Facial Imaging: The “Big-Data” Solution for Emotion Research?

September 2, 2013

We are entering an era where, thanks to technology like facial imaging, “soft-data” on emotions – traditionally the province of qualitative studies or smaller scale specialised surveys –  will become “big-data” that provides very hard results.

Facial Imaging Embeded & Automatic: nViso API in Cinemax site - 1 million views and counting

Facial Imaging Embedded & Automatic: nViso API in Cinemax site – 1 million visitors and counting

At first glance facial imaging (or “facial coding”) seems like just another variant of Neuroscience testing, but in fact it has some very different features. In earlier posts we’ve written extensively on the results obtained from this technology (e.g. see “Soft-Drinks, Soft-Sell“), but in this post I want to get across the point that the really big news is not so much how well facial imaging measures emotion, but how many people and how much emotion can be measured.  This makes it fundamentally different from hardware dependent methodologies like EEG or conventional survey based methods. Two thought experiments for market researchers might illustrate: Read the rest of this entry »


Instant Noodles in India –Tug of War or Tugging Heartstrings?

July 29, 2013

One of the fascinating things about using Facial Imaging when exposing competitive ads to the same respondents in a single category is that it provides unspoken clues about relative brand positioning issues (as FI does not rely on ‘rational’ recall or post-exposure rating).

 ToW foodles

 A brand with an inherently easy to understand positioning is more likely to get its message across clearly than a brand attempting to make consumers think about the product in a new light. Here, we see an example of an Instant Noodles brand, Horlick’s Foodles, taking on the task of provoking Indian mothers believe that instant noodles can actually be nutritious. It is done in a manner successfully demonstrated in previous posts (telling a story and building towards the main message), but the style of the execution is perhaps a little too dark. So it does not seem to reduce the high level of apprehension when the main ‘nutritious noodles’ idea is brought forward. Maggi, by contrast, takes an easily believable ‘life-saver’ message (that Maggi noodles are there and ready when you really need them) and pushes this in an engaging positive manner connected by a catchy jingle. Clever execution combines with a clear-cut positioning to evoke a strong response that builds positively.

The key lesson is if you are taking on a difficult advertising task such as proposing something that may conflict with common wisdom, the sequence and content of emotions evoked is critical.  Such ads are more likely to lose direction in emotional terms, and need more careful prior assessment. Read the rest of this entry »


Men & Women Watching Ads – Different Markets or Different Species?

June 22, 2012

The more I look at the patterns of response to advertising revealed by facial imaging the more I’m convinced that the old “Men are from Mars, Women from Venus” chestnut has validity. (Facial Imaging is the direct measurement of emotion via webcams – see www.nviso.ch if you want to know how it works).  In essence major contrasts in reaction are the norm in the ads we’ve studied and this seems to hold across cultures and is not confined to the obvious categories like cars or alcohol. Yes, MR studies have often noted gender differences, but I’m not referring merely to broad reaction to the ad but the fact that the level, pattern and type of emotional reaction to specific elements of the commercial can differ wildly.

Women (pink line) React To An Ad With Lots of Baby Shots. OK for a while, but get it wrong and you lose them.

Even when the overall  response of males and females is pretty similar (e.g. as measured by self-reported liking, impact etc.) the norm is that you’ll find that the build of emotion and reaction to specific scenes diverges markedly. While I’d be the first to admit we need to do a lot more systematic analysis to back up my initial assessments, I’m increasingly of the view that this has huge implications for advertising and research. Read the rest of this entry »


Watching and Listening – An Alternative to Direct Questioning?

April 10, 2012

In the early 90’s, the ‘tagline’ of a leading Japanese agency’s brochure was ‘the key to understanding was to read between the lines of what was not said’. Two decades later, Shobha Prasad re-visited the issue focusing on the layering and subtleties of Asian languages in ‘Listening to the Sounds of Silence’ at ESOMAR’s APAC Conference.

As Asia grows more healthy, wealthy, and wired, superficially its consumers resemble their Western counterparts. Sometimes even, having leapfrogged technological lifecycles, they appear more advanced. Nevertheless, although these new trappings bring an almost ‘stateless’ vocabulary to the world’s languages, the fundamental challenges of interpreting local cultural nuance and international comparisons remain.

Direct questioning methods is felt to be (even) less effective in Asia due to language structure, cultural norms, and social convention. Asian consumers are often not so forthcoming with opinions than Western ones – although researchers in Mumbai or Manila may well disagree. The real issue, though, is the huge variety of expression, both verbally and visually. As well as vocabulary and sentence structure, different expressions also support communication. Cultures displaying emotion less conservatively usually have the mouth as the main focus; a culture that masks its feelings focuses more subtly on the eyes when determining emotion. So, visual cues can be equally important to gauge underlying sentiments. Read On..>


The Landscape Of Emotion

February 22, 2012

Measuring emotion is increasingly straightforward – interpreting the results still requires some intellectual subtlety.

Emotion, and the research techniques that measure it, remain hot topics in market research. Many of you will have read of Brainjuicer’s Valentine’s day card to Millward Brown, celebrating the latter’s purported “embracing” of emotion as a key marketing driver. A lot of fun for those of us that are observers of course, but leaving aside the question of whether this unduly caricatures Millward-Brown’s approach to emotional analysis, I detect in the discussion, another caricature: the reduction of ‘emotion’ to something simplistic and monolithic. If only we can measure this emotion stuff, we will ‘have the answer’. Maybe, if we can find the right emotional measurement machine we researchers can all retire?

As some of you know, David and I are working (with nViso SA of Switzerland) with exactly that: an “emotional measurement machine” that directly measures people’s emotional response to stimuli via a method called 3D Facial Imaging. Here’s a chart based on 3D Facial Imaging data – I’ll explain it’s significance later in this post, for the moment just note we can directly measure specific types of emotive response with a standard computer and webcam.

Hills & Valleys in The Landscape of Emotion (See Below for Explanation)

This is, I would argue, much more accurate and granular than any questionnaire based method. Yet, despite being thrilled by the results we are obtaining, I would not argue that we have reached some sort of  “deus ex machinamoment, where researchers and subtle interpretation become redundant.

Read the rest of this entry »


Emotionally Preaching To The Converted

January 16, 2012

New Year, new start. As some of you who know us will realise, one of the reasons our blog postings have stuttered in recent months is that we’ve been far too emotional. Or at least far too involved in telling clients, MR and ad agencies about why emotional marketing matters, and why it’s not quite what they thought it was.

Preaching To The Converted: More Useful Than You Think?

We thought therefore, we’d start 2012 with a series of posts on what we think is the most important development in modern market research: our increasingly accurate ability to tap into consumer emotions.

In particular, we want to do our bit to move discussion of emotion measurement from methods and applications towards the more important area of marketing implications. Why measuring emotion accurately really matters.
Read the rest of this entry »


Simple, Scalable and in Shanghai: The future of research?

September 22, 2011

Allow me to blow my trumpet a little: David and I recently presented at the AMSRS conference in Sydney on automated facial imaging – the content must have been worthy, as it earned the ESOMAR-sponsored “Best Presented Paper” Award. But, truth be told, we felt the driver of the award was probably people’s excitement at seeing how much detailed information on emotional response to marketing stimuli can be delivered by a system that just ‘watches human faces over a webcam’. This is illustrated below:

Facial Imaging: From Faces to Reports, No Questions Asked!

The appeal of such systems also came up in a discussion I had recently with a senior colleague that was spurred by news of events at EmSense:  http://www.neurosciencemarketing.com/blog/articles/r-i-p-emsense.htm. While things may yet turn out for the best, it did seem to us that selling a system based on sophisticated hardware to US customers, in these tough times, cannot have been easy. As we tossed around the issues, it seemed apparent that as clients become ever more cost-focused and have to deal with massive amounts of data from multiple sources they become increasingly obsessed with research services that are both scalable and simple to implement and interpret.

Read the rest of this entry »