6

Red Herrings and Long Hauls

What Is, and Isn’t, Difficult About Adopting
the New Tools and Approaches

This chapter looks at the principal challenges of Enterprise 2.0, the main obstacles to an organization’s successful deployment of emergent social software platforms (ESSPs). I have noticed an interesting pattern: before an organization begins an Enterprise 2.0 effort, its leaders typically have a consistent set of concerns about the negative things that could happen. But most of these are not real or serious risks—I call these red herrings.

After an organization begins to deploy ESSPs, its leaders’ concerns often shift to one simple issue: How can we go faster? They are often surprised by the fact that Enterprise 2.0 is a long haul and puzzled that the effort is not progressing more quickly.

This chapter first examines why so many initial concerns are actually red herrings and then explains why deployment of ESSPs is a long haul.

Red Herrings

I’ve noticed that concerns around Enterprise 2.0 fall into two broad categories: fears that people won’t use the newly available ESSPs, and fears that they will. The latter, which stem from the lack of upfront control common to ESSPs, tend to crop up first. When first exposed to these technologies, business decision makers voice concerns about what happens when direct control is surrendered and many people can freely contribute to information platforms. The scenario of broad participation in these platforms behind the firewall gives rise to a consistent set of worrying questions:

What if employees use their internal blogs to post hate speech or pornography, or to harass a coworker?

What if blogs are used to denigrate the company itself, air dirty laundry, or talk about how misguided its leadership and strategy are?

What if nasty arguments break out in a discussion forum and the whole thing descends into name calling and “flame wars”?

Won’t people be tempted to use forums to talk about current events, review movies, ask for advice about camcorder purchases, and have other non-work-related conversations?

What if people waste time filling up their employee profile pages with pictures of their kittens and vacations?

Will people just use social networking software to plan happy hour, rather than to get work done?

Don’t Enterprise 2.0 platforms yield just another source of discoverable content—material that must be turned over as part of a lawsuit or other legal action?

If the information on these platforms really is valuable, won’t it be harvested by spies and sold to the highest bidder?

Won’t hackers break into our Enterprise 2.0 platforms and steal their content?

Don’t these technologies make it easier to leak secrets, deliberately or inadvertently, to the outside world?

Don’t they make it too easy for confidential information to leap over our internal Chinese walls?

If we give up tight control over our intranet’s content, how can we possibly avoid running afoul of all potentially relevant regulations and laws on information sharing in all the places we do business?

The list of concerns grows when an organization also considers extending Enterprise 2.0 tools and approaches to external groups such as prospective customers, actual customers, suppliers, and other community members:

What if unhappy customers use our community site to air their grievances, and to talk loudly and often about our lousy products or Kafkaesque customer service?

What if a supplier uses our site to complain about how we never pay on time?

Are we responsible and liable if people give incorrect information or bad advice on question-and-answer forums we host on our Web site?

This is a daunting list, and many companies conclude that whatever the benefits of Enterprise 2.0, it’s not worth running the array of risks indicated by these questions. I’ve heard from many executives that their legal and human resources departments have advised against pursuing Enterprise 2.0. (To be fair, I’ve also heard that some companies’ human resources departments are big supporters of the new tools and approaches.) And for many decision makers, risks and nightmare scenarios seem concrete and immediate, whereas benefits appear more nebulous and distant. In such cases Enterprise 2.0 never gets off the ground; company policies become hostile to freeform information sharing and the technologies that support it, and incipient efforts are shut down.

Whenever I’m doing research or consulting at a company that’s been pursuing Enterprise 2.0 for a while, I ask about horror stories—the worst things that have happened related to use of the new technologies. I also often solicit such stories when I speak at conferences. And I keep my eyes open for catastrophes in the blogs, sites, magazines, and journals that I read. In short, I have made a serious effort to collect examples of counterproductive, dysfunctional, or risky consequences of Enterprise 2.0. Yet my collection is almost empty. I have yet to come across any true horror stories—scenarios that make me question whether the risks associated with deploying ESSPs actually do outweigh the benefits.

To understand why this is, let’s first place the concerns listed above into a few categories and then examine how the technologies, communities, and leaders of Enterprise 2.0 have been able to address them effectively. There are a few legitimate risks associated with deploying ESSPs, but far fewer than most people expect.

The first and largest category of risk is inappropriate behavior and content, either deliberate or inadvertent. As the list of questions above indicates, widespread concern exists that people will use ESSPs to post content that is offensive, alienating, needlessly provocative, irrelevant, or otherwise out of line with the goals of the organization.

All Internet surfers have at some point come across Web content that they found offensive or disturbing, and many online communities and discussions seem to be dominated by trolls, which Wikipedia defines as people who post “controversial and usually irrelevant or off-topic messages in an online community, such as an online discussion forum or chat room, with the intention of baiting other users into an emotional response or to generally disrupt normal on-topic discussion.” ¹ It seems quite possible that trolls and offensive content would also eventually appear on an organization’s internal ESSP. In other words, if a company tried to make its intranet multivoiced and egalitarian (as the Internet is), some of these voices would be unpleasant enough to drive others away, and perhaps even to land the company in trouble for offenses like creating an environment of harassment in the workplace. Four factors, however, combine to make this scenario unlikely.

First is the fact that while anonymity is the default on the Internet, on the intranet attribution is the norm. People can launch blogs, post comments, and edit wikis on the Internet without revealing their identities, but in most internal corporate ESSPs it’s possible to tell who made each contribution. Some companies have allowed unattributed employee comments on certain blogs, but this is as much anonymity as commonly exists on intranets. When attribution is the norm, people are much more likely to be cautious and circumspect, and much less likely to “flame” their colleagues. And if workers do misbehave on an ESSP, they can be easily identified, counseled, educated, and disciplined if necessary.

The second factor limiting inappropriate behavior is self-policing. Participants in an ESSP usually come to feel a sense of community and are therefore quick to react when they feel that a member is violating community norms. Counterproductive contributions are often met with a flurry of messages and posts that highlight why the content was out of bounds, reiterate norms, and offer corrections. The communities formed on top of ESSPs often have informal leaders who exert a great deal of influence and can shape the behavior of other members.

In addition to informal leaders, the formal leaders of an organization are a third counterbalance to inappropriate behavior and content. Managers can intervene when one of their direct reports is being counterproductive in an ESSP, and often the simple awareness that “the boss” is observing behavior and watching contributions leads to changed behavior.

I saw this firsthand recently on the wiki I set up for my MBA students. Contributions to this wiki were responsible for one-half of their grades, but I gave them very little upfront direction except to say that they should look at the wiki as an online textbook for the course that they would write collaboratively during the semester. My goal with this assignment was, of course, to have students experience emergence for themselves. This semester, as is often the case, students set up discussion boards within the wiki in order to set norms and policies and talk about what the textbook should look like.

My policy was to contribute to this ESSP as little and as rarely as possible; I told students that it was their resource, not mine. As I watched during the early part of the semester, though, I saw that some of the discussions were in danger of descending into a flame war, which the online dictionary webopedia.com defines as a series of “posts or messages in a thread that are considered derogatory in nature or are completely off-topic. Often these flames are posted for the sole purpose of offending or upsetting other users. The flame becomes a flame war when other users respond to the thread with their own flame message.” ² A few of my students were acting inappropriately and threatening to weaken the ESSP and its community.

I posted the following to the discussion:

We need to constantly watch to make sure that our online discussions with each other don’t ever become too hostile, dismissive, provocative, argumentative-for-argumentativeness’s-sake, etc. By this point in your MBA careers, you have an excellent template for interacting with each other—the in-class discussion. A good ground rule is that if you read aloud a comment you’re about to post, and find that it would sound “off” if said within an HBS classroom during a case discussion, don’t send it. Please keep in mind that any wiki contributions that would have the effect of discouraging others from reading or participating in an online discussion are contributions that work against the goal of the technology itself, and will be evaluated accordingly. Also keep in mind that it’s often hard online for others to tell if you’re kidding or joking or giving a real-world friend a hard time; because text-only comments have little context, they’re easy to misinterpret. This is not to say that online debate is bad; it’s actually necessary and healthy. It’s just to say that we all need to pay particular attention to the tone of our online discussions, and strive here at least as much as in the classroom to be courteous and respectful.

This comment showed that I was paying attention, and my students certainly understood that the word evaluated is, in this context, very close to graded. The flame wars stopped quickly. I’ve had to intervene like this only once in the three years I’ve included a wiki in my course.

The fourth, and most fundamental, factor limiting inappropriate behavior and content is simply that most people know how to behave appropriately in ESSPs and are inclined to do so. They have well-formed notions of how to interact productively with other community members, and they know what kinds of content and contribution are out of bounds. There are exceptions, of course, and in a few instances people have been fired because of ill-advised posts to their external blogs; but my experience indicates that most people know how to act professionally in job-related environments, including digital ones.

I discovered evidence to this effect in late 2007 when I was teaching a group of senior human resources executives from very large organizations about Enterprise 2.0. One of the participants told a revealing story. Her company, which employed a lot of young people, had become concerned about how the firm was being discussed and represented in these employees’ blogs, MySpace pages, Facebook profiles, and so on. So she and some colleagues decided to have a look at all these environments.

She soon learned that many of the young employees mentioned their company as part of their digital identity, but virtually always in appropriate ways. The worst offense she found, after a lot of looking around, was a photo of a training session in which account numbers were visible on a blackboard. It turned out that they were dummy account numbers and that the person who posted the photo, when made aware of the concern, immediately apologized and took it down.

If a company believes that these four factors do not provide enough protection against inappropriate content, it can set up a review or moderation process in which contributions must be vetted before they appear on a content platform. This precaution is more common on externally facing sites, where vandals and spammers can wreak havoc. Finally, inappropriate content on company-owned ESSPs, either internal-only or externally visible, can always be removed if necessary.

Another common concern is that people will contribute inaccurate information to ESSPs. In this scenario contributors are not deliberately doing anything inappropriate; they’re just misinformed. This concern is particularly acute for companies that host communities on their public Web sites in which participants can ask and answer questions. Wrong answers in such forums could at a minimum hurt the company’s reputation, and could also lead to further problems.

As with inappropriate behavior, self-policing is often an effective antidote to inaccurate information. Community members, it seems, take it upon themselves to highlight and correct inaccuracies. There are few topics about which correct information is more important than in tax law, so this might seem to be an area where individuals would be better served by hired experts rather than a broad and unscreened community. Yet the financial software company Intuit has added a community to its Web site for TurboTax, its popular tax preparation application. ³ The experiences of many other companies that run community sites—including Dell, Cisco, Amazon, and many others—are similar: community members themselves do a thorough job of policing inaccurate content, which can be easily corrected or removed.

A closely related concern is the appearance of embarrassing information on an ESSP. This problem often takes the form of negative reviews of a company’s products or services by a customer, or a more general expression of displeasure or distrust. Negative reviews on a community site sound like unambiguously bad news, but they can actually be beneficial in two ways. First, they help assure readers that the site has not been whitewashed of all negative content—that instead of containing only advertisements and cheerleading, it actually includes a more representative range of viewpoints. Negative reviews, in other words, increase confidence that positive ones can be trusted. ⁴ Second, negative reviews and comments call attention to problems, giving the company the opportunity both to become aware of them and to address them. In class discussions on this topic someone always makes a comment like, “Doesn’t a smart company want to know about negative customer experiences as soon as possible? Community sites sound like great early warning systems.” In addition, these sites allow companies to respond directly, quickly, and publicly to the person posting the embarrassing information. If the “service recovery” is an effective one, the community will see this. The cable television company Comcast has begun using Twitter both to scan for customer complaints about its service and to respond to them. ⁵

Embarrassing information is not limited to external community sites; platforms behind the firewall can also contain information that some would prefer to keep hidden. I once visited one of the world’s largest technology companies to talk with the people who were responsible for deploying ESSPs, including internal employee blogs. On the intranet page that listed the most recent blog posts I saw a title along the lines of, “Why Our Recently Announced Strategy Is Misguided.” A little surprised, I asked the team if this level of dissent was rare, and if the people who wrote such posts found themselves in hot water.

They assured me that the answer to both questions was no. For this company, visible employee disagreement with official strategy was not embarrassing; it was welcome. It helped surface potentially important issues and made sure they were discussed.

One of my MBA students saw something similar at his summer employer, disguised here as “Chemco.” In response to a class assignment question about the benefits of Enterprise 2.0, he wrote:

I have been an eyewitness of the power of enterprise 2.0 (although I didn’t know that is what it was called at the time) technology in promoting collaboration and connectedness across disparate individuals and groups within organizations. While interning at Chemco this summer, we were briefed on the Company’s brand-new intranet, which included blogging capabilities. The site was relatively new, and only a few people (mostly more senior leaders) had created blogs at that point. About a month into the internship, my supervisor (the head of corporate development and strategy) forwarded us a link to one of the comments to the CEO’s blog.

The comment had come from a low-level marketing manager located in a satellite office. In his remarks submitted to the CEO’s blog, the marketing manager openly questioned Chemco’s sacred cow—its ability to wring costs out of a process and to successfully operate an ultra-lean efficient organization. Specifically, he questioned the importance of one of the company’s favorite metrics (something they are extremely proud of); I’ll call it Metric A.

In his post (which was several pages, probably 800-plus words), he broke down basic financial information for Chemco and its 3 or 4 main competitors. Chemco was the clear leader in Metric A. He then overlaid this analysis with metrics such as market cap per employee and other metrics of value (I have forgotten what else he used), where Chemco was a distant laggard.

He went on to say that Chemco needed to essentially reverse its strategy and begin adding significant additional costs in the form of additional sales reps and R&D professionals, which are the key drivers of value in large chemical firms. His analysis was insightful, extremely thorough, and bold.

I was amazed at how much buzz it created in the organization—my project team decided to use much of his material as a source reference for one of our deliverables. The office of the chief executive formed a small (informal) task group to investigate some of the marketing manager’s claims and test their potential benefit.

In short, without this new E2.0 vehicle, this manager’s voice would likely never be heard. From the look of his robust analysis, this was something he wanted to (and perhaps had tried to) share for a long time, and he now had the means of doing so. As the above example illustrates, I think that the potential benefits of E2.0 to an organization can be revolutionary—a step-function improvement in collaboration and efficient dissemination of information and resources.

Many decision makers are concerned that ESSPs raise the risk of noncompliance with laws, regulations, or policies. These platforms have many contributors, wide borders, and little, if any, up-front verification of content. Furthermore, all contributions are instantly and widely visible, which seemingly increases potential harm, exposure, and liability. If employees deliberately or inadvertently post information that violates the laws or regulations in any environment where the company does business, there could be serious consequences.

I first thought of these risks as I was conducting interviews for the first Harvard Business School teaching cases I wrote on Enterprise 2.0. These cases were set at the European investment bank Dresdner Kleinwort Wasserstein (DrKW), and focused on the use of blogs and wikis. DrKW CIO JP Rangaswami had led the effort to deploy these technologies throughout the bank at a single point in time, to make them available to most employees, and to refrain from issuing detailed guidelines or policy statements about their appropriate uses. ⁶

New York Attorney General Eliot Spitzer had recently made headlines for his successful prosecution of multiple U.S. investment banks, showing that some of their analysts had expressed pessimism about certain stocks in private to their colleagues while remaining optimistic about the same companies in comments to the public. There had also been high-profile cases about bank employees sending sensitive information over the Chinese walls set up to ensure that traders cannot gain access to nonpublic data about deals in progress. ⁷

As I was interviewing Rangaswami, it occurred to me that ESSPs were hugely risky and inappropriate technologies for investment banks, especially as they were being implemented at DrKW. These tools spanned the firm’s Chinese walls and provided forums for employees to voice their opinions behind the firewall; if these opinions were at odds with public statements, the bank could face serious penalties. And because DrKW did business in many countries, it was subject to a large and constantly shifting set of laws and regulations that could be difficult to monitor. I asked Rangaswami how he could possibly justify deploying ESSPs in such an environment, especially without issuing clear and detailed policies or conducting extensive training.

I found his response illuminating:

These technologies are our best defense against financial industry regulators and prosecutors around the world. These parties don’t typically come after you for isolated incidents or honest mistakes—they come after you for long-term patterns of misbehavior or illegal activity. And because our blogs and wikis are open to almost everyone, that means that everyone in the bank can be on the lookout for these patterns. As soon as anyone sees anything troubling on one of these platforms they can alert our compliance department, which can then take action. If there is a problem, we can show any regulator exactly what happened and when, when we became aware of it, the corrective actions we took, and whether or not the inappropriate activity stopped.

Now, contrast that with e-mail and instant messaging, our two most popular collaboration technologies at present. We keep a record of all of this traffic over time and we have the right to read it, but we can’t possibly monitor it comprehensively. And because e-mails and IMs are invisible to everyone except the senders and receivers, the bank’s workforce can’t look at them and help us spot inappropriate activity. Keep in mind that Spitzer used e-mail records to show that the analysts he prosecuted were saying one thing in public and something very different in private. When I think about compliance issues email makes me very nervous, and blogs and wikis calm me down. ⁸

I found Rangaswami’s answer very insightful, and his argument very compelling. He helped me understand that ESSPs actually reduce noncompliance risks, because their content is so widely visible. Channel technologies like e-mail, on the other hand, are risky because they are so private.

Rangaswami made another interesting point during the interview. He stressed that DrKW’s workforce had been amply educated about inappropriate behaviors and communications in the workplace prior to the deployment of ESSPs, and so could be trusted to use them properly. Employees who engaged in harassment or violated Chinese walls, he said, would already have been counseled or released, and it made no sense to him to think that people were just waiting for Enterprise 2.0 technologies to appear so that they could start misbehaving.

Some of the questions listed earlier in this chapter reflect concerns about theft, either by insiders who could copy content from ESSPs or outsiders who might hack into the platforms to steal content. Theft is always a possibility with digital information (indeed, with any information an organization possesses), but the platforms of Enterprise 2.0 seem to provide particularly rich targets because they aggregate many types of content. In other words, if it’s true that ESSPs centralize and distill previously disparate information and let the cream rise to the top, it should also be true that they contain content that would be highly valued outside the organization.

When assessing the magnitude of risks like theft and comparing them with the benefits offered by Enterprise 2.0, I find it particularly valuable to look at organizations where the risk is greatest. These organizations provide critical cases, and if they’ve decided that the benefits outweigh the risks, I conclude that the same is probably true for most other organizations as well.

I have trouble imagining that information theft can be more harmful anywhere than in a national intelligence agency. Yet the U.S. Directorate of National Intelligence (DNI) has supported Intellipedia and the other Enterprise 2.0 technologies that have been deployed across the sixteen federal agencies making up the nation’s intelligence community (IC). Despite the relatively recent counterintelligence examples of Aldrich Ames at the CIA and Robert Hanssen at the FBI, both of whom were convicted of selling highly sensitive secrets to other countries, the DNI has concluded that the benefits of better and wider sharing of intelligence information outweigh the risks.

It’s particularly interesting to note that the IC has not developed all of its ESSPs internally in order to ensure that they are secure. Instead, it has deployed a number of publicly available applications, including a blogging engine from WordPress, the MediaWiki software that underlies Wikipedia, and a Google appliance for enterprise search. The first two of these are open source applications; the third is a commercial product. This evidence indicates that publicly available technologies can be secure enough even for highly sensitive environments. The critical case of the IC therefore indicates to me that the benefits of Enterprise 2.0 outweigh the theft risks for the great majority of organizations.

Whatever the advantages of Enterprise 2.0, though, it is true that ESSPs typically do increase the amount of discoverable information within an organization. In other words, they add to the inventory of material that can be requested and reviewed as part of a legal action, just as e-mails and memos can. Many organizations understandably want to limit the amount of discoverable information they produce, but they also want to gain access to the advantages and capabilities of Enterprise 2.0 as described in this book. Because I have yet to hear of a case in which ESSPs and their content gravely hurt a company during a legal proceeding, I continue to believe that the benefits of Enterprise 2.0 outweigh the potential disadvantages associated with generating more discoverable content. For most organizations, in fact, I believe that these benefits outweigh all the risks described above.

Long Hauls

The questions discussed earlier in the chapter came from managers in organizations that hadn’t yet begun to deploy ESSPs. Once a company has made the decision to pursue Enterprise 2.0, the people in charge of the deployment usually ask a very different question: how can we persuade more of our people to use the new technologies? Many internal champions are frustrated by what they see as the slow pace of adoption within their organizations. In a common scenario a company, division, or work group installs one or more ESSPs, alerts its employees about them, conducts training, and then sits back to reap the benefits described in earlier chapters.

I thought I’d observe just this scenario unfolding at one of the first companies where I did research on Enterprise 2.0, a large financial services firm. One of the effort’s champions was a line manager who wanted to take advantage of the benefits of group editing and self-organization. He became aware of the firm’s wiki deployment and encouraged the people who worked for him to use it for their group-level work. However, it proved difficult to persuade his staff to use the new tool.

When I interviewed this manager, I expressed surprise that migration to the wiki was not faster and more spontaneous. I mentioned the extraordinary growth of Wikipedia and wondered aloud why his company hadn’t experienced the same thing with its wiki. He explained that the comparison was invalid. Wikipedia, he said, can potentially draw on all Internet users as contributors. Even if only a tiny percentage of these potential contributors ever become actual contributors, this group will still be large enough to generate and refine Wikipedia’s content.

A quick analysis showed that he was exactly right. In November 2005 (shortly after I did the interview) Wikipedia contained over 850,000 articles in English and 2.9 million across all languages. This content was generated by fewer than 50,000 contributors in English, 103,000 altogether. A contributor is defined by Wikipedia as someone with a user ID who’s made at least ten total edits. Even this population is skewed: active English-language wikipedians (more than five contributions in a month) numbered 15,600 in that month, while very active (100 or more) numbered only 2,081.

The community of Wikipedia contributors is fairly large in absolute terms, but it’s a tiny, almost negligible proportion of all English-speaking Internet users. When scaling down from the Internet to an intranet—in other words, when considering the use of an ESSP within an organization—if anything like the same percentage of users starts contributing, the platforms will remain almost totally empty, and the effort will be an abject failure.

It’s easy to be impressed by the large, dynamic, and vibrant Web 2.0 communities on the Internet and to overlook the fact that they’re actually quite tiny when expressed as a percentage of all Internet users. A key challenge, then, for all Enterprise 2.0 advocates is increasing the percentage of intranet users who contribute to their organizations’ ESSPs.

To do this, it’s critical to understand why the “ambient percentage” of contributors to organizational ESSPs isn’t higher. Are the technologies themselves too primitive, or are they difficult to learn and use? Do some managers in an organization actually act to block Enterprise 2.0, because they don’t want information to flow more freely? Or are the real roadblocks internal, rooted somewhere in the heads of individuals? Accurate answers to these questions are essential prerequisites to designing and executing successful efforts to deploy ESSPs.

In June 2008, at the Enterprise 2.0 conference, I moderated a panel of early adopters, including representatives from Wachovia Bank, Pfizer, Sony, and the IC. The first question I asked them, and the one on which we spent most of our time, was essentially, “If Enterprise 2.0 tools and approaches really are so beneficial and powerful, why haven’t they spread like wildfire?” I suggested three categories of impediment: managers, technologies, and users, and invited the panelists to hold forth.

In their initial responses all of them identified users, not bad managers or inadequate technologies, as the biggest barriers to faster and deeper adoption of Enterprise 2.0. Entrenched practices and mind-sets and some degree of technophobia combine, they said, to limit the pace of adoption. These factors slow the migration from channels to platforms and necessitate continued patience, evangelism, and training.

I hadn’t expected the panelists to say that the Enterprise 2.0 technologies themselves were so incomplete as to hinder adoption, but I was a bit surprised that in their first round of comments, none of them identified management as a real impediment. So I pressed the point by saying something like, “I didn’t hear any of you point the finger at the managers in your organizations. Were you just being polite, or are they really not getting in the way of Enterprise 2.0? The new social software platforms are a bureaucrat’s worst nightmare because they remove his ability to filter information, or control its flow. I’d expect, then, that each of you would have some examples of managers overtly or covertly trying to stop the spread and use of these tools. Are you telling me this hasn’t happened?”

That is in fact what they were telling me, and I didn’t get the impression that they were just being diplomatic. They said that managers were just another category of user that needed to migrate over to new ways of working, no more or less. In other words, the panelists hadn’t observed managers in their organizations actively trying to impede Enterprise 2.0.

This surprised me. I’d assumed that since Enterprise 2.0 tools and approaches have no inherent respect for existing organizational hierarchies and boundaries, those who had ascended through the hierarchy and within the boundaries might be actively hostile to them. For the organizations represented on the panel, at least, that did not appear to be the case. The most counterproductive behaviors mentioned were the reflexive desire to work in private and the temptation to build a large number of mutually inaccessible ESSPs.

The panelists represented large organizations, most of which had been around for a long time and had large numbers of managers who were used to, and probably comfortable with, the status quo. Yet no examples surfaced of these managers trying to thwart or sabotage Enterprise 2.0 efforts, and no panelist told a story about managers darkly hinting to their groups that participating in these platforms might not be the best thing for a career. They talked instead about how hard it was to get people to change the way they worked, and to change the interaction and collaboration technologies they used.

Individual knowledge workers’ failure to see the benefits of Enterprise 2.0 and adopt ESSPs can seem puzzling at first. After all, the new tools are both useful and easy to use. They require no advanced computer skills, offer novel capabilities, and benefit both organizations and people. They enable individuals to work more effectively with their strongly tied colleagues; build, maintain, and exploit a network of weak ties; and convert potential ties into actual ones. All of these activities also enhance the enterprise(s) where these individuals work. As Google’s example shows, at least some people also derive value from trading in a prediction market, so the combination of individual- and group-level benefit persists even at the outermost ring of the Enterprise 2.0 bull’s-eye.

It also seems as if enterprises should have little trouble increasing their “ambient percentage” of ESSP contributors to a level higher than that which exists on the Web. After all, organizations have a wide range of levers available to encourage desired behaviors. These levers include norms, incentives, objectives, policies, and whatever is meant by the important but vague notion of “corporate culture.”

Organizations also usually have internal Enterprise 2.0 change agents and champions who are eager to both demonstrate the new ESSPs to their colleagues and to explain the advantages of using them. In many cases these evangelists will be younger employees and new entrants to the workforce, since members of Generation Y are much more likely than older workers to be comfortable with ESSPs—in fact, this group is sometimes called the “Facebook generation.”

For all these reasons it seems that there should be relatively little persistent individual resistance to Enterprise 2.0, and that companies should quickly be able to turn their employees into competent and enthusiastic users of ESSPs. Yet this is not at all the situation that the panelists at the 2008 Enterprise 2.0 conference reported, and their experience appears to be typical. Many organizations, especially larger ones, have found that ESSPs remain a niche technology even well after their introduction, used by only a relatively small portion of the workforce, and lagging far behind the universal deployment of older channel technologies like e-mail. According to a global survey conducted by McKinsey, published in the McKinsey Quarterly in July 2008, only 21 percent of respondents expressed overall satisfaction with Web 2.0 tools, 22 percent expressed clear dissatisfaction, and 7 percent had tried at least one ESSP but had subsequently stopped using it. ⁹

Of course many factors contribute to the comparatively slow spread of ESSPs among knowledge workers within enterprises. The explanation I have found most comprehensive and compelling comes not from research on information technology, but rather from investigations of why some consumer products fail to gain mass acceptance even though they are clearly “better mousetraps.”

In a June 2006 Harvard Business Review article titled “Eager Sellers and Stony Buyers,” Harvard Business School marketing professor John Gourville traced that phrase to a quote from Ralph Waldo Emerson: “If a man can write a better book, preach a better sermon, or make a better mousetrap than his neighbor, though he build his house in the woods, the world will make a beaten path to his door.” Gourville concluded that this quote was compelling, but wrong; in fact, there are plenty of examples of products and services that offered clear advantages over those already in place—better mousetraps—yet never succeeded in replacing them. ¹⁰

The stand-alone digital video recorder TiVo is one of the examples Gourville cites. TiVo drew rave reviews as soon as it was introduced in 1999, but the company had trouble generating sales sufficient to cover its heavy development and marketing expenses. By 2005 it had accumulated $600 million in operating losses.

TiVo’s failure to gain mass acceptance quickly is especially puzzling, because it seems to be a clearly better mousetrap. When I speak to groups, I frequently ask TiVo owners to raise their hands. Fewer than one-third of the audience typically do so. I then ask these people to keep their hands in the air if they love their TiVo, if they can’t imagine going back to the old, pre-TiVo way of watching television, and if they are essentially unpaid members of the TiVo sales force because they are always telling their friends how great the device is and encouraging them to purchase one. Most hands remain in the air throughout this sequence. I then ask in mock exasperation, “Why don’t we all have a TiVo?”

Gourville drew on research in behavioral economics to answer this question. Among its many other accomplishments, this research has identified three consistent features of people’s psychological “equipment” for making evaluations.

•     We make relative evaluations, not absolute ones. When I’m at a poker table deciding whether to call a bet, I don’t think of what my total net worth will be if I win the hand vs. if I lose it. Instead, I think in relative terms—whether I’ll be “up” or “down.”

•     Our reference point is the status quo. My poker table comparisons are made with respect to where I am at that point in time. “If I win this hand I’ll be up $40; if I lose it I’ll be down $10 compared with my current bankroll.” It’s only at the end of the night that my horizon broadens enough to see if I’m up or down for the whole game.

•     We are loss-averse. A $50 loss looms larger than a $50 gain. Loss aversion is virtually universal across people and contexts and is not much affected by how much wealth one already has. Ample research has demonstrated that people find that a prospective loss of $x is about two to three times more painful than a prospective gain of $x is pleasurable.

When combined, these three elements lead to what the behavioral economist Richard Thaler has called the “endowment effect”: we value items in our possession more highly than prospective items that could be in our possession, especially if the prospective item is a proposed substitute. ¹¹ “We mentally compare having the prospective item to giving up what we already have (our “endowment”), but because we’re loss-averse, giving up what we already have (our reference point) looms larger. A related phenomenon is the “status quo” bias—our demonstrated preference for keeping what we already have even when equal or superior alternatives are available.

And Gourville points out three factors that make the situation worse for product developers who want their offerings to succeed. First is timing: adopters have to give up their endowment immediately, receiving benefits only sometime in the future. Second, these benefits are not certain; the new product might not work as promised. Third, benefits are usually qualitative, making them difficult to enumerate and compare.

As if all this weren’t enough, Gourville also highlights the fact that the people who develop new products are generally quite dissimilar to the products’ prospective consumers. You don’t become a TiVo executive or engineer if you don’t “get” the potential of digital video recorders and think they’re a great idea. And after working for the company for a while, having TiVo becomes part of your endowment; you think of other new products in comparison to TiVo, instead of in comparison to a VCR or DVD. Both these factors make it harder for technology enthusiasts to see things as their target customers do.

All of these elements lead to what Gourville labels the “9X Effect”—a disparity of 9 to 1 between what innovators think consumers want and what consumers actually want. This effect goes a long way toward explaining the high-tech industry folk wisdom that to spread like wildfire, a new product has to offer a tenfold improvement over what’s currently out there. For products that aren’t inherently ten times better, Gourville offers important cautionary words:

Many products fail because of a universal, but largely ignored, psychological bias: People irrationally overvalue benefits they currently possess relative to those they don’t. The bias leads consumers to value the advantages of products they own more than the benefits of new ones. It also leads executives to value the benefits of innovations they’ve developed over the advantages of incumbent products.

That leads to a clash in perspectives: Executives, who irrationally overvalue their innovations, must predict the buying behavior of consumers, who irrationally overvalue existing alternatives.

The results are often disastrous: Consumers reject new products that would make them better off, while executives are at a loss to anticipate failure. This double-edged bias is the curse of innovation. ¹²

This curse, I believe, directly applies to enterprises as they attempt to deploy ESSPs and promote Enterprise 2.0. This is because acceptance of a new piece of technology among people who are free to use it or not is very much like consumer acceptance of any other new product. Therefore, the endowment effect and status quo biases apply.

E-mail is virtually everyone’s current endowment of collaboration software. Gourville’s research suggests that the average e-mail user will underweight the relative benefits of a replacement technology like an ESSP by about a factor of three, while Enterprise 2.0 enthusiasts will overweight these same benefits by the same factor (since for them ESSPs are the status quo). This is the 9X effect deployers of new collaboration technologies will have to overcome.

In most companies groupware and knowledge management (KM) systems didn’t solve the problem, because they weren’t that much better than e-mail. The current generation of Enterprise 2.0 tools is clearly different from groupware and KM and offers more and greater benefits. But that’s not really the critical consideration. The critical consideration is brutally simple: are these tools nine times better than e-mail for collaboration?

Consider how high this sets the bar. E-mail is freeform, multimedia (especially with attachments), WYSIWYG, easy to learn and use, platform-independent, social, and friendly to mouse-clickers and keyboard-shortcutters alike. It would actually be a pretty tough competitor even if it weren’t the universally used incumbent, and so the beneficiary of the 9X effect.

Within enterprises, where the incumbent collaboration technology of e-mail is well established, ESSPs are what Gourville calls “long hauls”—products that represent significant technological leaps forward and are therefore potentially quite valuable, but require major behavioral changes from their target audience. Long hauls have the potential to become popular and widespread, but their success comes slowly. Champions of long-haul products must be patient and prepared to evangelize, demonstrate, coach, train, and explain for what seems to them a very long time. As Gourville writes, “The simplest strategy for dealing with consumer resistance is to brace for slow adoption … to be successful companies must anticipate a long, drawn-out adoption process and manage it accordingly.” ¹³