Does This Market Research Make Me Look Fat?

Does This Market Research Make Me Look Fat?

The following article is one I wrote about three years ago, when I was focused on the complications of market research.

One of the things I had taken notice of was that "the biggest problem with research is that people draw conclusions from research". That is to say, people draw conclusions based on a linear set of findings. Answers to their questions, in spite of any questions they failed to ask.

When gathering info the in the form of surveys like for elections or marketing research, you find that the way people answer questions has a lot to do with when and who asked the question, and in what setting. And you're only asking people who show up to the focus group meeting, subjecting the researchers to survivor bias, and confirmation bias.

I'll present the article as it was sent via my email newsletter in 2018.


Hi Friend,

As you develop your communications platform or marketing campaigns, should you spend any time on research or should you skip it and go with your gut?

I would consider this the area of marketing that is the most under-invested in by one half of the market, and over-invested in by the other half with very little in the middle. To be clear, I did not do any formal research for this email, so that is an anecdotal conclusion, but since I am an expert, I feel pretty confident in it (note that tongue is in cheek).

Confidence is the problem though. Most marketers say they know their customers, they know what they like and want, and they know that their product sells itself. However, this is almost never true.

When marketers have too much confidence and skip research, they hurt themselves. While most brand managers and product marketers know their customers better than any agency will, they still have blind spots because they are operating on lagging indicators. What I mean is they know insight about the people that are already customers after they become customers. Asking your customers if (and why) they like your product is like asking a spouse if an outfit makes you look fat. It’s not a good route toward getting the truth (you always look great by the way). Even if you are interested in expanding market share inside an industry, knowing why someone may choose NOT to buy your product could give you insight into how to market better. Too much confidence in your knowledge of the market can make you blind to the roadblocks. This could be a good opportunity to invest in deeper market research.

The rest of the world is spending too much time on research, and not enough on executing. To be fair to the over-confident crowd, one of the ways to know more about your market is to sell stuff to them. Seeing what promotions work and which don’t. Which products take off and which ones result in anemic sales? Which marketing statement resonates and what falls flat? Marketing itself is market research, and it happens to generate revenue in most cases too. Some organizations spend so much time trying to validate their message in a lab, when real life results are always more nuanced. When someone knows they are responding to a survey or are in a focus group session, they respond to questions within that context. Buying a pair of shoes while scrolling through Instagram doesn’t require the same kind of analysis as rationalizing an answer to a researcher's pointed questions. This might be an opportunity to pause your research and go ahead and execute to get some real data.

The biggest problem with research is that people draw conclusions from the research. It’s good to base your strategy on data, but the data is biased, and so is the interpreter. When you ask a customer “why do you like our product better than the other brand?” you get a different answer than if you ask “why do you like the other brand’s product less than ours?”. Most research is leading, and all research is built with prejudice. We’re fooling ourselves if we think that our research is not biased (including independent third party research). The truth is, we can validate anything we want if we are clever enough. We’re always better off knowing that research is necessary, but it can only go so far.

To be clear, my recommendation is to know the risks of either approach and develop a plan that fits the occasion.