Does neuromarketing analysis really work?

Despite some promise, neuromarketing research has a long way to go before the results live up to the hype. Dr. Andree K. Bates reports.

By Dr Andree Bates on Feb 23, 2010

At a dinner in Tokyo recently, someone was talking to me about neuromarketing and how a salesman from a company utilizing this approach told him that this could not only measure impact of advertising but could then also demonstrate the ROI and impact of the advertising, and make budget decisions around this. This fascinated me for a number of reasons: My interest in analytics and, therefore, whether this would be useful for Eularis to use in our analyses, and my general passion for neurology as I did my PhD in neurology and utilized ERPs, EEGs, MRI and PET scans to research aspects around neurological processing.

In researching the neuromarketing being discussed by various market research and advertising analysis agencies, I was interested to find it utilized for the same types of things I did in my PhD; i.e. ERP/EEG and MRI and galvanic skin response (although no neuromarketing analysis I have seen incorporated PET into the analysis).

Stimuli (pictures, sounds, tastes, etc.) create a reaction in our brain that is essentially firing off electrical impulses, and the patterns of these determine our thoughts, actions and responses. These electrical impulses can be measured in a variety of ways, as I will describe below.

ERP and EEG

As a way to explain it to those unaware, in ERP (Event Related Potentials) and EEG (Electroencephalogram) analysis, an electrode cap is placed on the subjects scalp and she/he is given stimulus to respond to; then the brain waves generated are recorded for analysis. Usually these have 64 electrodes in them. Also, one usually incorporates eye tracking software to know what the subject is responding to. In my day, we had to ask the subjects to refrain from blinking as this interfered with the recording activity. I still remember Simon, one of my subjects - he was very compliant and did his best not to blink but the stress from it all made him faint from the pressure. And if you are reading this - thanks for the effort, Simon! It was appreciated the cleanest recording I got.

MRI

In MRI (Magnetic Resonance Imaging) one takes a picture of the brain itself a little like an X-ray or CTI scan, but more complex. You can detect an increase in oxygen levels in different parts of the brain in response to certain activities in different parts of the brain. The brain essentially does many things in parallel, so the MRI collects data from across many areas of the brain.

In neuromarketing analysis, the companies use these ERP/EEG skull caps to measure brain activity in order to understand the response the subject has to advertising materials, Websites and other promotional activities. They also do galvanic skin response (GSR - similar to that used in lie detector tests) and MRI as described previously. This is done by the subject being fitted with the various equipment to measure responses, and then being fitted with goggles which will deliver the stimuli to be tested.

This method is being touted as the best way to move beyond focus groups and more traditional market research methodologies to get more insight into the reactions to materials, the effectiveness of the materials and, in some cases, the strength of the brand itself. Most of the techniques on the market today focus on what the subject is paying attention to, which elements are generating an emotional response, and memory retention. From this, the process decides on the effectiveness of the material from the level of engagement on the various factors, and makes decisions on the persuasion or purchase intent, novelty or newness and awareness.

Getting consumers' attention

After researching this in detail, limited results data is available but some data on testing a 30-second advertisement for the Weather Channel has been published. This was aimed at testing which components of the advertisement grabbed the subjects attention and produced a positive emotional response. This helped the advertising agency create an advertisement that should grab attention and appeal to the target audience.

The good thing about this is that the tester can actually see the emotional response in the brain - so if someone says they like something and they dont, you would be able to see this. Also, the reverse is true. Therefore, more insight could be offered into the reactions to materials than by simply asking someone. However, no straight line was able to be drawn to link the advertisement with the ratings for the show (the real ROI), so the actual impact the changes made was unable to be measured.

To the point of using this to demonstrate the ROI and impact of the advertising, and to make decisions around this, unfortunately (from the exhaustive research I have now done on this topic) this does not seem to be the answer. No one has successfully been able to link the two in any real life case studies to date. The companies have created lovely virtual case studies, but not real life impact case studies.

Marketing ROI

This reminds me of when I attended a conference in 2003 and one company demonstrated a case study for their forecasting tool which showed how, over their 10 year prediction, what they predicted actually happened. We were all amazed that even with such environmental changes, which the pharma industry continually faces, such accurate predictions could be made so far into the future.

However, one audience member asked, What brand was this case study based on? The muffled answer was, Brand? Well, actually we have never done it on a real brand yet this is a virtual case study. It seems that in the field of neuromarketing and pharmaceutical marketing analysis, we are still in the realm of virtual case studies with no real life case studies demonstrating the validity of this approach.

I feel I can comment, given my PhD in this area and my current field of work, that it would not be particularly accurate in making claims around marketing ROI yet, as it suffers from many drawbacks some clinical and some practical. The clinical being that, in neurology, although using ERP, EEG and MRI can tell us about activation and emotional response, it cannot tell us whether one will actually act on that emotional response, making the tying of this to results difficult to prove.

We already know the research around Intent to Prescribe studies that show you may convince a doctor that the drug is great and they fully intend to prescribe it, but when their prescribing is monitored, they do not do what they intended. We have emotional responses to things all the time but whether we act on them is a result of many factors (both emotional and rational).

Not yet a game changer

Millwood Brown, one of the companies pioneering this research, have spent the past five years trying to determine if using this kind of approach justifies the significant costs associated with it. (They cost several thousand in testing per subject with the equipment, excluding the research fee and incentive fee for participating.) So far, Graham Page, the executive vice president of Global Solutions, which runs Millwood Browns Innovation Centre, concludes that he is not convinced that this is a game change for marketers as he says this provides minimal information on top of what the existing traditional research techniques already provide, and the claims made by many of the companies are simply overstated in his five years of research on this topic.

Theres a huge amount of smoke and mirrors from some neuromarketing companies, says Page. Some greatly overstate what they can actually measure. And the implications are simply not true.

This kind of testing cannot be brought to a doctors office. The doctor would have to come into the test facility. With the decline of face-to-face research due to time constraints on the physician where interviewers travel to the doctor, it is unlikely that many doctors would travel to the facility and subject themselves to a battery of tests for the sake of market research. I polled my physician friends on this topic and all bar none stated that they would submit to an MRI for a medical diagnosis but not to test a Pharmaceutical advertisement or anything else.

An expensive form of research

Although the techniques are scientific, the interpretation I am seeing and the conclusions being drawn simply are not backed up by those in neurological academia and research. I have been told by my ex-colleagues in neurology research that the conclusions being drawn are entirely subjective and neurology is not yet at a state where an emotional state reaction from this can be determined. The claims being made are grossly exaggerated and not supported by the academic community in this area.

The costs to do this kind of research would add thousands per subject to the market research cost. In this current climate of cost cutting in pharma and in pharma market research and analysis, is any pharma company that cannot fully justify its expenses and if they fully research the results of this approach, they would not be able to able to increase market research spend significantly to accommodate this kind of research? I suspect not.

Although another firm involved in this research cannot substantiate its claims of tying this kind of approach to more results driven marketing, the founder argues, If nothing else, this will get marketers to rethink their marketing investments, as just bringing in a new technology based on science shakes things up a bit, even if we cant tell everything about what works. Yes if the costs are so much higher, they certainly will.

A long way to go

I am by no means writing off these techniques as a lost cause. The research I have done with the neurological community just shows that it is not as advanced as some fast-talking sales people will have you believe. It can tell you what the emotional response to advertising is, but when extrapolating to ROI on marketing investments, they have quite a long way to go to be able to scratch the surface on that one because emotional response does not tie in a straight line to actions.

However, as these techniques mature, and the field of neurology moves on, it is conceivable that eventually these techniques will improve and be able to reveal more. However, as of now, it is simply not the case. This kind of research is interesting and an important evolutionary step in customer perception analysis, but still has some way to go before living up to some of the claims being touted by their sales teams!

For more information on this topic, please contact Dr. Andree K. Bates of Eularis.