Market Research is Dirty. Avoid it or Fix it?
Listen to the Audio Version here
Market research has a bad reputation. And for good reason. It produces data that is like dirt in a gold mine. If you just take it as it is, you are often f... ahem, misguided. Here is how market research can give you unique insights that you cannot find anywhere else. I also discuss ways to separate the wheat from the chaff, the dirt from the gold.
This spring, I gave a keynote at a CMO peer network event. During the break, I met Jesko, the CMO of a major European confectionery brand. He said: "Frank, we don't do market research and we're growing strongly." During the conversation with him and some other marketing leaders, I learned a big lesson: Market research has a bad reputation. It is hardly seen as a strategic insight tool. When it is used, it is often for political reasons.
“Market research has a bad reputation. For good reasons.”
Why is market research not trusted?
"If I had asked consumers what they wanted, they would have said: faster horses" This is what marketers always say when asked about the limitations of market research.
It is one of many valid concerns: Simply asking a question and expecting the answer to be self-evident is naive. Insights professionals know this. This is why it takes more than a SurveyMonkey login to get useful insights.
“It takes more than a login to SurveyMonkey to get useful insights.“
There are other valid concerns that undermine trust in market research:
Reproducibility: Consumer responses often feel fabricated. If you have ever filled out a questionnaire yourself, you know they are. The Ehrenberg-Bass Institute at the University of South Australia had consumers fill out the same Brand Tracker questionnaire twice and found that only about 50% of respondents gave the same answer.
Not very specific: what do you do with "Quality is important"? What exactly does "quality" mean to the customer? What exactly does "important" mean in terms of buying behavior?
Representativeness & sample size: Most marketers do not understand the difference between representativeness and significance (spoiler: it is like water and fire). However, they are still uncomfortable with the idea that insights from a few, somehow selected people are representative of the entire market.
Interpretation is subjective: As marketers learn that research results can be interpreted in many ways, the discipline loses its appeal of uncovering objective truth.
Say-Do Gap: When consumers say they would buy a product, it does not mean they will. There are many reasons that can interfere with an initial intention. This gap can be very large and fosters distrust in market research.
Obviously, market research is far from perfect. What can we do about it?
Please choose:
Avoid market research because it is not reliable. Instead, try to get the insights you need from real customer data, such as sales data or web tracking information.
Fix market research
Avoid or fix market research?
My conversation with Jesko at this CMO event continued. He said, "Frank, your presentation made me realize that with the use of Causal AI, we CAN now use market research in a reliable way."
For me, his realization was a new way of looking at it. I did not know that market research had such a bad reputation. In the market research bubble, everyone confirms that it's the most important thing to do. I did not realize how much CMOs needed a solution.
How else (besides market research) are you going to understand how people see your brand, why customers choose a certain brand, what product features are missing or need to be communicated properly?
We "just" need to fix market research. We can fix it by using Causal AI and implicit measures.
“We can fix market research by using Causal AI and Implicit Measurements.”
Implicit measurement is measuring customer associations based on gut instinct (System 1) and eliminating rationalized responses (System 2). It does this by interpreting the respondent's reaction time. We know from neuroscience that valid responses faster than 0.3 seconds are impossible and that simple responses that take more than a second or two are the result of a thought process rather than an intuitive response.
Many aspects from willingness to pay to brand aspects are much better measured implicitly. It dramatically reduces the say-do gap.
Causal AI solves many of the other concerns about market research. Reproducibility issues are just noise that the AI model cancels out. Representivity issues can be controlled by including variables in the Causal AI model that define representativeness.
But the biggest concern CMOs should have about market research is rarely addressed: It only measures data and correlations. It does not focus on the underlying cause-and-effect relationships. This is exactly what Causal AI focuses on.
(You can read all about Causal AI in my latest book, Think Causal Not Casual. It is written for marketers.)
Companies are starving for insights. Let’s filtrate the data.
Market research is like salt water. We can not drink it right away. We first need to de-salt it.
“Desalt market research data and you can drink it.”
Supra.tools for instance is a new platform that provides standardized research tools by leveraging Causal AI and implicit measurements. These solutions are available as a managed service that includes consulting:
Strategy AI: Understand why potential category buyers really buy. Based on a special questionnaire, Causal AI distills the market's cause-and-effect relationships. This enables you to build powerful marketing strategies.
Product Optimizer: measures how much each product feature can increase willingness to buy and willingness to pay. It is an alternative to conjoint measurement and eliminates its drawbacks.
Price Optimizer: measures the price-demand curve to determine a profit-maximizing price. It is highly scalable and ideal for consumer products.
Brand Optimizer: measures how much brand archetype perceptions increase willingness to buy and willingness to pay in your domain.
Touchpoint Optimizer: measures how much each perceived touchpoint increases willingness to buy and willingness to pay.
CX AI: A standardized analytics service that leverages existing customer feedback data. It categorizes open-ended qualitative text feedback and then reveals the impact of each customer issue on NPS or CSAT.
Coming soon: Supra Mix AI and Supra Churn AI.
“Supra.Tools is a new platform that uses this cutting tech that “fixes” market research.”
Yes, market research has many problems. But instead of "starving", we can solve them somehow. Causal AI and implicit research solve many of the problems. With supra.tools, companies can access standardized solutions as a managed service. They get the latest insights at reasonable and predictable budgets with the security of expert advice.
You don't do much market research these days? Or you do, but you rely on traditional methods? Step up and use Causal AI and Implicit Research.
This is how you 10x insights.