Chapter 3. Research with Expert Resources

Brands are using the Internet to find the best PR research strategies for market and media intelligence, with the right partners to assist them. When budgets are tight, organizations keep research strategies in-house and use Internet searches for secondary (previously published) research. There are also "free to inexpensively priced" tools to construct polls and questionnaires in-house for primary (first-hand) research. Then, when funding is available, more formal research services can be contracted. Either way, you can rely on the Internet. It has the reach and the expert resources needed for your research program—whether you conduct the research yourself or you hire the pros to handle a program for you.

The Internet Meets Your Budgetary Requirements

Luckily, even when you're on a restricted budget, you can work with PR service providers who make it extremely easy and cost effective to use 2.0 research tools. Today, smaller companies are able to invest in a research program, sooner rather than later. Ted Skinner, Vice President of Public Relations Products at PR Newswire (www.prnewswire.com), explained that PR Newswire has spent years researching and probing PR professionals. Yes, even the research providers do their own research, which leads to more affordable products for brands. One such product is MediaSense, which offers a brand the ability to track and evaluate messaging and market perception, and also enables you to have more control over brand communication. The notion of having tighter control over communication in a PR 2.0 environment is appealing. The service works with a linguistic model that is able to analyze content electronically through artificial intelligence, which enables it to know the different meanings of the same word. As an example, the computer can distinguish from the words surrounding a keyword the difference between the bat of an eye, a baseball bat, or the animal. The computer reads information like a human being can and, through a natural language process, is able to decipher a wide range of words and terms; the computer can even determine the tonality of a story, whether it's positive, negative, or neutral.

You also need a research program that gives you the ability to research and analyze market intelligence. For example, it's in your brand's best interest to identify, track, and evaluate its top five competitors. You need to know how these companies are performing in the marketplace and whether your brand surpasses their performance or pales in comparison. According to PR Newswire's Skinner, MediaSense is used by many different organizations, from nonprofits, universities, and lobbying groups to high-end multinational corporations. One of Skinner's favorite success stories is about a PR agency that asked PR Newswire to assist them in a new business effort. The agency needed to get its hands on information, approximately 12 months worth of reporting on a tough challenger they competed against for a new business opportunity. Their intention was to use this high-quality report as a part of their presentation to win the new account. The information, available from MediaSense, enabled the PR agency to undermine the incumbent's position by obtaining a detailed report from a neutral third party on that agency's performance over the past year. The PR agency used this valuable intelligence to uncover its competitor's weaknesses, and won the new business account. Although MediaSense is an annual investment of $10,000, it's a considerably small investment if you break it down over the course of the year. Skinner goes into more detail on monitoring brand communication programs (especially how PR 2.0 needs to be monitored through blogs) in Chapter 5, "Better Monitoring for PR 2.0."

Gurus Need Not Apply Here

There are also inexpensive software programs for license or purchase to aid you in your media intelligence. You learn more about media intelligence and the importance of building relationships with the media in Chapter 4, "Reaching the Wired Media for Better Coverage." Among the interesting examples of less expensive and easy-to-use resources (as compared to the well-known service providers, such as Cision or Vocus) are CornerBarPR and Bulldog Reporters Media List Builder. CornerBarPR has been in business for about five years. Don't let the name fool you. They have a serious online tool that is simple and boasts approximately 60,000, and growing, media contacts for PR pros to generate their targeted media lists. CornerBarPR does not claim to provide the advanced features of the more widely recognized media databases. According to Richard Barger, who goes by the title of Chief Curmudgeon, "Our service is a cost effective resource for the entrepreneur or small start-up that needs to develop that targeted media list." CornerBarPR is not a fancy solution and the Web site isn't flash enabled, but the free demo alone is a good indication of how easy it is to use this media-generating tool.

Bulldog Reporter's Media List Builder is another quick and efficient tool that enables you to create highly accurate and targeted media lists. Because the program is online, you're able to develop your list in minutes without any training or complicated instructions. The service also provides you with exclusive pitching tips with each list you generate. Short blurbs offer you advice on different media contacts and the types of material they prefer to receive. Other tips from Media List Builder include when to follow up with an editor (the best time to call or e-mail), pitching pet peeves, quirks and editorial topics, and "hot buttons." Media List Builder has no membership or licensing fees. It's a pay-per-use service that's based on the number of editorial contacts you generate. According to Mike Billings, research manager at Bulldog Reporter, "The service is designed for small practitioners. You pay only $2.00 per editorial contact. Bulldog is currently working on a subscription service. This program is PR 2.0 friendly and will even include a wiki feature for PR professionals." Another excellent feature of Media List Builder is that there's no charge for building the list. You can review your list first and then decide if you want to purchase, receive, and archive it for future use.

If you are in a position to use media-gathering intelligence tools, you'll quickly experience how easy they are to use. Most of these tools don't require much more than a brief review of media outlet criteria and then the selection of the types of media outlets/editors you need to reach. Everything is accomplished online. And, you definitely don't need to be a 2.0 guru to manage the media list building process. A long checklist asks you to narrow down your criteria for the most effective search. You can search by interest/category, editor's title, circulation, regional or national outlet, type of information accepted, and if it's a paid versus unpaid subscription. The same types of criteria exist for online publications. In the case of a broadcast search, the criterion is, of course, different. For broadcast, you're looking for a specific program format, whether or not the station accepts interviews, and the program's reach. As long as the person or group in your organization has access to the Internet, you're ready to explore the research tools for today's communications professionals.

It's important to conduct continuous research for your brand. I believe you need to do research at intervals. Your audiences' habits and behaviors can change in an instance or eventually change over time. Primary online research—including polls, surveys, and focus panels—can be quarterly efforts for some companies and for others, monthly. Of course, that ties directly to your budget. Some companies prefer to do more research throughout the year in-house, and then use outside resources for one large study to complement their efforts.

Find the Right Research Partner

The Pharmaceutical Safety Institute (The Institute) is an independent, service-providing organization formed to restore and maintain consumer confidence in medicines worldwide. The Institute kicked off its research program with a prominent third-party research provider, Harris Interactive^™. Harris Interactive (www.harrisinteractive.com) is a partner that provides not only the technology for online research, but also guides you through the process every step of the way.

Harris Interactive is the 12th largest and fastest-growing market research firm in the world. The company provides innovative research, insights, and strategic advice to help its clients make more confident decisions, which lead to measurable and enduring improvements in performance. Harris Interactive is widely known for The Harris Poll, one of the longest running, independent opinion polls, and for pioneering online market research methods. The company has built one of the world's largest panels of survey respondents, the Harris Poll Online^SM.

Although The Institute gathers a great deal of in-house research, it still needed to use an outside source to validate and add credibility to its research efforts, in the eyes of the media and the public. The Institute researched several different research providers, but really wanted a partner that would help them to build trustworthiness in the market among the Life Sciences sector, healthcare professionals, and consumers who were all concerned with drug recalls and what's been labeled a "crisis" of safety in the U.S. and worldwide. By commissioning Harris Interactive to complete an online research study on Consumer Perceptions of Drug Safety, the Institute was able to gain the recognition and credibility it desired, with the Harris Interactive name behind some very interesting survey research findings.

It wasn't a difficult decision for Dr. Axel Olsen, President of The Institute, to invest in the online research program with Harris Interactive. In fact, Dr. Olsen knew an online survey would be an effective tool to reach the 21st Century consumer. Prior to selecting Harris Interactive to conduct the survey, The Institute screened many other research organizations. Most research providers were in the $10,000 to $20,000 range and offered to provide a research program with several research components: the development of an online quantitative survey instrument (15 to 20 minutes in length), a panel of respondents representative of the desired targeted population, 1,000 online completed responses, a tabular report and an SPSS data file, and a summary of key findings.

Although there were less expensive options, The Institute chose Harris Interactive as a partner because they wanted an experienced provider, one whose name they could leverage in marketing efforts. (You learn more about Harris Interactive in an interview with Humphrey Taylor, the Chairman of The Harris Poll at Harris Interactive, and two key methodologists for the company, John Bremer and Randall Thomas, at the end of this chapter). The choice to use Harris Interactive was also based on the fact that the organization is known to be a leader in conducting online market research. As the survey contract negotiations went back and forth between the two parties, Harris Interactive provided concrete evidence on successful web-based research that really helped Dr. Olsen and his team feel comfortable about the effectiveness of online surveys. One of the first questions to arise was whether the media would see an online survey as a valid instrument to gather information on consumer opinions. Would some of the medical journals and healthcare publications they were trying to reach accept online research, or did they include only telephone survey research in their publications? The Institute wanted to be certain that the Journal of the American Medical Association (JAMA), which is a well-respected medical journal, would accept online survey research. A quick call to JAMA informed Dr. Olsen that even the hardcore research publications accept online statistics as valid research, as long as the sample size was large enough. Harris Interactive proved that online research was an effective means to collect data on consumer drug safety perceptions.

What the Experts Have to Say

According to one article in American Psychologist, there are several preconceptions about Internet data.^[1] Some of these include

1. "Internet samples are not demographically diverse."

2. "Internet data are compromised by the anonymity of participants."

3. "Internet-based findings differ from those obtained by other methods."

An informal interview with the Harris Interactive executives helps to clear up these three misconceptions and really provides an in-depth look at why you should incorporate online research into your program. The professionals who participated in the interview were:

• Humphrey Taylor, Chairman of The Harris Poll, Harris Interactive
• John Bremer, VP, Global Representative, Harris Interactive
• Randall K. Thomas, Director of Internet Research and Senior Research Scientist, Harris Interactive

Q: What are the biggest differences among research methods today, as compared to the tools we used years ago?

Humphrey: I think the biggest difference, obviously, is that 10 to 15 years ago there was virtually no online research. Now, out of the worldwide spending on market research, a big and rapidly growing chunk of it is in fact done online. Telephone research, which was a substantial part at one time, has not only been on the decline, but has, in some ways, been getting worse because of the weakening penetration of traditional land lines. Many countries now have more people online, and there are fewer landlines because of cell phones. Also there is the increasing difficulty of getting people to actually talk to interviewers on the telephone. That's one big difference.

John: Yes, I think even though the research industry has always been at the forefront of using fantastic analytical tools, I also think there's a convergence of academic and some heavy hitters from the statistical, economic, and the psychological worlds. The tools we are using now are even more advanced than they were 10 to 15 years ago.

Randall: The research tools have changed fairly dramatically. A couple things we can see, going along with that visual theme, is that we've developed many tools which enable us to measure reactions using visual presentations, and some of those are ad concept testing and package testing. There's also discreet choice modeling. You can see many elements simultaneously that would be hard to do in any other modality. People are presented with more realistic choices using multiple variables and they make decisions very much resembling those that occur in real life. From that, we can find out how important various features are. So, we can get more complex information from the new types of tools we're developing. I work on new research tool development, so part of my job is to review tools as we develop them. We're developing new kinds of scales. When we think about visual analog scales, we can create scales of any visual form that we want to map out the human mind, and as long as people see them as meaningful, we can more accurately measure ideas involving how people feel, think, and what they're going to do. In other modes of research, you often have to simplify the scales, present them in very practiced ways, and as a result, might not be able to explore novel ways of thinking or reacting. For instance, for car interiors, we can actually show respondents a different lighting format within a car interior and ask them to pick the lighting formats they like. On the Internet, they can actually click on different buttons and see an overhead light come on within the car interior. Or you can see a side light come on, and you can ask, well what kind of combinations of lights or what types of lights do you like in the car interior? This can't be done very easily in other modalities. It's also very interactive, it goes with Web 2.0, so many tools we're developing are going after the interactivity of the Web, portraying events in a much more complex fashion, resembling real life.

Humphrey: Additionally, not only has the Internet enabled us to do many more interactive interviews, they are much faster, across many more countries, and in many more languages than previously.

Q: Do you think that the combination of the traditional and the online research provides really solid information and data?

John: I think it depends. There are a lot of things going on when you consider the research methodology. Can you reach the population or a proportion of the population you want to online? Is there some form of a social desirability bias? Humphrey and Randall wrote a paper on social desirability bias which talked about if you are doing studies online, you can sometimes get more truthful answers. So there are times when doing the research online by itself gives you better answers—more reliable answers.

Humphrey: Let me give you an example. When we ask people at the end of a telephone survey about their sexual orientation, we get up to two percent who identify themselves as gay, lesbian, or bisexual. When we ask the same question at the end of an online survey the number is at or close to six percent. When we ask about going to church, the numbers are different. When we ask about believing in God, the numbers are different. When we ask about drinking alcohol, the numbers are different. So, anything where there is embarrassment to telling a human being you drink or don't go to church or don't give to charity or whatever it is that makes you feel uncomfortable, then the numbers, we think, are much better with the online surveys.

John: There are times when doing mixed mode studies, in certain countries, is really to a brand's advantage. One of the things you'll hear a lot about is convenience. Convenience is an important factor in survey research. When somebody doesn't want to talk to a telephone interviewer, they can for the most part avoid him or her. You have the option of saying, "Look, I don't have to talk to you now." You can take a survey online, or you can take a survey via telephone. For physicians, we often give them the option of taking it via paper, via Internet, via telephone, anything that is convenient for them and enables them to take the survey and give us the most thoughtful answers they can. That's really the result we want.

Humphrey: That's particularly true, as John implied, in terms of professionals and business people, some of whom will do a survey online and not on the telephone and vice versa.

John: So, that's our mixed mode methodology. When you use a mixed mode approach, it helps to reach certain populations, particularly internationally. In certain countries, the Internet penetration isn't high enough to enable us to do a full online approach, but it is high enough for us to get some of our interviews online and the rest through more traditional modes such as face-to-face or telephone methodology.

Q: Just stepping back for one moment and thinking about your populations, are you finding that certain demographics are more apt to participate online?

John: I'm a statistician, so understand I'll always say it depends on a lot of things. It's not as though you can say men prefer to do surveys online and women prefer to do surveys on the telephone. There are certain types of people who have definite preferences and it gets back to what I was talking about before, which is convenience. We tend to see that professionals will take surveys online more than they'll take surveys on the telephone. For certain income groups, they might choose to take surveys online more. I haven't really found that the gender has played any role. I have found that the younger age groups do tend to be more comfortable online, so they do have a preference, making it difficult to get them in many survey modes at this point. Even though I have an inclination that they have an easier time, or a better time taking it online, they're still not the easiest to reach.

Humphrey: I think that in America, more college students are online than have landlines and can be reached by a telephone survey.

John: That's absolutely true. We know that, even in the 18- to 24-year-old age group, 24 percent have given up their landline at this point and traditional telephone interviews do not call cell phones just because there's an additional cost, and there's an inconvenience factor to the person taking the survey. So we're getting less data from telephone research.

Randall: We're able to present a lot more visual stimuli, whether its video or visual images for people to view. And, the thinking and answers we get are more visually based rather than orally based, as might be obtained in telephone or face-to-face interviews. It's a different way of thinking about research. It requires more interaction with the visual environment as well.

Humphrey: One important difference is that we can test television commercials or movie trailers online in people's homes, which we couldn't do before. We had to bring people to theaters to have them look at trailers or commercials.

Q: What do you feel are the benefits of 2.0 research?

John: I'm one of the biggest cheerleaders for online research. I think there are a lot of benefits out there. There are several benefits such as what Randall was just talking about, which is the ability to get more information. Again, it has to be appropriately analyzed. The fundamentals of market research and survey research haven't changed. Even though we have great technology and better ways of doing things, the fundamentals are still there. But you can collect more data from more people, so you have the ability to take that and get bigger sample sizes with more in-depth analysis of your population of interest. We've talked a lot about doing things better. There are analytic techniques that were out there but we weren't able to use them quite as easily as we use them now. We have the ability to get a very particular population of interest.

Humphrey: A specific example is the number of people who say they have been diagnosed with depression. It's much higher online than it is on the telephone. And, we can go back to the same people much more easily, quickly, and less expensively than you could if you did it by telephone.

Q: Do you feel that PR professionals have been skeptical of online research, or as a group have they really embraced new research methods?

Randall: From what we see, PR firms are embracing online because the results we obtain from online research are quite comparable to those obtained from phone research.

John: To address the skepticism issue, the one point I do want to make is that we're using the phrase "online research" but not all online research is created equal. One of the things that I'm very proud of is the work that we've done here at Harris. There are certainly some other really good firms that have taken the time to look into what produces a representative result online. Part of the problem is that there are some firms that are very quick and dirty. They provide a sample and you're not entirely sure where the sample is from. As a result, there are times when people do get burned.

Online research, when it's done well, is a fantastic tool. Online research when it's not done well can leave people with distaste for doing Internet based research. I think that many PR firms are savvy enough to understand now that you need a combination of looking at survey designs, sampling, and weighting together to figure out what really does constitute a good survey. And that's not only true for online; that's true for telephone and for face-to-face. The fundamentals of survey research have not changed. At one point people thought the fundamentals had changed. They really haven't. So I think there was some skepticism a while back when people were getting burned with some data that really was not representative of the populations they were trying to project to. I think many PR firms are much more knowledgeable and understand that the fundamentals haven't changed as long as they look at the research they're getting with regard to those fundamentals. Then they understand that well-done online research can be such an incredible thing.

Q: Tell me one of your favorite online research success stories

Humphrey: This is my chance to tell you my favorite story! A few years ago we were pitching to a small pharmaceutical company, and they were very happy with their existing research providers and they more or less said, "Thanks, but no thanks." As we were leaving we said what salespeople sometimes say stupidly, "Well, if you have a really difficult problem, why don't you come back to us." A few weeks later the phone rang and they said, "We have a really difficult problem."

It was Friday morning. They said, "Next Wednesday we have a Board meeting and our CEO wants to go into the Board meeting with data about how patients taking our drug compare with patients taking two other competitive drugs." We wrote a questionnaire, got it approved by 3:00 in the afternoon on Friday, and sent it out into the field that same day. By Sunday morning we had several thousand responses including more than 500 taking drug "A," drug "B," and drug "C." We wrote a report and when the CEO came in on Monday morning, the next working day (after the day they called us), there were results from a survey comparing the three different samples taking the three competitive drugs. He looked at that and said, "That's terrific, but if I had known these were the answers they were going to give, I would've liked to ask them two more questions." That was fine. We wrote the two additional questions. We went back to them on Monday afternoon. We had data in on Tuesday, and we wrote it up on Tuesday. The CEO went back into his Board meeting on Wednesday with two waves of interviewing with very large numbers of very hard to reach people. As a result, we gained a new, happy client.

One other example is also worth mentioning. The biggest survey we ever did involved 1.2 million people in over 90 countries around the world in maybe eight different languages. We were able to do that in about 10 days, which is sort of mind-boggling.

Q: Where do you see online research going in the future?

Randall: I think there's always a creative and economic tension there that some companies start to recruit market research professionals internally to help them figure out information and then they go through downsizing cycles, which then causes them to look to the outside. I think the market research field in general has seen companies try to have this internalized specialty, and then outsource it...and then try to have the internalized specialty again. I think it goes back and forth. I see it as cyclical, not necessarily as a growth concept for the next 10 years.

Humphrey: I've found there are two cycles that go on. I'm extremely old. I've been around much longer than young John and Randall here. I have seen many clients go back and forth between downsizing where they outsource research and then hire in people because they think it's too expensive to outsource it. Usually every time a new CEO comes along he will reverse what his predecessor did. You downsize, you hire in, you downsize, you hire in.

The second trend is what I call the decentralization and the centralization trend. In the '70s and '80s I was doing an enormous amount of work for a large financial institution, and every few years they would centralize everything into one big marketing research department for all their different divisions. Then they would have a new CEO or head of marketing and they would decentralize everything. Each team in every division had its own marketing research department. Then, another CEO or some manager would come along and they would centralize it all again. I think you can make a strong case for all these things but new managers like to make changes, so these two trends go on forever.

Randall: Now, in terms of online research growth, I see the tools getting increasingly visual. I see interfaces being more intuitive and potentially getting into peoples' brains a lot more effectively. In ten years, I predict that we will measure things a little more accurately than we can today. We'll be able to render decisions that people have to go through, or simulate them more adequately, so that we understand and arrive at better predictions. This is something that's been going on for the past ten years, but I see it continuing onward. I think that tools will get increasingly visual and orally presented as well, but through the computer.

John: The discussion of where research is going in the future always goes in a few directions. The first direction is technology: What advancing technologies are we going to see. The second direction is the respondent experience. Even the technology discussion goes in two directions. One, the computer itself is changing. I think today is actually the release of Windows Vista, the new software that's supposed to incorporate a lot more things in it. It's supposed to become part of your living room. Some would claim that Mac has already done this, but I'm not going to get into that argument. Really it's the advancement of technology. There have been some advances in cell phone technology that might lead to more people using different tools to do what we'll call "online research." We also talked about some things earlier, which is what sort of technology is going to enable the client and the respondent to see more information and have more convenience in the way that they respond.

I think the real benefit of a market research company is not just data; it's analysis of that data to provide better understanding and insight. I think the client is more involved now to some degree. We bring them more into the process. But it's still through the guidance of the market research community that gives them the right answers. There are some who just want to do it on their own, but as Randall and Humphrey have said, that's more statistical in nature. So you've got the application of being able to see the data in real time and then you've got the respondent experience, which is what Randall has just gone through. I see qualitative moves going back to being what we call face-to-face in virtual settings—with an ability to see everyone in the room. One criticism of online focus groups is that although they might provide excellent information, you're not really getting the emotion or the body language. As we can move to the visual video aspect; we are going to get to see emotion and things of that nature.

The analytic methodology is being used better with online research and will just continue to get better. This sounds sort of geeky, but at the major survey research organizations, we're just getting more capable statistically and better at what we're doing. We're getting better at taking advantage of all these powerful resources you can utilize online. That's going to really contribute in the next ten years. Even with the coming advances, I think the fundamentals of market research will remain the same—thoughtful consideration of sample, well-designed questions and responses, and appropriate analyses.

As you explore research with Internet resources, there are several key points you need to remember:

• If you haven't started your research program, get started. The Internet is the best place to start—it has everything you need.
• Your brand is never too small or too big to do research at different intervals.
• The Internet has incredible, interactive tools to research the 21st Century customer, from your in-house intelligence gathering to formal research with a third-party service provider.
• If you outsource your research program, make sure you take the time to select the right partner—one who guides you through the process.
• Take advantage of new 2.0 research strategies that will provide you with accurate and instantaneous intelligence.
• Online research enables you to collect more data from more people, and you have the ability to get bigger sample sizes. Be more interactive and do more in-depth analysis of your population.
• Online research enables you to test audiences visually in ways that were never possible before the Web.
• New research strategies are continually improving, and in the future you'll see interfaces being more intuitive and tapping into audiences' brains much more effectively.

Endnotes

1. Gosling, Samuel D. and Simine Vazir of The University of Texas at Austin, Sanjay Srivastava of Stanford University and Oliver P. John of University of California, Berkeley, "Should We Trust Web Based Studies?" Feb/Mar (2004), American Psychologist.