Welcome visitor you can log in or create an account

800.275.2827

Consumer Insights. Market Innovation.

blog-page
Rajan Sambandam

Rajan Sambandam

Chief Research Officer

Brand PerceptionsIs the Mini Cooper seen as an environmentally friendly car? What about Tesla as a luxury car? The traditional approach to understanding these questions is to conduct a survey among Mini and Tesla buyers (and perhaps non-buyers too, if budget allows). Such studies have been conducted for decades and often involve ratings of multiple attributes and brands. While certainly feasible, they can be expensive, time consuming and can get outdated over time. Is there a better way to get at attribute perceptions of brands that can be fast, economical and automated?

Aron Culotta and Jennifer Cutler describe such an approach in a recent issue of the INFORMS journal Marketing Science, and it involves the use of social media data – Twitter, in this case. Their method is novel because it does not use conventional (if one can use that term here) approaches to mining textual data, such as sentiment analysis or associative analysis. Sentiment analysis (social media monitoring) provides reports on positive and negative sentiments expressed online about a brand. In associative analysis, clustering and semantic networks are used to discover how product features or brands are perceptually clustered by consumers, often using data from online forums.

Breaking away from these approaches the authors use an innovative method to understand brand perceptions from online data. The key insight (drawn from well-established social science findings) is that proximity in a social network can be indicative of similarity. That is, understanding how closely brands are connected to exemplar organizations of certain attributes, it is possible to devise an affinity score that shows how highly a brand scores on a specific attribute. For example, when a Twitter user follows both Smart Car and Greenpeace, it likely indicates that Smart Car is seen as eco-friendly by that person. This does not have to be true for every such user, but at “big data” levels there is likely to be a strong enough association to extract signal from the noise.   

What is unique about this approach to using social media data, is that it does not really depend on what people say online (as other approaches do). It only relies on who is following a brand while also following another (exemplar) organization. The strength of the social connection becomes a signal of the brand’s strength on a specific attribute. “Using social connections rather than text allows marketers to capture information from the silent majority of brand fans, who consume rather than create content,” says Jennifer Cutler, who teaches marketing at the Kellogg School of Management in Northwestern University.

Sounds great in theory, right? But how can we be sure that it produces meaningful results? By validating it with the trusted survey data that has been used for decades. When tested across 200+ brands in four sectors (Apparel, Cars, Food & Beverage, Personal Care) and three perceptual attributes (Eco-friendliness, Luxury, Nutrition), an average correlation of 0.72 shows that social connections can provide very good information on how brands are perceived. Unlike with survey data, this approach can be run continuously, at low cost with results being spit out in real time. And there is another advantage. “The use of social networks rather than text opens the door to measuring dimensions of brand image that are rarely discussed by consumers in online spaces,” says Professor Cutler.

...

Recently I had lunch with my colleague Michel Pham at Columbia Business School. Michel is a leading authority on the role of affect (emotions, feeling and moods) in decision making. He was telling me about a very interesting phenomenon called the Emotional Oracle Effect – where he and his colleagues had examined whether emotions can help make better predictions. I was intrigued. We tend to think of prediction as a very rational process – collect all relevant information, use some logical model for combining the information, then make the prediction. But Michel and his colleagues were drawing on a different stream of research that showed the importance of feelings. So the question was, can people make better predictions if they trust their feelings more?

To answer this question they ran a series of experiments. As we researchers know, experiments are the best way to establish a causal linkage between two phenomena. To ensure that their findings were solid, they ran eight separate studies in a wide variety of domains. This included predicting a Presidential nomination, movie box-office success, winner of American Idol, the stock market, college football and even the weather. While in most cases they employed a standard approach to manipulate people’s feelings of trust in themselves, in a couple of cases they looked at differences between people who trusted their feelings more (and less).

Across these various scenarios the results were unambiguous. When people trusted their feelings more, they made more accurate predictions. For example, box office showing of three movies (48% Vs 24%), American Idol winner (41% Vs 24%), NCAA BCS Championship (57% Vs 47%) and Democratic nomination (72% Vs 64%), weather (47% Vs 28%) were some of the cases where people who trusted their feelings predicted better than those who did not. This, of course, raises the question of why? What is it about feelings and emotion that allows a person to predict better?

The most plausible explanation they propose (tested in a couple of studies) is what they call the privileged-window hypothesis. This grows off the theoretical argument that “rather than being subjective and incomplete sources of information, feelings instead summarize large amounts of information that we acquire, consciously and unconsciously about the world around us.” In other words, we absorb a huge quantity of information but don’t really know what we know. Thinking rationally about what we know and summarizing it seems less accurate than using our feelings to express that tacit knowledge. So, when someone says that they did something because “it just felt right”, it may not be so much a subjective decision as an encapsulation of acquired knowledge. The affective/emotional system may be better at channeling the information and making the right decision than the cognitive/thinking system.

So, how does this relate to market research? When trying to understand consumer behavior through surveys, we usually try to get respondents to use their cognitive/thinking system. We explicitly ask them to think about questions, consider options and so on, before providing an apparently logical answer. This research would indicate that there is a different way to go. If we can find a way to get consumers to tap into their affective/emotional system we might better understand how they arrived at decisions.

...

You may have heard about the spat between Apple and Samsung. Apple is suing Samsung for alleged patent infringements that relate to features of the iphone and ipad. The damages claimed by Apple? North of 2 billion dollars. The obvious question is how Apple came up with those numbers? The non-obvious answer is, partly by using conjoint analysis – the tried and tested approach we often use for product development work at TRC.    

Apple hired John Hauser, Professor of Marketing at MIT’s Sloan School of Management to conduct the research. Prof. Hauser is a very well known expert in the area of product management. He has mentored and coauthored several conjoint related articles with my colleague Olivier Toubia at Columbia University. For this case, Prof. Hauser conducted two online studies (n=507 for phones and n=459 for tablets) to establish that consumers indeed valued the features that Apple was arguing about. Details about the conjoint studies are hard to get, but it appears that he has used Sawtooth Software (which we use at TRC) and used the advanced statistical estimation procedure known as Hierarchical Bayes (HB) (which we also use at TRC) to get the best possible results. It also appears that he may have run a conjoint with seven features, incorporating graphical representations to enhance respondent understanding.

There are several lessons to be learnt here for those interested in conducting a conjoint study. First, conjoint sample sizes do not have to be huge. I suspect they are larger than absolutely necessary here because the studies are being used in litigation. Second, he has wisely confined the studies to just seven attributes. We repeatedly recommend to clients that conjoint studies should not be overloaded with attributes. Conjoint tasks can be taxing for survey respondents, and the more difficult they are, the less attention will be paid. Third, he is using HB estimation to obtain preferences at the individual level, which is the state of the science approach. Last, he is incorporating graphics wherever possible to ensure that respondents clearly understand the features. When designing conjoint studies it is good to take these (and other) lessons into consideration to ensure that we get robust results.

So, what was the outcome?

As a result of the conjoint study, Prof. Hauser was able to determine that consumers would be willing to spend an additional $32 to $102 for features like sliding to unlock, universal search and automatic word correction. Under cross examination he acknowledged that this was stated preference in a survey and not necessarily what Apple could charge in a competitive marketplace. This is another point that we often make to clients both in conjoint and other contexts. There is a big difference between answering a survey and actual real world behavior (where several other factors come into play). While survey results (including conjoint) can be very good comparatively, they may not be especially good absolutely. Apple used the help of another MIT trained economist to bring in outside information and finally ended up with a damage estimate of slightly more than $2 billion.

...
Recent comment in this post - Show all comments
  • Ed Olesky
    Ed Olesky says #
    how interesting! thanks for sharing this, Dr. Sambandam. i wonder how many price points they tested. and was it subsidized price,

What Does the Fox Say?

Posted by on in Market Research

Nate Silver’s much anticipated (at least by some of us) new venture, launched recently. In his manifesto he describes it as a “data journalism” effort, and for those of us who have followed his work over the last five years – from the use of sabermetrics in baseball analysis through the predictions of presidential politics – there is plenty to look forward to. Apart from the above topics, his website is focusing on other interesting areas such as science, economics and lifestyle, bringing data-driven rigor and simple explanation to the understanding of all these fields. It follows the template of the blog he ran for the New York Times as well his bestselling book, The Signal and the Noise: Why So Many Predictions Fail, But Some Don’t. As a market researcher, I found much to like in the basic framework he has laid out for his effort.

In critiquing traditional journalism, Nate describes a quadrant using two axes – Qualitative versus Quantitative, and Rigorous & Empirical versus Anecdotal & Ad-hoc.

qual quant market researchSource:www.fivethirtyeight.com

He is looking to occupy the mostly open top left quadrant, while arguing that opinion columnists too often occupy the bottom right quadrant and traditional journalism generally occupies the bottom left quadrant. For someone with such a quantitative background he is not dismissing the qualitative side at all. On the contrary, he argues that it is possible to be qualitative and rigorous and empirical, if one is careful about the observations made (and cites examples of journalists such as Ezra Klein, who occupy the top right quadrant).

For those of us in market research the qualitative versus quantitative dimension is, of course, very familiar. Somewhat less so is the second dimension – rigorous and empirical versus anecdotal and ad-hoc. But this second dimension is especially important to consider because it directly affects our ability to appropriately generalize the insights we develop. As practicing researchers, we know that qualitative research is excellent for discovery and quantitative is great for generalizations. But we also know that is not always the way things are done in practice.

...

research conference may13 2013We just wrapped up another of our client conferences and it was another successful day for all concerned. This conference stood out for the level of interaction between the speakers and the audience, a testament to the speakers, their topics, and the keen interest that practitioners have in these topics.  

The first speaker was Olivier Toubia from Columbia University. Olivier is a true leader in the area of innovation research and teaches an MBA course called Customer Centric Innovation. He gave a quick round up of four important questions that he has been able to address through his research – how to motivate consumers to generate ideas, how to structure the idea generation process, how to screen and evaluate the ideas and how to find consumers who have good ideas. By taking us through a variety of studies (including surveys and experiments) he was able to answer these questions and provoke a lot of interesting thoughts from the audience.

Next up was Vicki Morwitz from New York University. She uses surveys extensively in her research and is a leader in understanding the impact that survey responses have on subsequent behavior. She was able to present evidence about the unintended effect that surveys have on respondents, something that should be of interest to all marketing research firms and indeed all marketers. In some cases surveys have a positive impact in that they increase future purchasing behavior, but said Vicki, should be used with caution as overt efforts to influence consumers do not seem to work.

Vicki’s presentation was followed by TRC’s own Michael Sosnowski who discussed the idea of doing more with less in a mobile world. He talked about the increasing numbers of survey respondents who are attempting to get at surveys using their smartphones and why we as researchers should be aware of that. He questioned the conventional wisdom that mobile phone surveys should be short and simple and showed examples of more complex choice based surveys (using TRC’s Bracket) can be conducted on mobile phones and how it provides results similar to an online survey. We may not be ready to do conjoint studies on mobile phones, he said, but neither should we artificially constrain ourselves to extremely simple data collection. Using good design and sophisticated analysis it is possible to get good quality information from mobile surveys.

Following Michael was Joydeep Srivastava from the University of Maryland an old friend of mine from my graduate school days. He is now a leading consumer behavior researcher who has done especially interesting work in the area of pricing. His specific interest is in partitioned pricing (such as charging a separate price for shipping) and he was able to enlighten the audience with the results of his experiments. For example, he was able to counter the myth that charging a separate shipping price and then providing a price discount to offset it would stave off any damage to the company. On the contrary, it actually reduced the purchase likelihood compared to not providing a discount. This, he said, was because of people’s unwillingness to pay for shipping in the first place and the explicit reminder of it with the offsetting charge.      

...

My Evening with Daniel Kahneman

Posted by on in Consumer Behavior

Okay, so it wasn’t really just the two of us – there were a few hundred others involved. Still, it was a very memorable evening that I think is worth sharing.

The day started innocently enough. I was heading out to Yale for a guest lecture in the MBA Marketing Research class taught by Jiwoong Shin as I have done for several Spring semesters now. I like this trip a lot as it allows me to catch up with many of my friends in the Yale Marketing Department. One of those is Shane Frederick and I had emailed him to see if he was around. He replied asking if I was attending Kahneman’s lecture. I had no idea that Daniel Kahneman, Nobel Prize winner and godfather of behavioral economics was giving a lecture there. The day was already getting better! I quickly changed my Amtrak ticket to a later time and told Shane I would come by his office so we could walk over.

My guest lecture went off very well with the students asking plenty of interesting questions. Then I had lunch with Zoe Chance who is doing some very interesting work with leading companies, applying ideas from behavioral economics. After a couple more meetings, I went to see Shane and we walked over early knowing there would be a big crowd. And we were glad we did, as the auditorium was overflowing by the time the lecture started.

Daniel Kahneman (Danny to his friends) was introduced by another notable person from Yale, Professor Robert Shiller (yes, he of the Case-Shiller Index you may have heard about during the housing crisis). Shiller talked about the widespread impact of Kahneman’s work, especially after the publication of his best seller Thinking, Fast & Slow. Trying to find Kahneman’s connections to Yale, Shiller pointed out that two of his coauthors (Shane Frederick and Nathan Novemsky, both in the marketing department) were at Yale.

And then it was time for Kahneman to speak. His humility, thoughtfulness, and eloquence came through pretty much from the first few words. He started by saying that he doesn’t do university speeches anymore since he is not actively doing any research (he is retired), but could not say no to Bob Shiller. Most of his recent speeches have been about his book, and there had been so many that as a consequence he seems to have forgotten everything else he ever did (laughter!). And that, he said, makes sense because as he points out in the book, we like things that are familiar (more laughter!).

...

electoral map 2012 nov 7tWas the election outcome a surprise for you? It wasn’t for me.

In some ways election night was quite boring. And I blame Nate Silver, Sam Wang and others who predicted the outcome with such stunning accuracy that (at least for me) the drama was completely missing. While conventional pundits and partisans were making all kinds of predictions ranging from “Toss-up” to “Romney landslide”, a group of analysts (nerds, if you choose) were quietly predicting that Obama had a small but consistent and predictable lead. Turns out they were spot-on in their predictions (and were predictably smeared by vested interests).

In my last post I talked about Nate Silver and the approach he uses. This time I want to draw your attention to another analyst, Sam Wang of the Princeton Election Consortium. He is a neuroscientist who has been forecasting for the last three presidential election cycles and has been doing a remarkably good job of it. He nailed the Electoral College vote in 2004 and missed by just one in 2008. How did he do this time? Well, he had two predictions. One of them (based on his median estimator) was 303 for Obama, which is where the tally currently stands, subject to Florida being officially called. The second one (based on his modal estimator) was 332 for Obama which is where the tally is likely to end up if/when Obama wins Florida. Excellent calls whichever way you look at it, given the extremely close race in Florida.

I’ll give you the simple answer. Surveys!

No, I don’t mean looking at whatever survey happens to catch your eye or tickles your (or your favorite network or blog’s) ideological fancy. I mean, using a system that is powered by old fashioned surveys and making very, very good explanations and predictions based off that. There is someone who has been doing exactly that for several years now and it makes sense for anyone interested in surveys to understand how he is doing that. I’m talking, of course, about Nate Silver at fivethirtyeight.com.

Interestingly, Silver does not actually do a single survey himself. Instead what he has done is build a database of surveys (that contains thousands) and used some simple and clear rules to analyze them. Based on these rules and the statistical models he has built, he is able to provide the best, unbiased view of the race. All this from survey data. How does he do it? Let’s take a look at some (and by no means all) of his rules.

In Thinking, Fast & Slow, Nobel winner Daniel Kahneman (click here previous post about Thinking, Fast & Slow) talks about the two selves people have: the experiencing self and the remembering self. The terms are self-explanatory and vacations are a good way to think about them. The part of us that is enjoying the vacation is the experiencing self, while the part that is reliving it later (sometimes years later) is the remembering self. Neither one may be more important, but the emphasis we place on one or the other could determine our behavior. So, for example, you can enjoy the vacation or take plenty of pictures to relive it later, depending on the self that is more important. A way of finding out which self is more important is to ask ourselves whether we would go on a certain vacation if we could only enjoy it, but not take any pictures (or video, etc).

market research conference 2012Well, another conference is over, perhaps our best ever. A great roster of speakers, a room full of engaged attendees and a great location was a terrific formula for a memorable conference. Some highlights from the various sessions:

Lenny Murphy, Editor-in-Chief of the Greenbook blog opened with a wide sweep discussing the waves of changes rocking the market research world. Pulling from the GRIT survey, his discussion with emerging and established players, as well as his itinerant investigation, he was able to convincingly make the case that change in the MR industry is happening. Now. He talked about emerging technologies such as mobile, social media and text analytics and how academic expertise was a key to unlocking a future of new ideas. It was a perfect set-up for the group of academic presentations that were to follow.

The Outside View that Daniel Kahneman talks about in his book Thinking, Fast & Slow, is a specific remedy to a problem known as the planning fallacy (i.e.) the inability of people to make predictions. The planning fallacy is part of a larger problem of optimism bias. What is optimism bias? Simply put, people are generally more optimistic than they should be. For example, it is well known that most people think they are better than average drivers, an impossibility. It stems from a general dose of overconfidence not warranted by the situation on hand.

The best example of overconfidence is a study that Kahneman cites of CFOs of large corporations. They were asked to estimate the returns of the S&P Index over the following year. The data were collected over a number of years and hence there was ample opportunity to correlate it with the actual performance of the Index in the following year. Any guesses as to this correlation, given that the respondents should have been expected to have special insight in this matter? It was almost exactly zero, slightly less, in fact! And they seemed to have no idea their forecast was that bad.

Tagged in: Psychology

daniel-kahneman-thinking-fast-slowIn his opus Thinking, Fast & Slow, Nobel winner Daniel Kahneman (click here for previous post) relates a story from early in his career when he was leading a team to develop a curriculum and write a textbook on judgment and decision-making in high schools. He had assembled a group of experts and after working diligently for a year they had completed an outline of the syllabus and written two chapters. One fine day when discussing procedures for estimating uncertain quantities, it occurred to him that he should get an estimate from everyone on how long he thought this whole project would take. Being the clever psychologist that he was, rather than ask the group to guess publicly, he asked each person to make a confidential prediction. The mean was about two years and the range was about half a year on either side. In other words, the group was very consistent in its prediction.  

The Nobel Prize winner and the intellectual godfather of behavioral economics, Daniel Kahneman, has summarized a lifetime of research in his recent book Thinking, Fast & Slow. In the next few blog posts I will be drawing upon some concepts that he espouses and link them up to research to see what practitioners can take away from his four decades of work.

This post goes directly to the title of the work; fast and slow thinking. This is the foundation of his work. He and his great collaborator Amos Tversky, (who passed away and therefore could not receive the Nobel) see human thinking in two forms that they call System 1 and System 2. More aptly they could be called “automatic” and “effortful” systems, but Fast and Slow is a good shorthand description. According to Kahneman’s description,

System 1 operates automatically and quickly, with little or no effort and no sense of voluntary control

System 2 allocates attention to the effortful mental activities that demand it, including complex computations”

blackswanThe Black Swan is a book that was published a few years ago and generated much publicity and at least some controversy. It occurred to me that there are lessons market researchers can learn from that book, particularly about the relationship between qualitative and quantitative data obtained from a survey format. The idea is that the framework used to analyze such data is different from that used for directly obtained qualitative data through methods such as IDIs and focus groups. Understanding the difference between quantitative and qualitative frameworks for data analysis (and in particular, the difference between statistical and managerial outliers) can help derive more value when the qualitative data are collected in a regular survey. But first, let's take a detour.

A Brief Tour of The Black Swan

In his informative (and entertaining) book, Nassim Nicholas Taleb argues that real data are either distributed normally (from "mediocristan") or not (from "extremistan"). The former are characterized by data that follow the traditional normal distribution (or bell curve). The majority of the distribution is near the middle surrounding the average and as we venture further out the number of observations becomes increasingly scarce. It is a distribution that defines many phenomena in the natural world. In fact, basic statistics shows that with a reasonable number of observations most distributions start approximating the normal.

what am i supposed to doYes, it is a rather important issue and can be approached in a variety of ways. My purpose with this post is not to provide a comprehensive answer, but look at one specific solution based on what I recently read. The book is Thinking, Fast and Slow, the Nobel Prize winner Daniel Kahneman's excellent summary of a lifetime of research. He is perhaps the most accomplished psychologist around and could (among other things) justifiably be called the intellectual godfather of behavioral economics. It is always worth listening to what he says and in this particular case, it seems to me there is a nugget that applies to making quantitative research more actionable.

How to Be Happy by Spending Money Wisely

Posted by on in Consumer Behavior

It is that time of year when many people's thoughts turn towards buying gifts for loved ones. More generally it is a time when thoughts related to money and happiness occupy our attention. When thinking of ways to spend money either on oneself, for loved ones or even for complete strangers wouldn't it be nice if there was some actual research to provide data-based guidance on the topic? As it happens, there is. Researchers Elizabeth Dunn of the University of British Columbia, Daniel Gilbert of Harvard and Timothy Wilson of the University of Virginia have identified, through their research, eight principles designed to help consumers get more happiness for their money. Follow them as you will to enhance your life.

Tagged in: Psychology

Thoughts on TMRE 2011

Posted by on in Conferences

tmre_banner_250x250_nodisI recently came back from the 2011 The Market Research Event (TMRE) conference in Orlando, the biggest marketing research conference of the year. There was plenty to like, not the least of which was the scale of the event. Rarely, if ever, do we get to see an exclusively market research event that is so big. Kudos to IIR for putting it together.

The highlight of the event for me was the Keynotes, of which there were eight. I couldn't catch all of them, but my favorite was Sheena Iyengar from Columbia, author of the best seller The Art of Choosing (and sister-in-law of my friend Raghu Iyengar from Wharton). In a beautifully choreographed and clear presentation, Sheena (who is blind) talked about the problem of plenty in consumer choice and ways to avoid it for both sellers and buyers. The Keynotes were all held in a massive room and very entertainingly emceed by Cayne Collier, an actor and improv artist from Second City Chicago. Discussions with a variety of people indicated that the Keynotes were the favorite part of the conference for many.

Tagged in: Market Research
Recent comment in this post - Show all comments
  • Ed Olesky
    Ed Olesky says #
    Thanks for the comments Rajan! I agree with you, though I do think TMRE did a fairly good job reporting on the NGMR Disruptive In

As I sat down to write I realized that this is not a simple question. Consider the conventional meaning of necessities (defined as must-haves) and luxuries (defined as nice-to-haves). Which category market research falls into may depend on the eye of the beholder.

Researchers (or more accurately research sellers) may want to think of themselves as producing necessities rather than luxuries. But in the consumer world necessities are also generally commodities and often sold based on price. Researchers of course want to be seen as producing something valuable, something that is worth a premium -- in other words, a luxury.  So, which is it?

Now let's look at it from a research buyer's perspective. The buyer may think of research as a necessity, something that is indispensible for making good business decisions. But in keeping with the popular perception of necessities, perhaps they feel that more than one company can provide it and are hence unwilling to pay much of a premium for it. This view would support the many research sellers who complain about the commoditization of research.

center 4 neural decision makingRecently I was invited to attend a neuroscience conference at Temple University in Philadelphia, organized by their Center for Neural Decision Making, along with MIT and the University of Michigan. It turned out to be a very interesting experience with excellent speakers, great interactions and a terrific panel discussion. Some highlights:

  • Michael Norton of Harvard spoke with humor about the sensitive topic of racial paralysis. This is the tendency of people to refrain from making any decisions when faced with a situation where they could potentially be perceived as racist. His approach used data from experiments, surveys and neural imaging, a nice way to triangulate the results.
Tagged in: Neuroscience

Why MasterCard Ads Are Priceless

Posted by on in Advertising

You remember the MasterCard "Priceless" ad campaign, don't you? It first ran during the 1997 World Series.

"Two tickets: $28. Two hot dogs, two popcorns, two sodas: $18. One autographed baseball: $45. Real conversation with 11 year old son: Priceless."

It was an ad campaign that was so successful that it helped MasterCard move from a distant second to near parity with Visa. The question is, why? What was it about that ad that was so powerful, asked researchers Jeffrey LoewensteinRaj Raghunathan and Chip Heath  (who is incidentally a co-author of the best seller Made to Stick). What they found holds lessons for companies looking to create successful ads.

Tagged in: Advertising

Want to know more?

Give us a few details so we can discuss possible solutions.

Please provide your Name.
Please provide a valid Email.
Please provide your Phone.
Please provide your Comments.
Enter code below : Enter code below :
Please Enter Correct Captcha code
Our Phone Number is 1-800-275-2827
 Find TRC on facebook  Follow us on twitter  Find TRC on LinkedIn

Our Clients