Thoughts about the future: Market Research

Approaching people in person often doesn ’t work very well either. If a stranger comes to your door, your fi rst reaction is likely to be that this might be a salesman, a burglar casing the house, or a robber ready for you to open the door so he or she can barge in. Similarly, if someone approaches you in a parking lot, especially if they ask you to get in a truck to look at something, you may immediately think this could be a kidnapping or worse, and you scurry away.

Can virtual market research hold the key to coming up with better results more cheaply? Today, market research seems to have more hurdles to overcome than ever. When I fi rst worked in market research in the late 1960s, it was relatively easy to call people on the phone and talk to them, go door to door to conduct random samples, or take a truck to a local supermarket parking lot to get shoppers to try out and comment on new products. I even was an eager cooperative subject after I left the  field in the early 1970s and market researchers called me, came to my door, or approached me in a parking lot. I wanted to help out, since I had recently been a researcher myself.

But no more. Neither I nor many other people are receptive to market researchers. One reason is that you don ’t know when people call if they are who they say they are. And you often don ’t wait long enough to hear their credentials. You think they are selling something and/or are very busy multitasking. So if you don ’t know someone or don ’t immediately feel this will be a valuable call, you don ’t want to stay on the line. Also, once market researchers start asking questions, your suspicious antennae go up. You wonder: is this a con artist posing as a market researcher to get confi dential information?

Another problem marketers face is that it can be expensive to pay people to come in for focus groups—commonly $75 for each person, plus the phoning costs to get them to a one- to two-hour session. Commonly, it takes fi ve hours of phoning to get one person to agree to be in the session. Then, add on the expenses for food, office, equipment, the facilitator, and other materials for running a session and it brings the cost for each session to $5,000 or $6,000 or even more.

So, marketers need a more effective, less expensive way to fi nd out reactions to ideas for new products, prototypes, proposed product and company names, and other questions. Enter the Internet. For the last decade or so, there have been online surveys. I used to get such requests from time to time to spend 10 to 15 minutes answering questions in return for being part of a lottery to receive some small prize. But then the online questions stopped, maybe because many people like me didn ’t want to spend the time to answer the questions for a very small chance to win an inexpensive prize. Plus there was an anonymous questioner concern: was this really a researcher or was it someone seeking private confi dential information?

Now maybe there is another way, provided by researchers looking at the behavior of people participating in online games or buying virtual goods. According to a 2010 article in Newsweek, these activities provide clues to human behavior. The article describes economist Kristian Segerstrale who runs Playfi sh, a maker of online games and a top virtual goods seller. Segerstrale believes the virtual goods world opens up a new way to study human behavior by observing how people interact in this environment.

The way Playfi sh works, like many other social games, is that participating in the game is free. But then the company makes money by selling virtual goods inside the game, as do other merchants in the game environment. As of 2009, this was a $1 billion market, up from $500 million in 2008, and is expected to grow to $1.6 billion.

These virtual worlds also seem to be a perfect setting for market researchers to learn what different types of people do or might do in response to a new product or service. In Segerstrale ’s view, real-world information is always imperfect because of the small subset of people willing to respond to surveys and because people may often lie or say what they think the researcher wants to hear. But in virtual worlds, you can get a “perfect data set” because you know “every data point with absolute certainty” and you even know who all the people are, so you can group them by gender, age, or any other category.

Then, too, in a virtual universe one can “experiment in real time.” For example, you might learn what happens if you add a small tax to one product and not to others or if you set the tax at different levels. Then you can compare how different groups of people respond, such as whether men will be more receptive than women.

Increasingly, as the article describes, academics have started examining behavior in these virtual marketplaces to learn about virtual goods and currencies and how social factors affect how people make decisions. A key organization in this effort is the Virtual Economy Research Network started in 2006 by Vili Lehdonvirta, an economic socialist at the Helsinki Institute for Information Technology. One big problem may be that even though the data are perfect and complete since one can observe responses in real time from everyone in the virtual world, the people in this virtual world might not be a truly representative sample of the complete population. After all, there might be some unique qualities about a person who participates in FarmVille, PetVille, or Mafi a Wars, even if there are over 20 million or more people involved—which is a much larger sample than market researchers have ever gotten for anything in that 1,000 people may stand in for 100,000 or more, using random sampling techniques.

Yet, in today ’s world, where more and more of the population is online and don ’t have the resistance to in-person market researchers that is so common now, these virtual worlds may be the best source of research there is. Perhaps one way to overcome the representative sample problem is to take into consideration the demographics of people participating in different types of games or social networks and then target market research studies and offers to them.

For example, to get more people to participate, marketers might offer bigger and better prizes with a better chance to win. Or companies might team up with other companies to offer signifi cant savings in coupons on something those in the target demographic want to buy. Or researchers might use a few questions to determine what people want and then offer that in return for people answering a few questions.

Also, researchers might test out people’s behavior in the virtual world with what they would do in a similar situation in real life. For instance, besides varying a price increase in Pet Society, researchers might ask participants if they have a real pet and how they would respond to a price increase for a product they really would want to buy for their real pet. Or researchers might ask how one ’s participation in Pet Society might affect how they relate to their own pet.

In addition, researchers should test out how well their models for predicting behavior based on what people have done or on how they have answered questions in the virtual world compared to what they later do in the real world. Then, that will help researchers test out the accuracy of their virtual world models.

In sum, the virtual world does seem to offer a great opportunity for better market research, particularly in light of the real-world pitfalls of market research today. But it is a type of research that needs more testing. And researchers should particularly consider how active people are in these virtual worlds and whether they have continued to be participants. For instance, I may be on the rolls of FarmVille, but I haven ’t been very neighborly; I have turned down requests to be neighbors, contribute a plow, or receive gifts from others because I don ’t have the time, particularly now that I have over 3,500 friends on Facebook, many of them sending me messages, links, and writing on my wall. I suspect many other participants in these virtual worlds might similarly vary or drop their participation on and off over time. So researchers should take that into account, too.

Still, the more people participate in these virtual worlds, the more market researchers can learn from them. And perhaps, in the future, as things become increasingly challenging in the real world, people may spend more time in the virtual world, and the economy there may become even more real and robust.

Extracts taken from "The Very Next New Thing ~ Commentaries On The Latest Developments That Will Be Changing Your Life" by Gini Graham Scott (2011)