Ep #49: Elissa Moses and Neuromarketing Advances

Welcome to the Brainfluence Podcast with Roger Dooley, author, speaker and educator on neuromarketing and the psychology of persuasion. Every week, we talk with thought leaders that will help you improve your influence with factual evidence and concrete research. Introducing your host, Roger Dooley.

Roger Dooley:    Welcome to the Brainfluence podcast. I’m Roger Dooley. Our guest today is Elissa Moses. Elissa is in charge of the global Ipsos Neuro and Behavioral Science Center. In the past, Elissa has been with early neuromarketing pioneer EmSense, as well as big firms like Phillips and the ad agency Grey. Welcome to the show, Elissa.

Elissa Moses:     Good morning. I’m happy to be here.

Roger Dooley:    Glad you could make it. It’s great to have somebody with such diverse experience in both neuromarketing and consumer psychology here. Let’s talk about the Ipsos Neuro and Behavioral Science Center, which is a mouthful in and of itself. How big is this group and what kind of problems does it focus on?

Elissa Moses:     The group really has a focus of trying to bring all of the new kinds of non-conscious measurement tools that are of practical value to marketers and market researchers. We’re charged with a wonderful agnostic mission to be able to scout, identify and vet any kinds of tools that we put through different kinds of R&D paces, if you will.

One of the things that we really take pride in is not integrating any kind of tool into our standard offerings until we’ve really, really kicked the tires on them. We’ll often do a lot of pre-testing on our own and pilot testing which has to do with looking at test validity, being able to be sure that the measures or devices do what they say they’re going to do, and that we’re really getting the same consistent kind of results time after time.

Our focus is global and we have about ten people in the group right now plus some other people that kind of come in and out on a freelance basis, depending on how much experimentation we’re doing. The group is nicely divided. We have people that have strong academic backgrounds in cognitive neuroscience, PhDs, and they tend to lead our R&D effort. We also have wonderful people from a practical neuromarketing experience standpoint who have been working in this area for years and have footing both in market research and in neuromarketing tools to be able to bridge the gap.

Something that we really take part in as an organization is really to be able to build that bridge between traditional market research and neuroscience and understand where the added value is, where the practicalities are, and how we can adjust things.

We also pride ourselves in being able to know which tool to use when, because are agnostic. We don’t own any certain kind of company that might bias us to always wanting to say there’s a certain solution.

Roger Dooley:    Elissa, what are some of the different technologies that you’re using right now?

Elissa Moses:     Well, it’s quite a broad sweep. We do a lot of work with facial coding, with implicit reaction time, with biometrics, with eye tracking. With eye tracking we offer three kinds of eye tracking, from stationary to mobile to online. We also use EEG. We’re also experimenting with a lot of new tools in the wearable space, in the sound space. When you run a group like this, your work is never done. There are always new tools being developed.

Roger Dooley:    I guess that’s job security. Did you see the Temple study that just recently came out or is in the process of coming out that showed in their particular test only FMRI produced results that were better than simply asking people question as in a survey? Their test used a sort of very specific metrics of performance which was real world ad results to determine that.

Elissa Moses:     Yeah. I’m really familiar with …

Roger Dooley:    I’m curious what your take on that would be. I would guess that you would want to make the point that other technologies have their roles.

Elissa Moses:     I’m very familiar with that study because Ipsos was a sponsor and I was an advisor on the study, but I will say that I have … I want to overall applaud the initiative, because in the world of neuro we really, really need to have clients cooperate and give us in-market data so that we can validate in-market results and do modeling, so the study is quite wonderful from that perspective. I think it was very, very interesting, but I think in putting it together certain compromises perhaps had to be made in terms of how different tools were executed.

The people that executed the study, I think it fair to say were most expert in FMRI and I suspect that that may have influenced … It may have been a very, very good study for the FMRI part of it and being able to identify what it contributed. I’m not sure it was as good a study, for instance, from the implicit part of it. I think that they were kind of … that was a late entry, and I think they were kind of feeling their way on their methodology with that, did not include facial coding.

Ipsos was extremely happy because it also showed that the most important driver for predicting in-market results had to do with the questionnaire itself, which still supports the importance of cognitive measures, where the industry is in no position to abandon the importance of cognitive measures. Really the study supports our point of view, which is a very important …

Roger Dooley:    By cognitive measures, Elissa, you mean things that the consumers actually realize and can talk about themselves as opposed to things that are below their level of consciousness?

Elissa Moses:     Absolutely. Conscious measures very simply translates into old-fashioned surveys, questions and answers, being able to consciously respond to something. Even if you’re talking about what may be perceived more as an unconscious realm like emotion, you’re still filtering that through a conscious process.

Where the industry is right, and that study supports it very much, is that we really need both sides of the equation. I think that the jury is still out. We really don’t have enough studies and enough data to be able to link certain kinds of neuro or behavioral science measures to in-market results, and II think that Temple would agree that that study is not totally conclusive in that regard.

Roger Dooley:    I think it’s significant at least that respected academics took on this problem and took their solid first stab at it and we’ve got research that’s being done that isn’t be done by providers themselves. If you go to a skilled neuromarketing firm, any results they produce are going to be suspect simply because they have a dog in the fight. Here these folks did it on an independent basis and I think it opens the door for further study.

I think a lot of academics have steered clear … I talked to the researchers a couple of weeks ago and one of the things they sort of chuckled about was academic fear of being associated with neuromarketing for a while. It would be sort of like investigating parapsychology or something as a pseudoscience. Reframing it as consumer neuroscience or decision helps, but it really opens the door for other schools to take a look at this too and get more independent data out there.

Looking at the results of that study, one issue could be the fact that they had less expertise in certain types technologies than others, but also I wonder if in some cases different technologies are better for answering different kinds of questions?

Elissa Moses:     I think that’s true. I think that not all the tools do the same things. Some tools are more sensitive than others. Some tools are more practical than others. Some tools get certain types of response that different. In terms of designing studies, wanting to take System 1 metrics into consideration, you really have to have a deep understanding as we do about the pluses and minuses of different tools and when they’re best applied.

Roger Dooley:    Like implicit association testing, you mentioned that one. What would be a good application for that do you think? Describe the technique just real briefly for our listeners who may not be familiar with it.

Elissa Moses:     Sure. Implicit reaction time testing is based on a psychological phenomenon that the closer we hold things in association in our mind, the faster we are to react to that association. The way it’s measured is using a very structured kind of technique where you’re being asked whether you associate things, and by the speed of response we can tell if there’s really an emphatic close association or whether there’s some hesitation and uncertainty, which means that you might reply yes with lip service but not be so certain on the inside.

Roger Dooley:    Right. By hesitation you mean milliseconds; that really the individual wouldn’t be aware that they hesitated but it can be measured with the right equipment?

Elissa Moses:     Absolutely. It really takes place in … In the world of neuro, everything happens in fractions of seconds and the differences are really important.

Let’s say we went out to dinner last night to a new restaurant. If you say to me did you like the restaurant, if I say, “Oh yeah. Gosh, it was really great,” and I answer with passion very quickly, it’s more than likely that I have a lot of conviction in what I’m saying. If I skip a beat and, “It was nice. It was really nice,” that a very social kind of polite yes, but the truth is that maybe I didn’t think it was all that great. This is the way to quantitatively look at the difference between how we answer questions from a conviction standpoint.

Your question, as I recall, is what are the best uses of it and it kind of fits into the whole scheme of market research today. I’ll tell you, I led a panel last year of all of the top people at all of the large research companies that lead in neuroscience applications. I asked them which is the fastest growing tool and hands down everybody said implicit.

I think the reason is because implicit is so varied in its applications. It’s just so versatile, and it’s also the only tool in the neuromarketing practitioner’s toolbox that gets at impact. All these other tools that we’re used to, from facial coding, biometrics, EEG, really have to do with engagement. Are people getting more excited by what they’re looking at? Are they having an emotional reaction?

The way that implicit is structured really enables us to understand has change occurred? Do you feel differently about a brand after you see an ad? Do you feel differently about a product after you’ve tested it? A fragrance after you’ve smelled it? It’s really quite wonderful in looking at the potentials for change as well as getting a real understanding of what your internal emotional profile is about a brand or a product.

It’s also extremely versatile in terms of embedding it into surveys. With that respect it’s quite a wonderful tool because you’re able to at the end of the day have both the conscious and unconscious response and be able to integrate them, which we think is really the perfect way to do things. You have a really comprehensive understanding of how people really feel about a given topic.

Roger Dooley:    Right. I think Malcolm Gladwell famously described it in Blink and found out that he, himself, harbored racist tendencies, even though he’s multiracial and would have considered himself to be a liberal, unbiased individual. It can be quite …

Elissa Moses:     Right. So it was pretty shocking for him.

Roger Dooley:    Yeah. Yeah. But it really gets at the power of some of these nonconscious techniques. I don’t have the URL with me right now, but we’ll put it in the show notes. If anybody wants to try one form of the test, Harvard has an implicit association little demo thing that you can try on their site. We’ll link to that. It’s kind of fun just to see what your own biases might be.

Elissa Moses:     I think the Harvard place is a wonderful place to start to understand the basic concept of the tool, but there has been some serious strides in terms of simplifying the tool for embedding into studies so that it’s less wieldy. That’s what we offer today, a way of doing it that’s much more streamlined.

Roger Dooley:    One advantage of implicit testing is too that it’s relatively cheap and it’s scalable, compared to wiring people up with EEG; or worse, even sticking them in an FMRI machine. This is something that can be done even over the internet.

Elissa Moses:     This is true. However, it shouldn’t be mistaken that they’re getting the same thing. What implicit cannot do is give you any kind of moment-by-moment visceral reaction, unlike the other tools that you just mentioned. It certainly has its place.

Roger Dooley:    Oh, yeah. I wouldn’t say that it would replace those other tools, but it certainly … If somebody is looking to adopt a technology, it’s probably one of the least expensive ones that you can adopt, although it may provide different kinds of results than the other techniques.

Elissa, what do you think about the credibility of neuromarketing at big brands today? Are there still a lot of skeptics or is some of that doubt going away?

Elissa Moses:     I think it’s happening on multiple levels. I think that just about all of our clients really embrace the neuroscience behind neuromarketing. They really fundamentally believe and understand that emotions and the nonconscious has a huge influence on decision making. That is enough to propel them to want to at the very least experiment with various tools and see in translation which really has a powerful impact for them.

I just did a study, for instance, with a very major global client on fragrance using implicit and they were just blown away with how they were able to define some of the characteristics of the fragrance and what the fragrance means to people by using this tool.

The same has happened when I’ve applied it to package testing, and similarly with ad testing with facial coding. Clients are just so elated to be able to have insights that are really practical in terms of how they can optimize an ad or to understand how an ad perhaps really isn’t working.

I just did a study of all the winners of the Super Bowl. I shouldn’t say all the winners; I did a study of all the ads on the Super Bowl, and basically we were able to discern which ones we felt were best based on facial coding; not just on how engaging they were overall on facial coding, but particularly in the last five seconds of each of the ads.

The reason the last five seconds are so critical is because that’s when the big branding moment happens. Whether you’ve mentioned the brand or not in the whole buildup of the story, there’s usually that last brand reveal that ties everything together and is supposed to leave a lasting impression on you. If your story has been riveting but people check out at the branding moment, you’ve really lost your money’s worth, particular in a Super Bowl ad.

These kinds of tools clients are finding offer very great, practical value to them in terms of insight. Really the next horizon will be to try to tie some of these measures to in-market response that goes beyond the Temple study and really looks at things like facial coding, looks at implicit done in the best possible way, looks at biometrics.

There’s a lot to still learn, but I would say that the jury right now is that these tools offer great practical value.

Roger Dooley:    Elissa, on the subject of facial coding, are you a believer in automated facial coding analysis? That would be more or less by a camera or webcam connected to software that analyzes expressions, or are you more of a believer in the human expert analysis, using video or slow motion video?

Elissa Moses:     Believer is a funny word. One is an advancement …

Roger Dooley:    How about user for starters? What do you actually do in practice?

Elissa Moses:     We use both. One is an advancement built upon the other, I would say, by virtue of the fact that the traditional way of doing it with people hand coding and experts analyzing facial coding. The fact that that worked and was useful propelled the industry to look for an automated way to do it because it’s more expensive to do it the traditional way and the automated way has been highly successful.

We do use both because I feel that the traditional way is probably the best application for qualitative or for individual type responses. In terms of quantitative research and anything that’s scalable, we’ve been enormously satisfied with our partnership with facial coding where we’re getting tremendous discrimination and we have multiple country norms and we’re really able to compare apples to apples results with great discrimination, at low cost, and retest reliability.

It’s just a great tool, plus clients don’t have to wait. The results on facial coding are totally integrated into the report and you don’t skip a beat. It’s all passive measurements. Both have a role in market research and I support both of them.

Roger Dooley:    Elissa, the subject of scalability has come up a few times. Before Ipsos, you were at EmSense, who was one of the earlier firms in the neuromarketing space. One thing that they tried to do was scale EEG testing by distributing inexpensive little headset that could be used by remote web users as they viewed content and then measure their reactions and analyze that. Tell us a little bit about what you learned there, their technology and so on. EmSense I should add is not in business at the moment, but I thought that their approach was really interesting.

Elissa Moses:     EmSense was an extremely exciting time because it was when the neuroscience applications to the market research industry was just getting started and it enabled me to jump in and be a pioneer in that regard. When I was at EmSense, we saw a lot of firsts. We saw a lot of firsts with respect to marrying EEG with mobile eye tracking in a grocery store. Nobody had ever done that before. We were the first to create the partnership model between big market research companies like Millward Brown and PRS with neuromarketing firms and being able to integrate the tools.

From day one, we were able to put a stake in the ground and say that unlike some of our competitors at the time, that we really believed in an integrated approach. I insisted with every study with the Chief Analytics Officer that we have questions with every study; that we weren’t just looking at biometric or EEG response in the absolute, but that we wanted to look at it in the context of how people consciously felt.

I think that we laid a very good foundation there that a lot of the industry built itself on. We established a lot of firsts and a lot of precedents. It was quite a wonderful time for pure exploration and discovery.

Roger Dooley:    Is anybody trying to scale EEG at this point?

Elissa Moses:     Again, scale is a funny word. I think EEG is still a wonderful tool. It’s a little bit more labor intensive than some of the other tools. When I talk about scalability, there are wonderful tools like facial coding, like implicit, that you can embed in a survey and do anywhere in the world cheaply. It enables … I feel like studies should be entitled to have both a System 1 and a System 2 set of metrics. There should be two lenses to every important study.

Roger Dooley:    Elissa, just a moment of clarification. By System 1 and System 2, you’re talking about Daniel Kahneman’s split between non-conscious and conscious more or less?

Elissa Moses:     I am. Those phrases have become very common handles for people in the industry to refer to conscious and nonconscious measurements, System 1 mean nonconscious and System 2 meaning conscious measurement.

Roger Dooley:    I’ll interject for our listeners, if you want to dig into that a little bit more deeply, Daniel Kahneman, who is a Nobel Prize winner, wrote an excellent book a few years ago called Thinking Fast and Slow. We’ll link to that too. It’s really a great resource in understanding how people think, how people make decisions. It’s not a business book per se, but there’s a lot of actionable information in there for people in business trying to understand how their customers are deciding things. I’m sorry, continue Elissa.

Elissa Moses:     That was pretty much it on that.

Roger Dooley:    Okay. Let me change things up here a little bit as we’re kind of drawing to a close. Fifteen years ago or so when I assume you must have been a child, you wrote a book titled The $100 Billion Allowance, How To Get Your Share of the Global Teen Market. Maybe it was your first-hand perspective. At the point, the internet was just starting to get going and the .com boom, that was obviously a lot more imagined than reality. You didn’t have Facebook or Snapchat or WhatsApp or any of these things that are constantly used by teens. How do you think that market has changed since then?

Elissa Moses:     That’s such a great question. I think that you’ve put your finger on it; on how our world has changed, not just for that market. Young people tend to be early adapters of technology when they’re allowed to be and when they can afford to be. They’re certainly first in line to use it.

I’ll tell you an interesting anecdote. When I did the study … And you’re right, it was before that technology was developed. I used the DMB&B Ad Agency. It was a global agency that’s now primarily Publicis, because they were bought by Publicis. We had offices all over the world. I had offices in forty-four countries. We identified schools according to our sample specs and my colleagues around the world delivered questionnaires for the students to fill out and then collected them.

When I looked at the results, you know how every researcher has that gut feeling when something looks wrong?

Roger Dooley:    Uh-huh (affirmative).

Elissa Moses:     I looked at the data and I saw the use of computers was higher in Peru than it was in Japan. I said something is really dramatically wrong with my data. I dug into it and what I discovered at this point in time, a little piece of history, that the school system in Lima where we were testing had eagerly embraced computers as a way of elevating this third-world economy into a much better economy and that that was starting to happen. All around the world people were realizing that the internet was going to become the great leveler. You could be in a town where there wasn’t a great library and it didn’t matter. You had access to all the information all over the world.

At that point in time I predicted that education would change in terms of being able to take courses where you can hear not just the best professor at your university, but the best professor in the world. You might hear Daniel Kahneman talk about neuroscience or David Eagleman, or be able to really hear the experts. There wasn’t going to be quite the class structure anymore on knowledge.

Japan, at that point in time, was slower to adopt … Even though you think of Japan a being technologically advanced, at that point in time it was wary of computers in the classroom. It felt that education should be more the old-fashion way, and also the actual language itself was making it a little bit harder with the computer and they hadn’t quite worked that out as well. As a result, high school students at that point in time had less access to computers. It’s just a very interesting thing to learn.

Roger Dooley:    Of course these days everybody has a mobile phone so …

Elissa Moses:     Right. What’s changed is in terms of how teens are adapting to culture, but I believe that what still holds from this study is that there are definitely segments within youth culture. At the time everybody thought that a teen was a teen and they were making huge mistakes.

There’s definitely values that drive culture within a country. Two countries can have the same borders but be wildly different. In Asia, for instance, we saw tremendous swings. Japan was more like Europe. It was more hip and forward thinking and fashionable, and other countries were much more traditional in terms of how teens were expected to act and behave.

We’ve really seen fulfillment of the trends that we predicted about the global platforms for culture in terms of culture becoming much more focused. Everybody can see movies at the same time, hear music at the same time, and culture has become a global stage essentially, so there’s local, and if it’s successful it bubbles up to global.

Roger Dooley:    Interesting. We’re almost out of time here, Elissa. Briefly where do you think neuromarketing is headed next? What’s the next chapter?

Elissa Moses:     I think the next chapter … I feel that one of my predictions is coming true, which makes me so excited to see over about a ten year period, which is that research is not going to necessarily be designed with the question of should it be neuro or not. Research is going to be designed to be the best possible design to meet the objectives and it’s going to become somewhat standard to have these nonconscious System 1 type measurements included.

That is happening already. ASI does copy testing. It’s putting facial coding in a standard test. We’re starting to put more implicit into a lot of our studies. We’re starting to just think about these as measures, as metrics that are part of our arsenal and putting the right combination of metrics together, so that’s going to become increasingly true.

Then I think that we’re going to have a huge jump in our knowledge over the next couple of years. I’m doing R&D, trying to look and understand effects of things. I’m doing a big study on advertising now with various types of neurometrics to understand the power of sound versus the power of visuals in advertising. We’re going to start having a lot more meta-analyses and understanding aspects of marketing that we didn’t understand before because we didn’t have these tools before.

There’s going to be a lot more studies like the one that you mentioned at Temple to come full circle, where I think that we’re going to get a much better handle on the role that each of these tools really does contribute to being able to be predictive and be able to define that and use it and leverage it in a much bigger way.

Roger Dooley:    Great. Exciting times ahead. As a reminder, we’re speaking with Elissa Moses, head of the global Ipsos Neuro and Behavioral Science Center. We’ll have links to Elissa’s website and all of the resources we discussed today along with a text version of our chat on the show notes page at Rogerdooley.com. Just point your browser at Rogerdooley.com/podcast.

Elissa, how can our listeners find you online?

Elissa Moses:     They can reach me at Elissa, E-L-I-S-S-A, .moses, M-O-S-E-S, @Ipsos, I-P-S-O-S, .com.

Roger Dooley:    Great. Thanks very much for joining us today, Elissa.

Elissa Moses:     My pleasure. Thank you.
Thank you for joining me for this episode of the Brainfluence Podcast. To continue the discussion and to find your own path to brainy success, please visit us at RogerDooley.com.