PureSpectrum - Schedule A Demo
Qualtrics: Here to Help

Just Tell Me What You Want

Big data and surveys can be supplemented with additive processes to better understand your customers’ appeal factors and ultimately increase their motivation to your brand. Bridging the gap between perception and behavior really uncovers why customers make decisions.



By Luke Heffron

If big data isn’t the key to understanding what our customers’ want then why don’t we just ask them?

Because people can’t tell you what they want, but they know it when they see it, read it, taste it or hear it.

Traditional surveys focus on cognitive based questions and their resultant data, or in other words the rational, politically correct mind. Our challenge is that real purchase decisions are based on how our customers feel about a particular service, brand, or offer. The harsh reality facing us all (who are trying to sell) is that feeling is not something that can be intellectualized and therefore is nearly impossible to discern through a simple set of biased questions.

The explosive growth of Internet based research surveys that deliver fast results at very reasonable cost doesn’t mean we’re any smarter about our customers. We’ve just gotten more efficient at asking more questions to more people.

No doubt surveys provide a value. They can help, identify what someone likes or dislikes about your brand or service, and to what degree. Where it falls short is a profound understanding why you are turned on or off. It’s great to learn that they like this particular thing, but to ring a sale you need to know why they like it.

How someone perceives something and how the ultimately react to that is the basis for marketing (stimulus / response).

Perception is driven by everything around us including massive input from marketing messages, advertising, news and entertainment media and a slew of other sources. This perception influences what we “think” we like, or “why” we like it – but it often doesn’t accurately correlate to behavior. So as marketers, we listen to what our customers say they like about us, we react to that, and yet the results don’t reflect the data. Why?

What’s The Ideal Chocolate Bar Mix?

One of the big chocolate giants wanted to know specifically what makes a chocolate bar taste wonderful. Everybody in the business talked about a “rich chocolaty flavor experience”.

It seemed so natural to ask the consumer what would make the bar better. So the obvious solution is to conduct a survey and ask the consumer what they think would make the bar better. Not surprisingly, their response was “add more chocolate”. So, naturally more chocolate was added to the bar but regardless of how much more was added, test panels didn’t like what they tasted

Rather than follow what the consumer said, they went to food science expert, Dr. Howard Moskowitz for another viewpoint. He created multiple combinations of sugar and fat in the candy bar, then taste tested these with consumers, and watched what they did. The result of this study wasn’t more chocolate, the majority of everyone preferred an increased combination of the sugar and fat.

Dr. Moskowitz’ thoughts on this:

“Don’t ask consumers what to do, instead put something in front of them in terms of an experiment – that will tell you what to do, every time”.

We think and respond with our entire body, Visual, auditory, sensory and emotional stimuli all play a role in how we feel about a product, experience or brand.

The appeal of the candy bar is not just about taste either: it’s about how it melts in our mouth, how it smells, how the paper around it looks and how it unwraps, how the chocolate breaks off. There is no single magic bullet to appeal.

In my upcoming blogs we’ll look into how big data and surveys can be supplemented with additive processes to better understand your customers’ appeal factors and ultimately to increase their motivation to your brand.

We’ll begin to bridge that gap between perception and behavior and really uncover why customers make decisions.

Please share...

2 responses to “Just Tell Me What You Want

  1. Why do we need to know why people buy when big data and automated marketing just responds and learns to predict what to do next on average to get the sale?

    No human insight needed 😉

    Customers and their behaviors simply become resources to be harvested like ants on a hill 😉

  2. I think it’s a mistake to try to do everything in one go, especially in New Product Development. Research chemists at manufacturers for decades have back engineered recipes for competitor products and statisticians have related recipes to sales/share. Qual and small sample sensory testing have also been used for decades and conjoint/DCM also has a few grey hairs. Just asking consumers directly what they want can sometimes provide some direction but relying on this exclusively has always been considered research malpractices, hasn’t it?

Join the conversation