on various Web initiatives in the school.
KIRBY TURNER, FOUNDER, WHITE PEAK SOFTWARE
Micro-ISV: White Peak Software
Web site: http://www.whitepeaksoftware.com
Blog: http://www.thecave.com
Interviewee: Kirby Turner
What it sells: Email component and, in 2006, an email client.
Location: Armonk, New York; United States
Q. Can you tell me a bit about the software or Web services or products your micro-ISV sells?
A. We sell software that improves online communication specifically around email. The first product, SMTP Diagnostics, helps mail administrators, programmers, and power users troubleshoot problems with outgoing email. It is also useful for testing the configuration of outgoing mail (SMTP) servers.
Our second product is due out around the spring of 2006 and is code-named Vertigo. Vertigo is a new type of email client for the desktop designed specifically for the one- to two-person business that uses a POP3/SMTP mail server. It includes email, contacts, calendar, to-dos, notes, history, and more. Vertigo provides an innovative approach for managing information and addresses the issue of how to manage hundreds of thousands of email messages collected over the years.
Q. Is revenue a) less than what you hoped, b) about what you predicted, c) exceeding expectations, or d) you're going to buy a small Pacific island later this year? Can you share your sales per month?
A. Answer is b, about what I predicted.
Sales per month are between $3,000 and $5,000 USD per month. At the moment, the majority of monthly sales revenue comes from professional services, but the plan is to reverse it next year with product sales being the primary source of revenue.
Q. What has been the biggest surprise in getting your micro-ISV started?
A. I have had two biggest surprises. The first big surprise was how little I knew about running a company when I started. The second big surprise was how much I enjoy the business side of having a software company. I'm a developer at heart and will always be. But the challenge of running a company [and] learning what makes a product sell, how to market a product, and so on, are all new challenges that get me more excited today than writing code.
Q. Tell me about your business coach. How did you find them, what did they do for you, how much did it cost, and would you recommend them?
A. My business coach, Brian Harp of Class IV Solutions (http://www.class4solutions.com
), is someone I have known for years and who has been a mentor to me in the past. However, it wasn't until I heard him speak at a recent ICCA (Independent Computer Consultants Association) meeting that I realized 1) I needed a coach, and 2) he had started a business around coaching technology companies.
While I'm sure the cost for coaching varies based on the coach and the need, my coach charged me $500 per month. We met face-to-face for one hour each week during the three months of coaching. Also during this time, we talked over the phone and via email. And he continues to check up on my status from time to time.
The primary objective of my coach was to help me obtain focus regarding my company. Prior to this I was trying to be everything to everybody, and I wasn't happy one bit. With his help I was able to put together a plan for my company, and now I have a clear vision of what type of company White Peak Software will be in five years and in ten years.
In addition to my primary objective of becoming more focused, my coach has helped me by reviewing marketing collateral and content on my Web site. He has also provided helpful advice on how to handle various issues that can come up when running a business.
I definitely recommend a business coach to anyone who is looking for help. The experience has turned my company around for the better, and I am much happier now than I've been in a long time.
SMTP Diagnostics by White Peak Software
Q. Should micro-ISVs use business coaches in general?
A. No, micro-ISVs should not use a business coach in general. Going to a business coach is like going to a therapist. You have to acknowledge you need help, and you have to have a desire to change. Also, the area that you need help with may be better served by a specialist. For example, someone who specializes in marketing could provide you better help if you need assistance in that one area.
In my case, I realized I did not know how to run a company. Heck, I struggled with describing my company to others. "I run a software company" doesn't say a lot about what you do to prospective customers. I needed a coach to help me work through many small general problems like this. And like a good therapist, my coach was able to pinpoint my problem—the lack of focus—which helped me overcome the problems.
For those micro-ISVs that struggle with the transition from tech person to businessperson, a business coach can be helpful.
Q. How do you like eSellerate, and would you recommend them to other micro-ISVs?
A. My experience with eSellerate has been great, and I highly recommend them to other micro-ISVs. Their SDK and services have saved me countless hours, and the customer service has been top-notch with most of my questions answered within an hour, if not sooner. Setup for a Web store is easy and not time-consuming at all. But one of my favorite services from eSellerate is the integrated eSeller that comes in the SDK.
With an integrated eSeller, customers who are evaluating your software can buy a license from within the program. There is no need to redirect your customers to a Web site. At the moment, almost half of all licenses sold for SMTP Diagnostics came from the integrated eSeller. While I cannot prove it, I believe I have made more sells as a result of having the integrated eSeller.
eSellerate also has a nice reporting system, and all information can be exported should you wish to keep customer information in your database. They also have the ability to post transaction information to your own server using XML, which is useful if you want to automatically update your own database.
Another feature I really like about eSellerate is the generation of Armadillo license keys. This means I can protect my program using Armadillo and still use eSellerate for key generation. And for an added bonus if you use the integrated eSellerate, the license key is sent back to your program after the transaction is complete. You can then programmatically register the Armadillo license key (using the Armadillo API) within your program, eliminating data entry errors from your customers.
There are many other features from eSellerate that I have not yet explored. They have an affiliate program allowing you to cross-sell your products with other products. And you can customize the look of the Web store to your liking. I can ramble on more about features of eSellerate, but I will spare you. However, I will say that I find eSellerate well worth the per-transaction fee they charge.
RUDOLF F. VANEK, FOUNDER AND EXECUTIVE DIRECTOR, QUANTICUS S.A. DE C.V.
Micro-ISV: Quanticus S.A. de C.V.
Web site: http://www.dbxtra.com
Blog: None
Interviewee: Rudolf F. Vanek
What it sells: Reporting and query tool.
Location: Querétaro, in the state of Querétaro; México
Q. Can you tell me a bit about the software or Web services or products your micro-ISV sells?
A. DBxtra is a stand-alone reporting and query tool that connects to unlimited databases, including Microsoft Access, Microsoft SQL Server, MySQL, Oracle, DB2, FoxPro, Pervasive, Excel, and any other database through ODBC. No programming or database knowledge is required to explore and report your data with our user-friendly report tool.
Our main market is the United States and Europe, and the product is mainly for database administrators, although we are working on an API right now, which will allow embedding of DBxtra in other software applications. This new version will be released in December 2005.
Q. Is revenue a) less than what you hoped, b) about what you predicted, c) exceeding expectations, or d) you're going to buy a small Pacific island later this year? Can you share your sales per month?
A. b, about what I predicted.
Q. What has been the biggest surprise in getting your micro-ISV started?
A. Well, I guess the marketing experience. We sell mostly by the Internet, and we had to learn a lot of online marketing. We experienced almost anything imaginable but now know that the best advertising won't be as effective as, for example, product reviews, press releases, and word of mouth.
Q. Given your main market is database administrators, how do you market to them?
A. Mostly through AdWords campaigns and also advertising on IT sites.
DBxtra by Quanticus S.A. de C.V.
Q. Were database administrators the market you thought DBxtra would appeal to?
A. Yes, absolutely.
Q. How do you deal with tech support?
A. We offer free email support through our help desk, http://www.dbxtra.com/hd
.
The next group of eleven interviewees are showing solid success and revenue, sometimes because they've put the time and effort into building, nurturing, and growing their customer base and their products and sometimes because they've found a well-defined need and have worked tirelessly to meet it.
Again, certain commonalities come to mind as you read these interviews with successful micro-ISVs:
Of these interviews, you may be most surprised by two. Yes, Clay Nichols does talk like he's channeling Rodney Dangerfield, but Mr. Analogy, as he's known on the Joel on Software forum, knows how to run a business too. Speaking of business, ever hear of a micro-ISV successfully selling to the biggest of the big banks when it opened for business? Neither had I, which makes Mike Schoeffler's Profitdesk something of a phenomenon and his insights into selling to large enterprises very interesting reading.
ANDY MILLER, OWNER, STRUCTURED SOLUTIONS
Micro-ISV: Structured Solutions
Web site: http://www.structured-solutions.net
Blog: http://www.structured-solutions.net
Interviewee: Andy Miller
What it sells: Add-ons to two popular e-commerce packages.
Location: Bethany, Pennsylvania; United States
Q. you seem to have several different Web sites. Why?
A. I have several URLs. http://www.structured-solutions.net
is the main company Web address. Most of the content is a Web log. I also have two stores. Since I sell add-ons for two e-commerce packages (StoreFront and BVC2004), I have created a separate store for each package. StoreFront Add-Ons are sold on http://www.sfaddons.com
. BVC2004 Add-Ons are sold on http://www.bvcaddons.com
. Our main site is a blog.
Q. Can you tell me a bit about the software or Web services or products your micro-ISV sells?
A. Both StoreFront and BVC2004 are complete e-commerce packages. They can be installed on a Web site and used as is. However, nearly every merchant wants various tweaks and customizations. Many customizations are so common that they make sense as products. This is the genesis of most of my add-ons.
There are two things wrong with this model: a) add-ons inherently have a low price point (merchants are unwilling to pay very much for what they perceive as a visual or usability feature), and 2) the package vendors will likely incorporate the features into future versions of their base product, rendering my add-ons obsolete.
So, I have shifted most of my effort to a new product called Shipper. This is a stand-alone product that automates shipment processing for the same two e-commerce packages. Shipment processing is a critical and time-consuming task that all merchants face. Shipper significantly reduces the time it takes to process shipments and eliminates errors.
Q. Is revenue a) less than what you hoped, b) about what you predicted, c) exceeding expectations, or d) you're going to buy a small Pacific island later this year? Can you share your sales per month?
A. Revenue is definitely less than I hoped but about what I expected. I am currently running about $2,500/month in sales. I am shooting for $10,000/month.
Q. What has been the biggest surprise in getting your micro-ISV started?
A. The biggest surprise has been how nasty some (luckily few) people are when the relationship is purely over the Internet. I suppose the anonymous aspects of the Internet let some people feel that can behave in ways they never would at home…or perhaps they are just that way.
Several other surprises all fall into the category of "harder than I expected," for example, marketing, answering the phone, and switching hats (related to answering the phone).
Q. What's it like working/partnering with UPS?
A. Right now I am a nonentity to UPS. They have a developer program, but it is geared toward using their Web services. My program (Shipper) works directly with their client application called UPS WorldShip. UPS WorldShip seems to be developed and supported out of an arm of UPS that does not have developer relations.
The next version of Shipper will include support for FedEx Ship Manager (their equivalent client app), and the version after that will include support for Endicia Galaxy (Endicia is a U.S. Postal Service vendor, and Galaxy is their equivalent client app). Both of those companies have contacted me to inquire whether I would support their programs.
Q. Before starting your micro-ISV, what did you do professionally?
A. I was a civil engineer back in the '80s for a paper company headquartered in New York City. That led to an IT position first at the paper company and later for a utility based in New Jersey. We eventually moved (back, for me) to Oregon, where my wife and I both went to work for Intel. The lower cost of living gave me an opportunity to leave the corporate world and play Mr. Mom for a while. Once the kids were in school all day, I started the ISV.
Q. What's your marketing plan currently?
A. When I started the ISV, I was primarily hiring myself out—a fairly typical bootstrapping technique, I think. Eventually I installed an e-commerce site for a customer. It received some attention from other merchants that wanted similar enhancements made. That turned into the first set of add-ons and a lot of custom work. At that time, the only marketing I needed was word of mouth via the platform vendor's newsgroups/forums. I answered every question I could so that my name would appear frequently. That led to more than enough requests for work.
About a year ago I decided that it was too hard to switch back and forth between bidding/working custom jobs and creating packaged products. And I decided that I preferred the packaged products, so I dropped the custom work. It has turned out that answering questions in forums is great advertising for custom work (I still get requests daily that I turn down) but not so good for pitching products. The marketing plan has been slow to change. Eventually I added some advertising (via Google AdWords) and some resellers. But that has not been as effective as I would have hoped. That is where it stands, but it is clear to me that I need to develop a new marketing plan.
Shipper by Structured Solutions
Q. How do you manage tech support (and nasty customers)?
A. All support is provided free via email or through a Web-based ticket system. This works well for the most part, but some people (perhaps especially merchants) want to talk. I tried accepting phone calls for a while, but it quickly became too expensive both directly (phone charges) and indirectly. Some days I spent literally hours on the phone walking people through simple steps ("click on the Start button…"). So I stopped accepting phone calls. Since I never advertised that I accepted them, this was not a shock to anyone.
Nasty people tend to stay anonymous (which makes me think they know they are behaving badly), so it is fairly easy to ignore them…although it can still be frustrating (see "Arrgh!" at http://www.structured-solutions.net/ArrghShouldIKeepTheErrorReportInShipper.aspx
).
My vision of the future is headlined by more income, but it also includes a well-designed marketing plan that is being executed and a well-rounded support plan that includes phone support. I suspect there will also be a subscription component that will affect both income and support.
BRIAN NOTTINGHAM, VICE PRESIDENT, INTERAPPTIVE
Micro-ISV: Interapptive
Web site: http://www.interapptive.com
Blog: Not at this time
Interviewee: Brian Nottingham
What it sells: ShipWorks, a shipping management system for online sellers.
Location: St. Louis, Missouri; United States
Q.Can you tell me a bit about the software or Web services or products your micro-ISV sells?
A. We sell a single software product called ShipWorks. ShipWorks is a desktop application, written in C#, that connects to online shopping carts and selling platforms to download a merchant's orders. Our focus is on helping people ship their orders as quickly as possible. Through our relationships with UPS, FedEx, the U.S. Postal Service, and online postage services such as Endicia and Stamps.com, ShipWorks is able to connect directly to these services to get rates, download labels, and track shipments.
ShipWorks automates many tasks, such as emailing tracking numbers after a shipment is processed and printing "pick lists" for pulling inventory when new orders are downloaded. Sellers using ShipWorks are able to print reports, compose email, [and] manage their customers, and it can be used across multiple computers on a network.
Q. Is revenue a) less than what you hoped, b) about what you predicted, c) exceeding expectations, or d) you're going to buy a small Pacific island later this year? Can you share your sales per month?
A. Our revenue is what we expected it to be at this time. Our goal has always been to "grow big slowly," and that's exactly what we are doing. We have recently switched from a one-time fee to a recurring price model, and we expect that to have a great impact on our future growth.
Q. What has been the biggest surprise in getting your micro-ISV started?
A. It's a lot of work! During the first months, the excitement of starting a new project is plenty to keep you going. Everything is fresh, you're feeling good, and it's just a great time. Then at some point between working long hours, losing sleep, and sacrificing time with family and friends, you do start to wonder if it's worthwhile—if you're really going to make it. It's that feeling that surprised me, that feeling of "Is this worth it?" that would seem to just sneak up on me.
But, with plenty of encouragement from each other and support from those around us, we just keep pushing on. And each day we get a little bit closer.
Q. How should a micro-ISV maximize its relationship/partnership with "big companies" like FedEx and UPS?
A. I think a big part of being successful is leveraging others that are already successful. We leverage "graduated" micro-ISVs, such as the articles written by Joel Spolsky and Eric Sink. We leverage local small-business groups, college professors, and business advisors provided as a free service by the Chamber of Commerce. And we definitely leverage our partnerships with companies such as UPS, FedEx, Endicia, Marketworks, and the like.
The integration of their services into ShipWorks is obviously a key part of our product, but to leave it at that would be really missing out. We do joint marketing, promotional offers, [and] case studies, and [we] listen to any advice they are willing to give. And to have our name associated with companies that already have customer mind share is a big deal.
ShipWorks by Interapptive
Q. How did you find/start/create these relationships?
A. ShipWorks [was] called ShopInvoice, and basically all it did was use HTML-based templates to print invoices and packing slips. Based on feedback from the few customers we had, their biggest pain point was printing address and shipping labels. So we started to contact shipping carriers and postage services. Some companies such as the U.S. Postal Service and Endicia have freely available APIs that we integrated right away.
Other companies such as eBay have developer programs that we became apart of. UPS and FedEx had forms on their Web sites to request information about doing third-party integration. We pursued UPS quite heavily for a few months just to get our foot in the door. And FedEx didn't show interest until almost three years later.
Q. It sounds like you have a partner or partners…. If so, how did you work out who does what and who gets what?
A. Wes and I met our first year in college. He was in the business school and doing Web sites for small businesses in the St. Louis area, and I was in the math and computer science program. Several years later, he knowing my software background, asked me to put together a small application for one of his Web site clients. It turned out well, and we wanted to see if we could start selling it as a product.
Wes worked up the contract between us, and we both played a big part in revising it. Since he had initially started the Web site business, we agreed that officially he would be president and I the vice president. More important, we agreed that each of us brought something unique and critical to the company—he on the business side and myself on the technical—and that all decisions, profits, and aspects of the company would be shared equally between us. Four years later, and that is still working well for us. I do all of the product development and technical parts of our Web site; he manages our relationships, finances, and customer support; and we each play an equal role in spending, branding, and business decisions.
Incidentally, the product we initially started our company on didn't even make it off the ground.
CLAY NICHOLS, PRESIDENT, BUNGALOW SOFTWARE
Micro-ISV: Bungalow Software
Web site: http://www.StrokeSoftware.com
Blog: None
Interviewee: Clay Nichols
What it sells: Software for speech and language therapy.
Location: Blacksburg, Virginia; United States
Q. Can you tell me a bit about the software or Web services or products your micro-ISV sells?
A. Software for speech and language therapy. Our programs are interactive educational software that helps (primarily adult) stroke survivors regain speech, language, and cognitive skills impaired due to brain trauma (commonly due to a stroke).
Q. Is revenue a) less than what you hoped, b) about what you predicted, c) exceeding expectations, or d) you're going to buy a small Pacific island later this year? Can you share your sales per month?
A. Does anyone share their monthly sales? Maybe I should approach this like Enron: our reported revenue depends on who is asking. Enron told the IRS they were losing money (thus paying little or no tax) and told shareholders they were making it hand over fist.
And, please don't lean over and tell me to whisper it in your ear and it'll be "our little secret." They didn't do much for Connie Chung. But, seriously…sales are now where we thought they'd be in the very beginning. We were naive in the beginning to think we'd get to this point that fast. I didn't understand that you had to market to your customers. I thought they'd all be looking for us. I think it was Arthur C. Clarke who said that humans extrapolate linearly but development happens exponentially, so in the short run humans overestimate progress and in the long run they underestimate it. (That anecdote works much better with a simple graph visual.) But it's also how most growth processes happen, especially marketing.
By the way, that's why I believe so strongly that you better love what you're doing as an ISV because the money may take a while to develop.
Q. What has been the biggest surprise in getting your micro-ISV started?
A. That I'm very good at sales. I never even considered a career in sales. Always [I] kind of looked down on it, for the same reason I think most people do: most salespeople we experience are lousy salespeople. They do it all wrong. They try to coerce you into a sale. A good salesman facilitates the sale. And that was quite by accident. In the beginning, we were very concerned with "Is our software working well? Does it really help people?" So, I naturally approached sales as "OK, what's your difficulty? Do you think XYZ would help you? Did it help you?" I stumbled onto the right way to do sales, which leads to the second thing I learned, but I'll get back to that in a minute.
Looking back, sales should have been an obvious career path for me. I spent about six years on the speech and debate team in high school and college. I had a debate team scholarship for college, which I got by selling myself (figuratively) to the college's debate coach, who I met at the national finals in Baltimore, which illustrates the point I made earlier: do what you love. I was lousy at debate my first year. We didn't win a single round. My second year I went to the national finals. I'd have never stuck with it if I'd not loved doing it. I've always thought educational software could do more, loved education (I taught for a few years), enjoyed language, and [enjoyed] programming. I guess my subconscious was working overtime to guide me to what is very nearly my perfect career. (Next is a career in usability engineering, but that's another interview.)
I've also discovered that I really need a lot of independence in my work environment. My daughter, Claire (four years old), is very shy. If you bring her to a party, she clings to you. But the other day she was playing with some friends and telling a story. And she did the most amazing thing: when she forgot part of the story, she'd ask the other kids, "And what do you think the princess found?" and then incorporate that into the story. She was fantastic, quite by accident and quite naturally. If we tried to help her, she wouldn't have been able to hear her own quiet internal voice whispering that ingenious improvisation.
I'm kind of the same way. Throw me in the water, and I swim just fine. If you give me a chance to think about what might happen (or give me a crutch to hang onto), I'll flounder.
This is silly, but there's a great line from The Incredibles (hey, I have a four year old and a two year old) where the supermom, pursued by the villian's henchmen, tells her doubting superdaughter, "Doubt is a luxury we cannot afford now."
Q. How do you market your extensive line of speech therapy products?
A. Lots of different ways. Print ads, which you have to be frugal about; Internet advertising (AdWords, etc.); and of course people just find us on the Internet.
Q. You sell to both consumers and speech therapy medical professionals. What's that like?
A. It means you're selling to two distinct market niches at the same time. We have to be very careful about our use of jargon. I think it's very difficult to have two different Web pages (one for therapists, one for patients), and you don't always control what page the customer gets to.
Q. Few non-health-care programmers know that much about the health-care industry. I don't! Do you think there are opportunities now there for micro-ISVs?
A. That depends. There are two distinct market segments (maybe more) in health care: software for the system itself and software that a clinician would use as a tool. We fall into the latter category. The first category has higher risk and higher reward—things like Electronic Medical Records (EMR). It's an enormous market but also hard to sell to because it has to satisfy more different use cases, [satisfy] more different customers, [satisfy] more political issues, and interface to more systems. And it's mission-critical.
Speech and language recovery products by Bungalow Software
DAVID MICHAEL, OWNER, DAVIDRM SOFTWARE
Micro-ISV: DavidRM Software
Web site: http://www.davidrm.com/thejournal
Blog: http://www.joeindie.com/blog
Interviewee: David Michael
What it sells: Journaling or diary software for Windows.
Location: Tulsa, Oklahoma; United States
Q. Can you tell me a bit about the software or Web services or products your micro-ISV sells?
A. The Journal is journaling or diary software for Windows. Originally released in 1996, The Journal 4 is the most recent version. The Journal has evolved from a simple daily entry journaling program to be a robust personal organizer where people can store just about any information in just about any digital form—text, images, [and] objects of all types. Throughout the growth of the software, though, I've worked to keep it simple to use.
Q. Is revenue a) less than what you hoped, b) about what you predicted, c) exceeding expectations, or d) you're going to buy a small Pacific island later this year? Can you share your sales per month?
A. For 2005, revenue has been about what I expected. Of course, I always strive to exceed my expectations. Summer was a bit sluggish, as it has tended to be over the past eight years, but sales always pick up again in fall and build towards January, historically our best month of the year.
Revenue for 2005 has averaged over $7,500 per month. Comparing the first eight months of 2005 to the same period in 2004, 2005 has seen a 32 percent increase in new sales. 2004 had a 41 percent increase in new sales over 2003.
Q. What has been the biggest surprise in getting your micro-ISV started?
A. Well…the first surprise came in 1996 when I realized my "learn Delphi" project interested other people, who found the software as useful as I did. Then there was the whole "people will pay me for this" period of amazement and wonder.
On the "nasty surprise" side of the aisle, there was the effect of this new income on my taxes. First while I was still working for another company full-time, and again when I went full-time working for myself. FICA (the so-called self-employment tax) can be painful if you don't plan for it properly.
As sales began to trickle in, as I tried to be as professional as I could with this sudden new business venture, I realized that my computer science degree (of which I am still immensely proud) had been deficient in several areas, specifically, business management, accounting practices, marketing, and sales.
Q. Where are you based, and where did you get your degree from?
A. I live in Tulsa, Oklahoma. I received my computer science degree from a local, private university (Oral Roberts University). I've been in Tulsa since 1996 and for 19 of the last 20 years.
Q. How have you marketed, and did you/do you have a marketing plan?
A. In the early days, the extent of my marketing was getting listed on Delphi pages and on shareware pages like ZDNet and Download.com (remember when they were separate sites?).
Over the years, though, I've tried to improve my marketing in a number of ways:
I don't have a single, cohesive marketing plan that I follow. Mostly [I have] a collection of things I want to achieve, and when a chance to achieve one of those comes up, I act on it (or try to).
My plan is like what I mentioned for the Web page: try to accumulate good stuff and good decisions over time.
Q. What do you attribute both the success and the longevity of The Journal to?
A. I think that a big reason for the growth of The Journal over the years is that I listen to my users. This takes a few different forms:
The Journal by davidrm Software
The Journal started out as a more convenient daily journal for myself. I had been using Microsoft Word for Windows since 1993. While Word provided an adequate solution to my main computer journaling goal (copy and paste; I'm not kidding; I hate having to retype anything), it was too general-purpose to be an easy journaling tool. Because I made it a point to listen to my users from the beginning, though, that very personal design goal has been expanded into something I could never have created on my own.
In other words, The Journal is not just my vision of what journaling software should be. It's the combined vision of over 7,000 people, accumulated over 9 years. I take a lot of the credit, of course, but I acknowledge my influences.
Another reason The Journal has been around so long is that I've never grown tired of working on it. That's probably related to my using the software daily. I live in this software, tracking progress on my development projects, planning my week, writing articles for blogs, planning projects, writing books…everything. (I'm not the only one, either; I hear from users who say much the same thing.) As I use The Journal, I find new ways to extend it and sometimes recognize in a user's suggestion something that has been bugging me, as well, and even see a possible solution.
Yes, I can talk on and on about The Journal.
BEN RICHARDSON, COFOUNDER, SWITCH I.T. PTY
Micro-ISV: Switch I.T. Pty
Web site: http://www.campaignmonitor.com
Blog: http://www.campaignmonitor.com/blog
Interviewee: Ben Richardson
What it sells: Email newsletter software.
Location: Sydney, Australia
Q. Can you tell me a bit about the software or Web services or products your micro-ISV sells?
A. Sure, we offer email newsletter software as an ASP solution. We're a little bit different [from] the masses of other email newsletter software in that our tool is designed to meet the requirements of one specific group of people, Web designers.
Q. Is revenue a) less than what you hoped, b) about what you predicted, c) exceeding expectations, or d) you're going to buy a small Pacific island later this year? Can you share your sales per month?
A. b, about what we predicted, and our revenues are currently growing at about 15 to 20 percent a month, which we're pleased with. One good thing about having a steady growth rate is that it makes it much easier for us to scale with the increase in customers, which is pretty important for an ASP.
Q. What has been the biggest surprise in getting your micro-ISV started?
A. By far the biggest surprise for us was how long it took us to get to version 1 of the software. We'd already built something very similar for our internal use, so we figured we were almost there. We realized pretty quickly that converting something that's built just for your needs to software that any company can start using in a couple of minutes isn't easy!
I still remember how depressing it was sitting down with a customer and watching them try to use an early beta; before the run-through we figured, we [had] a few weeks for launch. Five minutes in, and a few months was looking more realistic.
Q. Has your blog helped your marketing?
A. Definitely, and in a couple of different ways. Firstly, it's helped us become more transparent, which means our customers realize there are actual people behind the product. I even remember someone recommending our product on a forum (http://discuss.joelonsoftware.com/default.asp?joel.3.85147.13#discussTopic85240
), and they mentioned we had a blog, which made them think we would be more reachable for any presales questions (which is true).
Secondly, we try to post content that's of interest to our target market, Web designers, which means they'll link to it from their blog, add comments to ours, and keep coming back to check for updates. We also try to use our blog as an unobtrusive way of keeping our customers in the loop about any new features and the future of our software.
Campaign Monitor by Switch I.T. Pty
Q. You have a very attractive Web site (naturally enough). Do you think micro-ISVs pay enough attention to how their Web sites look?
A. Thanks, Bob! We're lucky in that we have designers in our team, but I definitely think a lot of micro-ISVs could benefit from getting some outside design help for their Web sites. I know how much more confidence I have when buying a product from a professional-looking, easy-to-use Web site, and I don't think I'm alone there. Yet there are a heap of ISVs trying to sell great software from terrible-looking sites.
I guess with most ISVs it's a cost issue, and since they can whack a half-decent site together themselves, it seems like a no-brainer to keep it in-house and save the money. But this is your one and only chance to impress the customer before they move onto the next result from their Google search—maybe it's worth a few extra bucks?
Q. How do you track/manage your tech support?
A. We currently handle tech support through email but are looking to move to something more sophisticated soon—probably HelpSpot by UserScape. We also use support requests to keep any eye on any areas of the application or help documentation that can be tweaked so users aren't put in a position where they have to contact support.
Q. Are there other products in the works (you can share as much or as little as you want…)?
A. We're currently working on a something similar to Campaign Monitor, which we know a few of our customers are keen to get their hands on. But getting to version 1 is a long road, and the end is still a while off, so we'll be revealing more when the launch date is a little closer.
Q. How many people now work at your company?
A. We've a small team of four full-time staff, which was only two when we first started developing Campaign Monitor. We still do quite a bit of Web site development for clients, so between developing new features for Campaign Monitor, support, marketing, working for clients, and developing a new product, we're all kept pretty busy.
KEITH CASEY, CEO, CASEYSOFTWARE
Micro-ISV: CaseySoftware
Web site: http://CaseySoftware.com
Blog: http://blogs.CaseySoftware.com
, http://CodeSnipers.com
, and http://ProjectManagementBlog.com
Interviewee: Keith Casey
What it sells: dotProject hosting, customized open source applications.
Location: Virginia, United States
Q. It looks like you have a main Web site and several blogs?
A. My company Web site is http://CaseySoftware.com
, which has been online since July 2004. On the blogging front, there are actually three efforts with which I am involved:
http://blogs.CaseySoftware.com
, which has been online since December 2004. I focus on general development principles, business issues, and experiences with customers and projects.http://CodeSnipers.com
. It is a group blog with approximately 10 to 12 developers posting at least once/week discussing a wide swath of development issues. This site has been very successful so far and has begun to sprout a community around it after only six weeks of being online.http://ProjectManagementBlog.com
. It is based on a group blog model similar to CodeSnipers but is only two weeks old and is therefore in embryonic.Q. Can you tell me a bit about the software or Web services or products your micro-ISV sells?
A. We have three major areas of development that are going on at any given time:
Q. Is revenue a) less than what you hoped, b) about what you predicted, c) exceeding expectations, or d) you're going to buy a small Pacific island later this year? Can you share your sales per month?
A. c, with a caveat. Sales started above what I initially hoped but have grown significantly slower than hoped and projected. When I quit my full-time job, I had enough of a customer base with enough revenue on contract to completely replace my income. Within a month, I had enough work and revenue to bring on someone part-time. It took another three months to bring a second person on part-time. Now, after five months, it looks like a third person may be needed part-time.
Q. What has been the biggest surprise in getting your micro-ISV started?
A. What surprised me more than anything else was how little software development I'm doing. When I started out by myself, I was putting in a steady 60 hours/week every week. Now that I have two additional people on staff, I've found that I spend 35 to 40 actively doing development while the rest of the time is used writing specifications, talking with current and potential customers, building connections with like-minded ISVs, and generally learning and studying trends, competitors, and potential partners.
Q. How/why three blogs? (Put more nicely, "Do you see blogs as being a major part of your marketing efforts?")
A. Yes, blogging is a major part of our marketing efforts. From the evaluation of our opportunities at any given point, approximately 60 percent of our business consists of repeat/ongoing customers, but the next biggest segment (30 percent) consists of people who have found CaseySoftware through our blogging efforts. As one of my customers noted last evening, "Your blogging has been note-perfect. I'm an easygoing geek who gets it. And can write. I love it."
I think blogging on a variety of technical and business topics serves as a résumé for myself and my team. When I'm speaking with a potential customer, I make a point of letting them know if I've written something that may be of interest or applicable to their situation. If I can demonstrate to a customer that I a) understand their problem already, b) have dealt with a similar situation, or c) have skills above and beyond the competition, the one to two hours I spend each week writing becomes a business development activity.
Q. Do you find managing your part-time development staff hard when you're the owner and any advice to other micro-ISV about hiring, finding good people, or managing?
A. I believe one of the fundamental strengths of the open source community is the code repositories. It allows you as a recruiter with a technical background to actually investigate a candidate's claims and contributions. In addition, you can most likely browse whatever help forums, mailing lists, etc., to see how the person interacts with others and their general attitude. A candidate who has worked exclusively for private organizations does not have the walking, talking résumé that involvement with an open source project provides.
Open source implementations by CaseySoftware
MIKE SCHOEFFLER, PRESIDENT, PROFITDESK
Micro-ISV: Profitdesk
Web site: http://www.profitdesk.com
Blog: No blog
Interviewee: Mike Schoeffler
What it sells: Bank deposit pricing software.
Location: Piscataway, New Jersey; United States
Q. Can you tell me a bit about the software or Web services or products your micro-ISV sells?
A. We make SmartRate deposit pricing software. It statistically analyzes historical bank depositor behavior to help the bank optimize deposit rates. This is a big deal because no one in the industry has the right tools to do this job—they are forced to rely solely on their experience in a very complex arena. Underpaying customers can drive them to the competition. Overpaying depositors is even worse; it restricts a bank's ability to provide what customers really want—more branches and better service. There are trillions on deposit in America, and banks can make a lot more money by optimizing their rates.
Q. Is revenue a) less than what you hoped, b) about what you predicted, c) exceeding expectations, or d) you're going to buy a small Pacific island later this year? Can you share your sales per month?
A. About what I predicted. Initial sales are minimal as we work with our first beta customers. Once the kinks are worked out, we're primed to roll out to the rest of the industry. We have developed a deep pipeline of customers who want the final product.
Q. What has been the biggest surprise in getting your micro-ISV started?
A. I originally figured, I'm a banker who knows how to make software, not a salesman. Marketing to billion-dollar institutions will be the hardest part of my job. [It] turns out to be pretty simple.
I had the most amazing salespeople working for me when I was a banker. I used one, Steve Janaszak, as a role model, partly because he is successful at sales but also because he rises above what I used to think of as "sales."
When I first met Steve, my boss had bought $2 billion in CMOs (Collateralized Mortgage Obligations) from Wall Street. I had no idea how to model them! Steve had sold only some of the CMOs, but he was happy to help me get better at my job—even though it took away from his time focusing on selling more bonds. He saw an opportunity to help and ran with it. Eventually, my boss moved on and so did Steve's major relationship at the bank. When I had the chance to repay the favor by relating Steve's hard work, I was glad to do so.
I remembered how good it felt to work with someone helpful, and I've focused our company's marketing on the principle of good karma. For example, there are no good sources on our field of expertise (deposit pricing). So, we write articles in American Banker, US Banker, etc., on arcane topics in deposit pricing, and I speak at industry events. Sure, these articles help future competition learn the ropes. But the chance to be useful to the industry is too good to pass up. Likewise, I'm always telling bankers about resources that can help them with odd problems or telling vendors about bankers with odd problems.
The end result? No advertising, no cold-calls. We're too busy working with the banks [that] call us up or come to our booth at trade shows.
I can't see this approach working too well if you sell french fries. But I think it applies broadly to software, even beyond vertical niches like banking. Whether it's Google's amazing free tools, Fog Creek Software's great articles about making better software, or Microsoft's terrific books like Code Complete, the companies I admire try hard to move past the quick buck and be helpful.
It only occurred to me afterwards, but there's a rational reason for using "Do Unto Others" beyond making your mom happy. We've all bought software (or signed up for a bank account) and felt jerked around afterwards. Maybe it's the $24.95 you paid to find out how to get around a simple bug in your checkbook software or the time your bank charged you extra to see a teller. All of these experiences left a bad taste in your mouth and subconsciously make you seek out products that you can trust.
Software is an extreme example of this effect because so much of the total value is visible only after you work with the program for a while. Will the user interface be easy to master? Will the company help me use the software as effectively as possible? Is the software tested well for bugs? Will the company improve the software when customers point out the inevitable flaws?
Help people out. Help out your customers. Lend a hand to potential customers. Assist other vendors. Even help your competition. Not only will you have more fun, but it's your best shot to let the market taste how good your total product is.
Q. You sell a "big-ticket" item to a market famed for its conservative nature. How do you get them to try your product?
A. Everyone likes a free taste. But for our software, a trial period is pretty expensive to set up; you need to train the customer, work with the IT department to set up data feeds, etc. We're thinking about it, but for the moment, we give other "tastes." The articles in trade journals help people understand what we are doing and give some free advice to apply without buying the software.
We also sell with a 100 percent money-back guarantee, which is unusual in big-ticket software. We're pretty confident in our product and are happy to absorb as much risk as possible. We are even investigating moving to "zero-cost pricing," where we charge nothing for software or maintenance but take a cut of realized profits. This is complicated to measure accurately and audit, but I think it will be a real winner. Banks like reducing their risk and like buying from companies that stand strongly behind their products.
Finally, it's important to be able to back up your promises with some financial muscle. It looks increasingly unlikely that we will need more funding than our angel round; good software is strongly profitable. But there's a certain comfort in having relationships with the VCs. If we ever needed to, we could get a much bigger round of financing.
Q. Would Profitdesk.com succeed as well as it has if you'd not established credibility and "karma points" with your market?
A. We're young and loathe to give categorical statements on limited experience, but I strongly believe "no." Credibility is just that important.
SmartRate by Profitdesk
Q. How do you market to billion-dollar companies?
A. In some ways, marketing to large companies is easier than marketing to smaller companies or consumers. If I want to drive demand for an iPod, I better have the marketing smarts of a Steve Jobs. It takes genius to break through the constant surf of companies shouting to the general market. And let's face it, there are plenty of interesting products for the mass market.
However, there are many small niches within any large company. For example, banks have experts that know all the technicalities behind escheatment (turning over dormant accounts to the state), the Graham-Leach-Bliley Act (protecting consumer data), and a thousand other specialties. Only a few people in the corporation know much about each niche. The flip side is that there are only a few products vying for their attention. The product that hits their needs can easily get noticed.
Otherwise, big companies are like any other small companies or consumers—only more so. You have to do everything extremely well, or the decision makers wonder about the risk in working with you.
I believe every detail is important, from professional artwork in logos and brochures to a short, meaningful URL to toll-free phones, answered by live people. These are all important signals that you care about being easy to work with. By implication, your actual product is at the same level (or will quickly get there if problems are found).
This perfectionism can mean that marketing projects take a while to go live (for example, we've been rolling around blog ideas for six months), but sometimes it's better to offer less. You're forced to plan better and sometimes even decide that a project is not worth the effort of doing well.
A perfectionist attitude is especially tricky in the software world, where the best is truly the enemy of the good. Many software projects have gone south because they could not be released until everything, including the kitchen sink, was thrown in. The best software is built through evolution. Figure out minimum functionality to make the customer happy, find out what they really want once they use it, and put it in the next release. So, while an eye on quality is very important, shipping is the only way to eventually get truly great quality. You almost need to be schizophrenic to market and build with these somewhat conflicting attitudes.
PHIL WRIGHT, FOUNDER, CROWNWOOD SOFTWARE
Micro-ISV: Crownwood Software
Web site: http://www.dotnetmagic.com
Blog: None
Interviewee: Phil Wright
What it sells: DotNetMagic, a .NET component.
Location: Berkshire, United Kingdom
Q. Can you tell me a bit about the software or Web services or products your micro-ISV sells?
A. My micro-ISV consists of selling a single product called DotNetMagic. This is a library of user interface components for .NET developers and is written in C#. It contains easy-to-use gadgets such as menus, toolbars, docking windows, and other controls that are not present in the base .NET Framework. It actually started as a spare time project in order to learn all about .NET and C# when the technology was first released. I gave it away for free during the first two years, but as I started receiving requests from companies to include it in commercial projects, I decided it was time to start charging for it.
Q. Is revenue a) less than what you hoped, b) about what you predicted, c) exceeding expectations, or d) you're going to buy a small Pacific island later this year? Can you share your sales per month?
A. I only ever expected a small revenue for a few months. It has exceeded this by generating consistent sales of around $12,000 per month.
Q. What has been the biggest surprise in getting your micro-ISV started?
A. It is actually much easier than you ever realize. I always imagined that only a few entrepreneurs actually start a company or try to create products to sell. With the advent of the Internet just about any programmer can create a product and sell it through a simple Web site. My big fear is that the mass of programmers out there will also realize this, and suddenly the competition will become much more intense.
DotNetMagic by Crownwood Software
SANJAY BHATIA, FOUNDER, IZENDA
Micro-ISV: Izenda
Web site: http://www.izenda.com
Blog: None
Interviewee: Sanjay Bhatia
What it sells: Data engine and development framework.
Location: Atlanta, Georgia; United States
Q. Can you tell me a bit about the software or Web services or products your micro-ISV sells?
A. Through our partnerships, we have done nearly a hundred projects in the past three years. Our original focus was content management and e-commerce. Over the past 18 months, we have made an aggressive transition to business intelligence and reporting, and most of our business is now in that area. Our strategic focus has also shifted from small add-on projects to solutions that critical parts of an entire company or department depend on that are highly visible to decision makers. Our focus on business intelligence has also allowed us to create value by developing reusable components that can be used by multiple clients. We are now in the process of selling our platform on the open market to developers.
You can find a customer list here: http://www.izenda.com/iws/Company/Customers/tabid/91/Default.aspx
.
Q. Is revenue a) less than what you hoped, b) about what you predicted, c) exceeding expectations, or d) you're going to buy a small Pacific island later this year? Can you share your sales per month?
A. Revenue has certainly exceeded expectations. I'm not heading to an island quite yet but will be doing trips abroad every year for the foreseeable future. The great challenge has been revenue stability. While this month will be the first unprofitable month we've had since the company's birth (due to a project being delayed because of Katrina), revenue has been very volatile with good months bringing in five times as much revenue as bad ones. Cash flow has also been a challenge. We've aggressively financed our customers, which has been great in the long run and on paper but has caused some crunches when it comes to the bank balance.
Q. What has been the biggest surprise in getting your micro-ISV started?
A. The biggest surprise on the positive side has been learning the power of using networks of partners. I've gone from spending 100 hours a week keeping the business running to 20 hours (and spending the rest of the time growing it). I never would have imagined that you could get so much accomplished by spending the entire day doing emails and going to meetings. These were things that used to feel like a waste of time. I come from a coding culture where progress was measured in lines of code and features implemented. Instead of delivering as much code and as many features as possible, I now think in terms of delivering the smallest set of features that would create the most value and keeping the customer satisfied with as little code as possible.
On the negative side, it has really surprised me how quickly other small business that I've worked with have gone out of business or had serious cash flow problems. It seems a lot of entrepreneurs don't understand basic accounting and finance on a strategic level. I've had a number of customers as well as partners who have run out of money. I am a believer in erring on the side of caution and setting up capital and lines of credit that you don't intend to use but are there in emergencies. I know my HELOC (Home Equity Line of Credit) has certainly saved us more than once when a customer doesn't pay a bill on time.
Q. How many people work/own Izenda?
A. I own 100 percent of it. I have one part-time intern. We work with about a dozen people on different projects. In a way the model is that of a virtual corporation that utilizes networks of partners to function as a larger firm.
Simplido by Izenda
Q. When you say "Through our partnerships, we have done nearly a hundred projects in the past three years," can you tell me a little more about your partnerships?
A. Mainly we've partnered with companies like DreamStudio and MediumBlue and let them focus on their core competencies while we've focused on the back-end software. We've also worked with other one-person companies.
Q. How do you develop/nurture these partnerships?
A. I meet people over the Internet and at networking events. I just met somebody at a TAG (Technology Association of Georgia) event [who is] developing an application builder that complements our Ad Hoc technology well. I'm meeting him next week to brainstorm ways to do business together. It's a lot like dating; you just have to put yourself out there and play the numbers game and not waste too much time with connection that will be unprofitable. If there is a system to it, I've never discovered it.
Q. How do you market Simplido?
A. In truth, we have abandoned Simplido and are just now bringing it back as a part of our grand strategy of a software development methodology that allows for model-driven architecture based on relational metadata. Simplido was the first piece. It did not really make business sense as an independent product because Microsoft had announced ObjectSpaces, which would render it obsolete. As it turned out, ObjectSpaces somehow got lumped in with the Longhorn stack and is therefore not shipping with Whidbey. If I had known this two years ago, we would have invested in and marketed in Simplido more aggressively. It's really interesting how powerful an announcement from Microsoft is. I used to work there and wish that I had investigated what was on the back burner better, but that was a different life.
We used to market it with Google AdWords and sites like ASP.NET.
Ad Hoc is what's next for us. The product makes sense on its own, and IT people (business people, that is) buy into the concept within seconds of seeing it in action. You don't have to sell the concept. People are getting accustomed to editing their own Web content without needing IT; we are tying to empower the user in the reporting space.
Our marketing plan includes bloggers, articles, AdWords, SEO, and potential partnerships with other component vendors. It depends on how the organic growth works. Really, that's the part we are figuring out.
We have some Ajax-based forms technology that is in the pipeline. These three pillars will make possible what I consider the Holy Grail of software flexibility. Essentially this means that if the client needs a new field added, you add a new field to the database (possibly with additional metadata), and the application detects this and adapts without a recompile or reconfiguration. While purists and n-tier folks may cringe at this concept and big consulting firms will fear the lost billable hours, it is really what small enterprises need. You want to add a field, do one thing (maybe even from the browser), and poof!—you have fields, validators, [and] report fields, and it's all laid out cleanly and just works a second later. We have this working for one of our main customers, but it needs a lot of work before it can become a reusable technology. We believe we can extend the VS/SQL 2005 generation of technologies to make this a reality.
While we have a long-term vision, each component of that vision makes sense on its own and [needs to] be a useful concept. Every decision I make takes the future into consideration but solves an immediate client need today. I think that's where a lot of software entrepreneurs fail. They let their vision blind them from taking care of immediate customer pain. It's a challenging balancing act indeed.
TONY EDGECOMBE, FOUNDER, FROGMORE COMPUTER SERVICES
Micro-ISV: Frogmore Computer Services
Web site: http://www.frogmorecs.com
Blog: http://www.pdf-software.com
Interviewee: Tony Edgecombe
What it sells: Network printing software.
Location: Faringdon, Oxford (a small market town about 15 miles from Oxford); England
Q. Can you tell me a bit about the software or Web services or products your micro-ISV sells?
A. I sell software [that] helps businesses improve their network printing.
This ranges from keeping archives of all printed documents through to applying rules to printing, for instance, blocking big documents from going to small printers and sending them to a larger centralized printer instead.
Q. Is revenue a) less than what you hoped, b) about what you predicted, c) exceeding expectations, or d) you're going to buy a small Pacific island later this year? Can you share your sales per month?
A. It took longer than I originally hoped to reach a level [that] would support me and my family. I started at the beginning of 2002, and my first orders arrived in the middle of that year. It was another 18 months before I could drop the contract work I was doing and focus on my own products full-time. This wasn't a get-rich-quick scheme for me, though; I saw it as an opportunity to get some balance back into my life so was quite happy to plug away for as long as it took to succeed.
The Pacific island is still some time away yet!
Q. What has been the biggest surprise in getting your micro-ISV started?
A. The biggest surprise is the community support I have found; groups like the ASP or AISIP have newsgroups with valuable discussions and regular conferences. Just by following these groups and going to the conferences I have picked up useful knowledge [that] has increased my sales volume.
Q. How do you market your products?
A. The most successful method for me is Google AdWords. Because my software is fairly specialized, most of the keywords aren't that competitive, and so I get a really good return on my investment. The second biggest source comes through some articles I wrote [that] are related to the features in my software; these have been picked up by the search engines and drive a lot of traffic to the Web sites. After that it's a mixture of press releases, posting to download sites, and some activity in forums and newsgroups. I'm also planning some very targeted direct mail in the near future.
Q. Do you find that many of your customers buy more than one product over time?
A. Not very often. They usually have an immediate requirement or problem [that] needs to be solved. Once they have bought the software, that problem is out of the way. Most of my software is installed on servers so it doesn't get rolled out in the same way a client tool like some zip software would.
Q. How has PayPal worked for you in the United Kingdom?
A. Most of my orders come from businesses, and they don't use PayPal very often.
Q. What type of business entity are you (sole trader, etc.), and how is that working out?
A. The business is a UK limited company. There are some small tax advantages to structuring it that way in the UK, and it does mean you have limited liability. My house isn't on the line in the way it could be if I was a sole trader.
Print Distributor by Frogmore Computer Services
Q. Did you find it manageable to deal with the various government bits of red tape?
A. I do moan when I'm dealing with the paperwork, but in reality it's not that bad in the UK; a lot of the tax forms can be handled online now, which makes things a little easier. I deal with most of the financial work myself because I think it's important to understand just how much money is coming into and going out of the business. You do need to be quite thorough to do this, though. If you are the sort of person who never keeps the tickets from the cash machine, then you need to get an accountant doing it for you.
VLADISLAV UKHOV, FOUNDER, CETERA
Micro-ISV: Cetera
Web site: http://www.cetera.ru/
and http://www.cetera.ru/english
Blog: http://www.ageofWeb.ru/blogs/users/svyatoslav
Interviewee: Vladislav Ukhov
What it sells: E-commerce software.
Location: Yaroslavl, Russia
Q. Can you tell me a bit about the software or Web services or products your micro-ISV sells?
A. We develop Cetera e-commerce software, a small e-commerce product for SMB that's written on PHP+MySQL. It's very simple to use and well-documented. It costs $450 USD with a lifelong guarantee and updates.
We also provide development, integration, and support services for this product.
Q. Is revenue a) less than what you hoped, b) about what you predicted, c) exceeding expectations, or d) you're going to buy a small Pacific island later this year? Can you share your sales per month?
A. Licenses: less. Services: exceeding expectations.
Q. What has been the biggest surprise in getting your micro-ISV started?
A. First year: We need to think more about sales and marketing about less about development. Second year: Support is the key! Development waits.
Q. It sounds like the services customizing and installing your product are exceeding your revenue expectations.
A. Yes. A fact.
Q. Would you recommend this approach to other micro-ISV, or do they need to be selling a server-based app to make it work?
A. In our market (Russia, CMS products), revenue is in Consulting > Projects > Products > Support. As I see [it], customers in other countries also value projects more than products.
Q. How do you market Cetera?
A. • Word of mouth
• Products advertising in search engines (Yandex-Direct is a local Google AdWords analog) with a budget near $100 to $150 USD/month
• Two blogs (15 to 20 posts per month)
• Cold calls (near 30 to 40/month)
Nothing more. We don't have any printed materials and so on. This doesn't help in our market. Our Web site is not a sales channel, only a sales support instrument.
Cetera CMS by Cetera
The final group of five micro-ISVs have achieved a level of success that is definitely not micro. Some have grown out of what you could call micro; some have realized their hard-won equity and been sold.
At this level, it's harder to make generalizations about what works and what doesn't. One clear fact from all these interviews is, as Joel Spolsky puts it, you have to find a little bit of magic to make it to this level.
DHARMESH SHAH, FOUNDER, HUBSPOT
Micro-ISV: HubSpot
Web site: http://www.hubspot.com
Blog: None
Interviewee: Dharmesh Shah
What it sells: Platform for solo entrepreneur Web portals.
Location: Boston, Massachusetts; United States
Q. Your Web site seems a bit vague right now. How come?
A. My site's URL is http://www.hubspot.com
. The information there is intentionally broad and vague (as the product has not been launched yet). This is my third micro-ISV start-up. [The other two are]
http://www.pyramidonline.com
.Q. Can you tell me a bit about the software or Web services or products your micro-ISV sells?
A. The company is building a platform to allow for the efficient creation of powerful Web portals for solo entrepreneurs (and other micro businesses). The idea is to do what Salesforce.com
did for mid-market CRM but broaden the functionality and focus on the micro-business market (fewer than ten employees).
Q. Is revenue a) less than what you hoped, b) about what you predicted, c) exceeding expectations, or d) you're going to buy a small Pacific island later this year? Can you share your sales per month?
A. Revenue has not kicked in yet but will begin to do so in November (when the product goes through a "soft launch" to our alpha/beta customers). The pricing will be done on a per-user subscription model. We also have some creative ideas for marketing the product (given the challenges of distributing to the small-business community).
The first product (SoloPortal) will likely also be submitted to the MIT $50,000 Business Plan competition this year (which will help further refine the idea and get some additional visibility). I'm a graduate student at MIT, and my thesis work will be focusing on the solo entrepreneur community and its economics.
From a personal perspective, I have recently sold my enterprise software company (which has given me the luxury to go back to school and focus on "doing it right" for HubSpot). In this vein, my goal is to focus intensely on creating the right solution (something I would be proud to put my friends and colleagues on) and later worry about the revenue, earnings, and other details.
Q. What has been the biggest surprise in getting your micro-ISV started?
A. The biggest surprise I've found in getting the company kicked off is the difficulty in "narrowing the focus" of the product offering. I think it is a major temptation to create as broad a solution as possible (i.e., build something that millions of people could use), but my business experience tells me a narrow, focused solution on a specific segment of the market is much more likely to succeed. However, it's hard to figure out which customers to focus on first (example: lawyers, software companies, graphics design firms, CPAs, etc.) I'm still not there yet in terms of picking an early vertical.
Q. Why focus on the micro-business market when it seems most attention is on enterprise?
A. I'm currently a graduate student at MIT. One of the courses I'm currently taking is "The Software Business," taught by Michael Cusumano, who (literally) has written a book on the topic (called, The Software Business). Having started, grown, and sold a company in the enterprise space, I understand the pros and cons of such a model. The big issue is that competing for enterprise customers is a difficult challenge. Strategically, the customers have too much "power" (since likely, in this space, the customers can to a large degree dictate their terms). As such, it makes it difficult to create a revolutionary product. You start out innovating, but ultimately, your largest customers (from which most of your revenue is coming) have an inherent need to request "incremental" features and improvements. They are not looking for a revolution. And, more often than not, small businesses with large customers must listen (if they are to survive).
On the other hand, the micro-businesses present a completely different set of challenges (which is often very conducive to technology founders/CEOs). In this market, success is largely driven by one's ability to first narrow the focus and create a solution that is compelling for some discrete set of clients (in most cases, you are not replacing an existing solution but creating an entirely new "market"). After that, the challenge is being able to "generalize" the solution so that you can shift to "adjacent" markets over time. This particular model has major, major risks—but also a very large upside.
Though the small-business market has historically been impenetrable (even by the largest of software companies), I think a number of things have now changed the landscape:
HubSpot platform, HubSpot
Q. What do you see as the challenges of distributing to the small-business market?
A. The biggest challenge in distributing to the small-business market is that the clients are highly "fragmented." For example, in my enterprise software company, it was easy to reach 90 percent of my target market simply by going to two trade shows a year. That is simply not possible in the small-business market. As such, this market takes a lot more creativity to reach efficiently.
The fact that price points are very, very low (by necessity) makes it imperative that the cost to acquire customers is maintained at the lowest levels possible. This precludes "in-person" meetings (at least one-on-one meetings) and many other forms of traditional sales/marketing.
Q. What would you advise other micro-ISV (in school or out)?
A. My biggest advice to micro-ISVs [is]
JAMES SHAW, FOUNDER, DOZING DOGS
Micro-ISV: Dozing Dogs
Web site: http://www.dozingdogs.com
Blog: http://www.dozingdogs.com/news/blog
and http://www.coveryourasp.net
Interviewee: James Shaw
What it sells: Content management system.
Location: Roswell, Georgia; United States
Q. Can you tell me a bit about the software or Web services or products your micro-ISV sells?
A. Dozing Dogs sells a content management system written in ASP.NET. Source code is available, and our Enterprise Edition sells for $999. Our Personal Edition sells for $249.
Q. Is revenue a) less than what you hoped, b) about what you predicted, c) exceeding expectations, or d) you're going to buy a small Pacific island later this year? Can you share your sales per month?
A. a and b. Less than what I hoped but about what I predicted. I'd sooner not share actual sales, for reasons discussed next.
Q. What has been the biggest surprise in getting your micro-ISV started?
A. The biggest surprise was recently accepting an offer to sell the company. I never thought I'd do that, but the right offer came along at the right time. Although all rights are being sold, I will continue to work on the code to integrate it with their product. So, I get a big wad of cash plus the ability to work on my product full-time without the sales/marketing/support pain. It has just been me doing the whole thing (with some independent contractor help here and there), so this next phase is great.
Q. I see you have both a company and a personal blog. Which came first, and how has it worked out for you and where do you get the artwork for your personal blog?
A. The personal blog came first. I wanted to blog about the creation of the product and company. It has been great and gave a place for the old fans of Coveryourasp.com (my classic ASP site) to go to catch up. I heard from many people from the "old days of CYA" (i.e., 2001–2002), and it was great to get to know them again. As for artwork, I just made something up, probably loosely based on http://www.ourroswell.net
, which is the city network that my CMS evolved from.
CMS V2 by Dozing Dogs
Q. Can you tell me more in general terms about the sale—did they find you, were you recommended, etc.?
A. I have always worked full-time as a C++/C# programmer/manager throughout my ISV time. However, my employer moved to Denver from Atlanta, and I didn't want to move. I worked from home for a few years, which was a huge benefit for running an ISV, but then management changed and I was out of a job. No worries—being an ASP Insider has made me many friends, and they put the word into their companies for me. I ended up at Telligent because I knew all the principals there quite well.
Q. How did you determine what a fair price was for your micro-ISV (not so much numbers, but method)?
A. As for the sale, it came from necessity. There was too much possibility of conflict of interest with Telligent's Community Server product and my CMS. They wanted a CMS for CS, so as soon as I got used to the idea of giving up my baby, it all went smoothly. Fair price? LOL, I simply picked a number that would be enough for me to give up my dream. We worked from there, and Rob came up with a formula that fitted it. In the end, we both wanted the deal so it was easy for both sides to compromise a little. It was all over in a few emails—the hardest part was the week I spent wondering if I wanted to do it at all.
Q. Can you explain this a little more? It sounds like they forced you to sell!
A. There was no forcing involved, but it was obvious that working at Telligent was going to create a conflict as their road map (http://communityserver.org/i/roadmap.aspx
) showed them having CMS in v2.0. I had to choose: continue with Dozing Dogs and another "normal" job or a great job with people I knew and respected (mostly fellow MVPs and ASP Insiders, http://www.aspinsiders.com
) and sell my CMS product to them for a very good price. It wasn't a hard decision in hindsight, but you'll probably understand how it can be difficult to give up "your baby." I had to really step back from Dozing Dogs and look at it objectively. I pretended I was advising a friend about his business, and in fact I did get advice from other ISVs, notably good friend Dave Wanta at Advanced Intellect. It took a weekend probably to realize which was the better route.
MARK HOFFMAN, CHIEF TECHNICAL OFFICER, AUTOREVO
Micro-ISV: AutoRevo
Web site: http://www.autorevo.com
Blog: None
Interviewee: Mark Hoffman
What it sells: Web-based auto dealer inventory system
Location: Plano, Texas; United States
Q. Can you tell me a bit about the software or Web services or products your micro-ISV sells?
A. Our product offering is a Web-based tool, billed as a monthly service, that allows auto dealers to manage their inventory. With AutoRevo, a car dealer enters a vehicle's VIN and its description, and that vehicle can then be sent to eBay Motors; classified Web sites such as AutoTrader.com, Vehix, [and] Cars.com; and their own Web site. The benefit to the dealer is we provide one place for them to manage their inventory and Internet sales.
Q. Is revenue a) less than what you hoped, b) about what you predicted, c) exceeding expectations, or d) you're going to buy a small Pacific island later this year? Can you share your sales per month?
A. Revenue is about what we anticipated.
Q. What has been the biggest surprise in getting your micro-ISV started?
A. Biggest surprise so far is the difficulty in finding qualified developers. I've recently sifted through over 60 résumés, and only a handful was worth further consideration. I've interviewed people who claimed to be senior-level .NET developers who couldn't answer even the most basic technical questions. I've spoken to people who claimed on their résumés to be "great communicators" [who] could barely construct a single sentence in English. With all of the "doom and gloom" you hear from developers about the job market, I was surprised to learn that the truly talented developers are just as hard to find as they were in the late '90s during the boom. I really expected to be able to find people fairly quickly, but I soon learned about the true state of the job market.
Q. What has your background before launching your micro-ISV?
A. There were three of us who started the company. Myself and another partner worked as consultants at an IT consulting firm as developers for the same client. He had a friend [who] owned a car dealership, and the three of us started the company.
AutoRevo by AutoRevo
Q. Do you think that focusing on a very specific market made it easier or harder to get your revenue numbers where they are?
A. The fact that we focused on independent car dealerships was crucial to us getting traction quickly. We were also very fortunate in our timing. For years, small dealerships (think used-car lots) could compete only locally. Along comes the Internet and eBay Motors, and suddenly they can get national exposure. The problem was, car dealers are very nontechnical, and they didn't want to have the ability to create their own Web site or automate pushing their inventory to eBay or to the major classified sites. Large dealerships, such as your giant local Chevy dealerships, have the money and the expertise to do this themselves. The little guys can't. That's where we stepped in.
Q. How do you market to that market?
A. Our primary marketing method is direct sales via the phone. It's anything but high-tech, but it's effective for this market. You can always get a car dealer on the phone, unlike most other decision makers. We have a customer referral program that is quite effective because car dealers interact with each other quite a bit, and it's an easy sale when another dealer recommends us. We're also featured on eBay Motors Products and Services page, and that generates quite a bit of traffic.
Q. How many people besides yourself are there at your micro-ISV?
A. Today, there are four other people here besides myself. (The partner [who] owns the dealership is basically a "silent" partner and continues to run his dealership and isn't directly involved in the day-to-day operations.) We have a salesperson, administrative assistant, a developer, and my other partner. My partner and I work together on the business decision making, marketing strategies, the direction of the product, etc. He primarily focuses on the day-to-day operations while I focus on the development of the product.
BYRON MATHESON, FOUNDER AND CEO, CLEARNOVA
Micro-ISV: ClearNova
Web site: http://www.clearnova.com
Blog: None
Interviewee: Byron Matheson
What it sells: RAD platform for Web 2.0 applications.
Location: Alpharetta, Georgia; United States
Q. Can you tell me a bit about the software or Web services or products your micro-ISV sells?
A. ClearNova sells a Rapid Application Development (RAD) platform for building Rich Internet Applications (RIAs) called ThinkCAP. With that statement full of techno-babble and acronyms, ThinkCAP is a development tool for quickly building Java applications that utilize the hottest Internet technology: Ajax. Think of client/server-type applications but Web-based. Google has been at the forefront of the Ajax movement with the Google Maps project. (We have a demo of an Ajax-enabled application on our Web site…pretty cool stuff, even if I say so myself!)
Q. Is revenue a) less than what you hoped, b) about what you predicted, c) exceeding expectations, or d) you're going to buy a small Pacific island later this year? Can you share your sales per month?
A. Revenue is about what we expect this stage of our growth. We sell both our product ThinkCAP and professional services using ThinkCAP.
Q. What has been the biggest surprise in getting your micro-ISV started?
A. How much of a "herd" mentality there is in the software development space—Ajax being a great example.
How much indecision there is in corporate America. When we can deliver a proof of concept in two days that has a four-week timeline by the customer development team, and we can't get management to make a decision to buy…amazing! We have done this on several occasions but find that IT managers are afraid to take chances on a small company with a superior product!
Q. How many people work at ClearNova?
A. Thirty people and growing!
ThinkCAP by ClearNova
Q. Given your company's focus and success, would I be wrong in thinking you think desktop apps are obsolete?
A. Desktop apps will be around awhile, but Web technologies are moving closer to client/server usability. Apps that used to require desktop app productivity are beginning to be delivered via the Web. What I think is obsolete is the current page-based Web-based applications. Once users begin to experience Ajax-enabled Web applications, they will shun those applications delivered under the current page-based format. I believe that we are going to see a major push to rewrite our current Web-based applications; obviously we believe we have the product to assist in that effort!
Q. How do you market ThinkCAP?
A. Currently, our marketing exists of Google ads, speaking engagements, and great PR campaigns. I am shamelessly including a recent article headlining ClearNova with Microsoft. It has had an incredible impact on inquiries. The article is at http://www.infoworld.com/article/05/09/15/HNajax_1.html
.
Q. What's it like marketing to developers?
A. I have always said that developers are like painters: there are a lot of developers [who] can "paint" but few "artists." (I used to be a developer, so I can say this!) Unfortunately, everyone is an artist in their own mind! So where am I going with this…most Java developers have a "not invented here" mentality. The reason we are as successful as we are is because most Java shops are failing to deliver. To be honest with you, we have had an uphill battle convincing the Java development community to adopt a product like ThinkCAP. They have a herd mentality, which is good if you're leading the herd and hard if your not. Fortunately for us, Ajax is on fire right now, and we are actually leading the herd right now. ([Let's hope] we won't get stampeded.)
JOEL SPOLSKY, COFOUNDER, FOG CREEK SOFTWARE
Micro-ISV: Fog Creek Software
Web site: http://fogcreek.com
Blog: http://fogcreek.com
Interviewee: Joel Spolsky
What it sells: FogBugz, a Web-based bug-tracking application; Copilot, a remote desktop control service; CityDesk, a desktop Web site manager.
Location: New York, New York; United States
Q. Fog Creek Software started as a consulting company first, right?
A. Right. In fact, everybody told me, "No! You should be product!" We always wanted it to be a product company in the long run, but consulting seemed to generate a lot of money, and the margins were huge. And our thought was a consulting firm would throw off enough spare cash so we could develop products and thus sort of build a software company in the background.
Q. Do you think that's a good approach in general?
A. I have to say that I still do, despite the fact that it really didn't work. It makes a ton of sense. Here's the theory: the story you tell yourself is, I get a group of friends. We're all great at computers and computer stuff, but we don't know what to do. So we'll bill ourselves as computer consultants; we'll charge whatever the going rate is—$100–$200 an hour to develop [and] $300 to $400 an hour for top-notch software developers. And that will generate some cash and give us some business, and as we do consulting work, we'll start to discover common themes and problems our clients are having. At some point a lot of clients will say we all need X, and we'll tell them all, "Listen, why don't you let us develop X for you, and instead of charging our full rate, we'll charge you our discounted rate. And in exchange for the discounted rate, you'll allow us to keep the rights to productize the source code we developed for you."
It's a way of finding a business—of developing a product on somebody else's dime and bootstrapping. And as that evolves, you have this big product that your consulting firm exists to implement and customize and enhance. Theoretically, you want to get into a situation where you have a bunch of developers who develop the product and a bunch of developers who customize, install, and sell it and do consulting. And that's a good way to build what I call a consultingware company.
Q. OK, but that's not the same as a product company?
A. Well, if you're disciplined about it, a product does kind of emerge. Now realistically? I say I haven't seen that succeed, but the average start-up—the average venture-funded enterprise start-up—is doing this. They just don't admit it. They think they're developing a product.
So they have their idea: "We're going to make a security widget." So they go around and talk to people, and in order to make their business plan work, they're in the $100,000 and above per installation ticket price for their product market. And at those prices, there's a long sales cycle, and the customer demands all kinds of things, and half of the customers don't really need the product. They really kind of need to have some developers working on this problem they have, and they're going to pretend to buy the product in exchange for having basically put your developers on retainer.
Q. A commonality of wrong expectations I guess?
A. Kind of. And they deserve each other. But this happens a lot. There's a lot of companies out there with very specific software needs and just no ability to get the program no matter how much they're willing to pay. They can't find good programmers to develop it. Somebody walks in the door and says, "We're starting a software company. And we're going to make a product that kind of does this, and we have no customers yet. And we want you to be our first customer, and it's $500,000 plus $400 an hour for developer time to get it to work for you."
And then they say, "OK, how much developer time can I get?"
Q. Now does this actually really happen?
A. That's what usually happens with these companies.
Q. So Fog Creek Software is a product company. How much are you making?
A. Well, we're under $5 million a year in sales.
Q. And how many customers do you have?
A. These days, we're getting about 1,500 a month. But the point I wanted to make is that we're a company that sells to about 1,500 people, but these companies sell to about one person a month, if they're lucky, and they make $200,000 doing that.
For us, it's nice and stable and easy, and we don't have to do anything for anybody to get that money. But for them, in order to make a $200,000 sale to one customer, or a $400,000 sale to one customer every two months, it's a big rigmarole. and it's very dependent on it. And they really need that money, the customer asks for all sorts of extra things, and they promise them. And they wind up developing one-off versions for each customer.
Q. And at that point, you really are talking consultingware….
A. Yes, and they think of themselves as a product company, but they're slipping down the slippery slope of not having a product. It's the typical VC-enterprise software model. I don't know why it's so common or how they get out of that cycle. They have a ton of customers they are losing money on because they have to customize their product so much, and you have customers who are so good at negotiating that [the software company] is seriously losing money on the deal, but they convince themselves that they will be a good reference customer or something.
Q. They'll make it up on the next one—famous last words….
A. And they never do.
Q. What about your software? You're selling to about 1,500 people a month; that's kind of the midpoint. Do you think midpoint to lowpoint—the under $80 piece of software—works? Or is it only at the midpoint can you see the type of growth that Fog Creek has experienced, doubling revenue every year?
A. If we were doing shareware at the 20 to 30 buck level, it would take forever to build that kind of volume. I think we have a relatively low-priced product, but it's not consumer priced really. I'll give you another example, and numbers for it, but the numbers are completely meaningless since it has been out only a month.
We have the Copilot project—remote desktop software (http://www.copilot.com
). And it's about ten dollars a day. This is a consumer kind of $10 product. And right now, we're getting about eight users a day. It sounds low; it's significantly up from five users a day. And if you look at the trend, it's a really significant statistical growth. Although the number of users is tiny, the trend line is surprisingly clear, and the number of users we get per day is surprisingly constant.
Copilot by Fog Creek Software
Q. Let me ask you about Copilot—in a way it's a micro-ISV product.
A. It's a classic micro-ISV product, and I'm happy to talk about the numbers because anybody can figure them out. The total cost to me to develop Copilot was $80,000. That $80,000 included computers for four programmers (top-of-the-line Dell PCs with two CPUs, 2GB of memory, RAID drives, and dual 20-inch monitors, costing about $5,000 each), salary for four interns, [and] the housing for the four interns, which basically brought the cost of these interns up to that of full-time employees, and we paid an undisclosed amount of money to buy the domain name, Copilot.com, and that was about it.
So, total cost was about $80,000.
Q. It's going to take a long time to pay that off if you're talking eight sales a day….
A. Basically we figured if we get to about ten sales a day, we break even in about two years. That's not an unreasonable amount of time to make your money back; everything after that is profit. And we've already about hit that with organic marketing [and] no marketing except our blog. [This is] insignificant amounts of marketing, so we're getting insignificant sales, but we are getting organic growth. People are elated when they use Copilot, and it solves their problem in a brilliant way. It saves them $150 to call a tech support person or one of those "Geek Squad" kinds of people or to spend two hours on the phone trying to explain something that they can do in two minutes with Copilot. And these people tell their friends.
Q. So when you say organic growth, you mean there is no marketing?
A. Right. Just word of mouth. So, I already absolutely convinced we're going to make our money back. And the rate at which we've been growing in just the first month or two we've been live convinces me we're going to do a heck of lot better than just make our money back.
Q. That's a good thing. Are you going to stick with word of mouth, or are you going to actually start marketing it?
A. I think I'll stick with the word of mouth.
Q. OK, but you already have the advantage that you have an established company, you have other products, and you have a very influential blog. That's lots of presence. Let's say you're living in the sixth-floor walk-up and you've just come out with this product through the dint of working after your day job for six months, would you go word of mouth, or what would you do at that point?
A. No, that wouldn't work. Let's put it this way. The things that the geeks don't understand about sales, marketing and business, etc.…the first thing they don't understand is that there's three components here: there's PR, there's marketing, and there's sales. And they confuse those things.
PR is generating free or low-cost awareness that your product exists, through publicity. And the distinction between PR and marketing is that PR is on the editorial side; it's not on the advertising side of the page. When you look at a newspaper, any child can take a newspaper and show you want the ads are and what the editorial is, and if your story is on the editorial side of the newspaper, that's PR. And there's an editorial side of the Internet too.
Q. Do blogs fit there?
A. Tell me whose blog. On my blog, I don't attempt to blog for the purpose of marketing.
Q. So PR is free and low price. It may be your blog, or it may not be.
A. I don't think a blog can do marketing.
Q. OK, how come?
A. Because who's going to read it? How are they going to find out about it? Who's going to come to your blog?
Q. Well, what would you say about Joel on Software then?
A. Ah! You see this is sort of different. Joel on Software is editorial not on the subject of my business. It's editorial about software development. It's valuable editorial; it has an audience. So, you could tell somebody I've done this little software that does digital photo albums, and you could make a blog and people will hear about it. Well, if you make one of those blogs about your cat, you're going to get the same audience as the average blog about your cat, which is zero or fifty of your friends.
There's nothing wrong with that, and it's a beautiful way to keep in touch with your friends and family. It's just not marketing.
Q. OK, so instead, let's say you are a photo whiz maven, and you put out a blog about photography and all that stuff?
A. If you can make that work for you, that's a great model. And that's my model. But a lot of people can't pull that off, and they have an ersatz-like blog. By ersatz-like blog, I mean they are doing one of two things wrong. One, they are using their blogs in a very pure marketing way, when all they do is talk about their products. They're never even going to get an audience. Usually companies do that.
And then there's that the average programmer doesn't understand the distinction between marketing and sales. Marketing is creating demand; sales is fulfilling it. Marketing is any kind of process that you follow, usually some derivative of advertising or trade shows or Web sites or any of that kind of stuff that exists for the purpose of getting people to want to buy your product. And sales is whatever processes that you have in place that exist for a person once they want to buy your product.
Joel on Software
Going Back to that Sixth-Floor Walk-Up
Q. Going back to that sixth-floor walk-up for a second, this is their first product. It's a company of one. What would you suggest they do?
A. Let's cross off things here. They're not going to be able to do an advertising campaign. An advertising campaign doesn't work that well either. Advertising is very questionable, and I think if you ask people out in the field, they have no idea why they're advertising. Print ads/banner ads don't work.
Google AdWords will actually work, and they'll work very well, depending on how specific your product is. If your product is "nichey" enough and has keywords associated with it, Google AdWords might be a brilliant way to get your product out there.
Q. So the more of a niche product you have, the better Google AdWords work?
A. Right. So if you have a product that keeps track of the genealogy of your dairy herd, Google is going to be the best way in the world to advertise that particular thing. If you want to have an online photo album, forget it.
Q. OK, so on generic or horizontal products Google is less effective?
A. Because there's so much noise for those keywords. And actually, I think that's the best bet for a micro-ISV that wants to bootstrap. Figure out how to narrow your audience. I always make fun of the micro-ISV or the person in the sixth-floor walk-up that's doing a photo album start-up, because that's what it seems they're all doing. Programmers look at these things, and they say, that's easy. If other people are doing it, they must be making money. I can do that. But it doesn't work.
What I've told countless people who've come to me about this online photo organizer (and not one of them has taken my advice, and they should because then they would make a lot of money) is don't make the online photo organizer; make the online photo organizer for professional wedding photographers. It is shocking how much money is involved in that particular industry. So, they have money to spend. It's a much easier marketing target—all you've got to reach is the professional wedding photographers. Where are they? They're all in B&H (a very large professional photography store in New York), in the lights section.
My point being, how do you market to people having digital cameras? That's everybody! There's nothing to grab onto there. But wedding photographers? They talk to each other, they know each other, and you can look in the Yellow Pages and get the 12 wedding photographers in your city and call them up. They spend money on this kind of stuff—they have money to spend on this kind of stuff.
Now you've taken a product you know perfectly well can be used by anybody, and you've narrowed down to a very specific field. And all the sudden you'll start to discover that there are very specific problems that wedding photographers have. I could go on and on about this, but when you take a generic or horizontal product and narrow it, you almost always increase your opportunities as a micro-ISV.
Q. That's one of the best points I've heard yet.
A. It's also probably one of the most nonintuitive. And we made that mistake—that's why I know this. We made CityDesk, and it's horizontal, and lo and behold, FogBugz sold a lot better, and it's a vertical application. And if I had taken my own advice, I would have changed CityDesk, and instead of making it a content management system, I would have made it a real estate listing system for realtors who want to put pictures of their current properties up on the Web. And instead of saying headline, I would say address. And instead of saying author, I would say price range. And I would add a field for bedrooms and bathrooms. And it would be the same product, but first of all I would know how to market it [and] who my audience was, and I could charge a lot more money for it and serve their needs.
And I'm never going to be Microsoft with the content management system for realtors, but again, we're talking about micro-ISVs, and we are talking about bootstrapping. And that's the way you do it.
Q. So, going back, we were talking about the three things that geeks don't get: PR, marketing, and sales.
A. And we don't get them either. Let me talk about them in the context of Fog Creek Software. PR we had—Joel on Software was very successful PR, and that drives awareness and a lot of turnover basically. A blog is a good cheap way to get PR, but there are other ways—you can do things in your community; there's lots of books about guerilla marketing…about getting awareness for your product by doing stunts that cause the local media to show up. Although Joel on Software generated traffic, it took four or five years before we got the interest of Crain's New York Business that now is consistently running stories about us.
Q. It's one of the things you have to work at for a while….
A. Whether by luck or whatever, we're strong in that area because of Joel on Software. Marketing? We are an absolute zero. We've never tried to do it. We don't know how to do it. We don't know if works. We don't believe it works most of the time. A bootstrapping company never had money to pay for ads, so we didn't do it.
Q. So do you run Google AdWords?
A. Nope. No advertising, zilch.
Q. So for purposes of Fog Creek, marketing just isn't a factor?
A. Right. Traditional marketing and advertising and that kind of stuff—we've never had a marketing person; it's never been done here. Now the third part is sales, which traditionally has been a weakness of ours. Only a few weeks ago did we hire a person to figure out the sales thing. And his job is to figure out what procedures we are going to start following and who we are going to hire so we follow up more reliably with people and actually make more sales. That's one of the things we've been bad at, and I hope we're going to improve.
Q. Let's say you had to do it all over again, right now, in 2005. Things have moved along. Where do you see this whole desktop versus software as a Web service thing going?
A. I wouldn't touch the desktop anymore, but that's just me.
Q. So the whole Ajax, Web 2.0, smart client thing is the end of the story for desktop software?
A. I think so. It's kind of weird. Copilot is software as a service, and it's something that just cannot be done on the Web, i.e., remote-controlling somebody else's desktop. Copilot essentially downloads a little app that makes the service possible. That's the core of the service, but it's wrapped in this Web site you go to. And that's what makes it so easy and simple. It makes it so easy for us to upgrade, and every time we find a bug we fix it and get out there the next day, because nobody keeps these little apps on their desktops; they delete them as soon as they're done using them.
So even though it's something that just cannot be done on the Web, we do it on the Web.
Q. Maybe the question isn't desktop versus Web, but service versus product?
A. You know, there are still people with mainframes. And there are still people who go to movies even though we have DVDs; it takes a while for the world to change. And the world does change. I really feel like we had the mainframes, then we had the microcomputers, and then we had the Windows GUI Macintosh generation, and then we had the Web generation. But the Web didn't completely take over from the Windows GUI thing because we really didn't have Ajax, and now we're starting to have Ajax, and I'm really starting to think, a couple more versions of the Web browsers and some clever work, and that will be where the bulk of development is going to be done.
I would say at this point any kind of database in-house development anyone does, it's just nuts to do that any other way than on the Web.
Q. Based on going from you and your partner in 2000 to seven staff now, what should a micro-ISV founder be looking for from that first hire?
A. Well, gee, I'm really good at telling people who not to hire. If you have any kind of doubt, then no. With your first few hires, the more well-rounded they are, the happier you're going to be and the more different holes they can fill. When you're a micro-ISV, you're doing what would be at General Motors 370 different jobs. You're a specialist in all 370. And so, if you're good in only 216 of those, then obviously you want to hire someone who can do at least 100. And that's really what it is—trying to get things covered.
It's sort of a hard problem as a micro-ISV. Chances are the average micro-ISV that doesn't want to do consulting to pay the bills is going to have to hire people before they really have the ability to pay them. They might think they have the ability to pay them based on incoming revenues, but there's no sort of stability there yet. That means you're probably going to have to hire somebody who is willing to take equity and has an entrepreneurial spirit. And that's a whole different type of people than those who answer the Help Wanted ads on Craigslist.
Q. How do you with a small staff of owners and employees maintain who's in charge?
A. That one I can't really give good advice on because it depends so much on the situation and the particular personalities.
Q. What about outsourcing and offshoring? I've read what you've written that a software company that outsources development has given away why they exist.
A. Kind of, yeah.
Q. On the other hand, there's a report out today that one-half of the start-up companies in Silicon Valley outsource….
A. Wow. That's great! It's good to know I'm competing against idiots who have no control of their software—that has an employee pool that has no stake whatsoever in their success. That's great, for me; suicide for them. The only company I can think of that was a successful start-up with employees not in the same country, or wherever, is Skype. They had their employees in Estonia, and they were in the Netherlands. But you know what? They were employees, and they had stock.
Q. So any start-up that goes with outsourcing development is doomed?
A. They have a huge hurdle to overcome because they're working with one hand tied behind their backs. That's really what I feel.
Q. How about the typical way software gets written; you know the "eat pizza and sleep under the desk and work every hour until it's done" approach? Do you think that's the way micro-ISV should develop? Or is there a better way?
A. We did it the nine-to-five, professional way. We took off weekends; we didn't work very hard on weekends kind of stuff. But I think the "sleep under the desk" approach does have a lot to be said for it for the version 1.0 launch. There's something to be said for both ways. The truth is, if I saw a company like Fog Creek, that's starting a company, and they say, "We're not going to work ourselves too hard. We're very strictly nine-to-five company. We believe in the 40-hour thing, etc.," you sort of question the dedication a little bit. Maybe.
I can give you an example. There was a start-up called Westside. It was a start-up created by about 45 ex-Microsoft employees, mostly from the Visual Basic and Access teams. These guys idea was to make something like Access but entirely on the Web. The whole software as a service idea, way before its time, in 1998. They were all ex-Microsoft employees with tons of stock options, so they could afford to go an awfully long time without getting any kind of revenue.
One of the things I remember about them is that they built a company that was operationally almost like Microsoft, in the sense that they gave everybody private offices with doors and windows, and they gave everybody excellent benefits, vacations, and stuff like that. And these people were not like young Microsoft people; these people we kind of like Microsoft middle-age people who don't quite work as hard as the people right out of college who don't have friends.
There's sort of two demographics at Microsoft. There's the people who just got out college [and] don't know anyone in Seattle, and they have nothing else to do but just work around the clock.
Q. You make it sound like a cult!
A. It kind of is. And they love that kind of stuff. And there's a good reason people love to work at Microsoft—they love to program. So they do that for a while. And eventually they find a girlfriend somehow, and they get married, and they have kids, and now suddenly they're all in their Saabs going home at exactly 5:05 p.m., choking up the highways around Redmond, Washington. And that's sort of a second category of people, and those people sort of move into middle management and don't really get as much done as they used to in their golden days. And those are the people who created Westside to a large extent; they were already past the energy stage.
They did build a product, but it took them quite a while. They had all the trappings of success—the very expensive offices. But their product wasn't that good; I don't think they ever had any revenue.
Q. OK, [a couple] more questions. How old are you?
A. 40.
Q. Well, you're past the "don't have any friends/sleeping under the desk" stage. So here's the question: are we too old to do this? I'm 48 by the way. There are people out there who argue that starting companies should be done by adolescents, if not pre-adolescents. Where do you stand on that?
A. I look at these people…very smart people like Aaron Swartz who is starting companies with Paul Graham, who is pre-adolescent, whom I consider to be very smart, but he doesn't know shit. I think my summer interns who are Aaron's age were able to build a whole product and get it selling and shipping (and it's going to break even pretty soon) in one summer because they had a 20-page spec that I wrote for them. All I did was point them in the right direction and say "Go." And if they started to veer off-course, I'd say, "Let me tell you a little story about something that happened to me 16 years ago." And they went back on course.
Paul's young crew, with all power to them, there's something to be said for very young people, because they don't know what they don't know, so they extrapolate like crazy. They can't interpolate like me. And when you're dealing with technology that's changing all the time, it causes them not to be blind to certain things that I'm blind to. So, for example, they would have made a Web content management system while I was still trying to make a Windows content management system, because I'm too old. Because I'm interpolating and they're extrapolating.
Q. Well, you're saying two things at once. Or is there is something to be said for the young person model because they don't know what they don't know?
A. So they invent the future often.
Q. OK, so how about us older folk. Give us some reason not to be whining in to our tank of oxygen over here!
A. Well, let me tell you about how I feel personally. Which is this: what keeps me interested in Fog Creek is the fact that every year it grows, and therefore the problems we have to face on a daily basis change and evolve. And the numbers grow. To the same extent I'm no longer impressed by "Woo-hoo, I made $100,000. That's OK, I made $200,000." To some extent, you might be desensitized by doing the same thing again and again, but the great thing about having a growing company is that the growth itself means that it changes all the time so that your job is different everyday. And you evolve in your job.
Last year, everything for me was about recruiting [and] finding the next generation of people. And then there was the whole summer of Copilot, figuring out how to get interns all excited about launching a quick, fast product in a new business. And right now, for the next few weeks, I am going to be obsessing over reducing the amount of work it takes to process purchase orders and checks. So, the business as it grows creates new challenges for you.
Developing at the Bleeding Edge
Q. On a different subject, should micro-ISVs focus on the latest hot technologies? Should they go where the cutting edge is being cut?
A. Probably not. It probably means you're talking with people at conferences instead of your customers. Your customers probably couldn't care less. Let's say you're doing the photo album software thing. Can you imagine talking to a wedding photographer and having them say, "It would be cool if you had some Ajax here"? They're going to say, "I need a way to convert a color photo to a black-and-white photo at the click of a button." Those are going to be the terms they use.
There is one slight exception: if you're first with some exciting new trend, you do have an opportunity to be talked about or passed around. So you've got this one minute to do something, to catch some PR, if you do something that happens to catch on really fast and draw some attention. But that's not a strong or sustainable advantage.
Q. So your customers aren't interested in the latest and greatest, and it may not give you much of an advantage unless you're first?
A. Right. There are ways of using the latest and greatest building tools to build something cool, but everybody has the same tools too, so it doesn't necessarily get you a competitive advantage.
Q. So how as a developer do you keep up with all the new technology?
A. You don't and nobody is, and that's actually a serious threat to Microsoft at this point. It's something I've said again and again. They are turning out new developer technologies faster than developers can pay attention to them. And they really need to calm down. They are adding features to C# [that] are completely unnecessary, that nobody has time to even read about, so they can give Microsoft the feedback to say, "These features are unnecessary and just clutter your language, and we don't need them."
Q. So what should a micro-ISV developer do?
A. Pick one tool and know it obsessively well, and you're going to get stuck with it for five or ten years. The thing about being an ISV and using technology is that you just can't use the latest and greatest. We're stuck using VBScript because we're not going to port our application. It's not worth it. There's no monetary value in porting to new tools. And so we're using very, very old technology here and it works fine, but it's not the greatest and there are advantages to moving, but there are even bigger disadvantages. So we're sort of stuck with this tool, but that's OK.
When you're an ISV because of the cost of porting [and] because your application is going to go for ten or twenty years, you are almost inevitably going to be using what appears to be crappy old technology at any given time. But that's OK—that's the life you choose. If you want to use the latest and greatest at all times, go do in-house development for an insurance company, because they are always doing a new projects, and you can probably convince your boss who's probably not a programmer that porting to C# 3.0 is the most important thing you can do. He'll probably fall for it.
Q. Any advice on anything for a micro-ISV or somebody who wants to be a micro-ISV?
A. Don't do it, you're crazy!
Q. OK, why?
A. I don't really know how to put it. You have to know some advantage you have. If your idea of why your business is going to succeed doesn't have a little bit of magic in there somewhere, that you're sure is going to work, it's not really going to succeed.
Q. By magic, you mean…?
A. Let me see if I can try to explain this. The reason that I knew that Fog Creek was going to work was that we had the following piece of magic: we were going to do consulting, and that would generate cash that would keep us in business long enough to be able to become a software company. We could fail 27 times at making a software product, and if we succeeded on the 28th time, that would be great. The only possible way that Fog Creek could have closed down is if we didn't have enough consulting revenue to keep even one person fed, which is a pretty low barrier. So that was our piece of magic.
The Skype's piece of magic was that people have to pay per minute for phone calls and yet they have unlimited access Internet, and nobody had made a tool for them to actually make phone calls for free. The Kazaa piece of magic—pulled off by the same people—was you could steal records, and you wouldn't have to pay for your records. And they were successful because of that piece of magic.
The PayPal piece of magic was that when you sign up for the account, you get ten free dollars, which you could then take out. The other piece of magic was that you didn't have to wait for the check to arrive for your eBay purchase.
One more example: the Copilot piece of magic was that most of the code we needed to do the remote desktop feature that's the core of Copilot was available in a GPL [General Public License] open source piece of source code we could use as a starting point. And without actually violating their license in any way, we were able to launch a service with very little effort, and we knew that we had searched the world high and low, every possible Web page, to accomplish this particular feature without complicated setup, and we knew we could make a faster, easier setup.
Just by making the setup easier we knew that we could make something that would solve a known problem in a way we couldn't find that anybody else had solved. So that was the magic there.
The other magic that's going on obviously for Fog Creek Software is Joel on Software—a built-in audience that doesn't cost us anything. It's some extra thing that gives us a competitive advantage that has enabled us to be successful. But if you can't tell me what the piece of magic is your micro-ISV is going to have, it probably won't really get there.
My editor, Jonathan Hassell, rightfully pointed out the first time I submitted this chapter that I had neglected to come to a conclusion in this book. He's absolutely right. As a micro-ISV myself in the "emerging" category, I don't know how the story ends.
But I do know, thanks to all the people who kindly gave of their time for this chapter and all of the other people interviewed for this book, that most likely, the ending is going to be a pretty good one.
Micro-ISVs aren't a business fad that will run its course in a year and disappear. Micro-ISVs are a real alternative to the venture capital companies that come and often go. About one billion people are sharing the same connection called the Internet now, and if you can find a problem just a tiny part of that one billion have, and solve it, you too can succeed.
But you have to try.
Walt Disney, a man who knew a considerable amount about making something out of nothing with some imagination and a lot of hard work, is cited as the person who first said, "Better to have tried and failed than never tried at all."2
Maybe so: of course, it's best not to believe everything you read on the Internet! But one thing is for sure: if you've got the drive and the need and the perseverance, you too can go from vision to reality.
I wish you the best of luck; drop me a line at [email protected]
, and let me know how things have worked out for you.
-30-
(That's old reporter-speak for "You've come to the end of the story; now get out there, and do something!")
3.145.184.117