When testing trumps intuition in marketing
One of the biggest challenges in our business is the tendency that many marketing executives have to, on one side, underestimate the extent to which data can be used to obtain market insights and, on the other, overestimate their ability to take good decisions based on their intuitive understanding of what will work and what won’t.
We recently came across a McKinsey Quarterly interview with Duncan Watts, principal research scientist at Yahoo! Research and author of Everything is Obvious, Once You Know the Answer, that touches upon the theme of intuition vs. data in marketing. Mr. Watts is a terrific expert with a gift for clear thinking and delivery.
Some insights we can draw from the interview:
1. Most businesses have (or can access) much more data that they could just a few years ago. New tools and approaches are needed, but the good news is that any business can start small, by harnessing data that they already have. For example, any company that relies heavily on online marketing programs to fill their sales pipeline can investigate the performance of each acquisition channel, use analysis to segment the market, uncover key decision points in the buying process or qualify leads based on users actual behavior. The science is rapidly developing and the research tools once available only to large corporations are becoming accessible to small and medium businesses.
The kinds of questions that we can ask are much more sophisticated and require a whole new science. (0'36")
2. We are wired to trust our intuition, but we do not always need to base our business and marketing decisions on it. In fact, we shouldn’t. A simple rule of thumb we use with our clients: is the cost of a bad decision bigger than the cost of the minimum test you need to run to determine if your intuitive solution is right with the accuracy you need? Any time the answer is yes, you should test your hypotheses. Data and tools make it possible to test assumptions quickly, reliably and at a relatively low cost. Mr. Watts goes on to describe a glaring example of the fallacy of intuition that is particular relevant to online marketers: word-of-mouth influence. When looking at how information and content moves through networks, you find out that true diffusive “viral” adoption is a lot less frequent that our intuition would make us believe.
Whatever hypothesis you have, whatever intuition you have, however self-evident it may seem, you test it against the data, and, bottom line, it’s wrong — like, every time. (2'35")
3. So if intuition is unlikely to give us reliable answers, it can still give us working hypotheses to test. With today’s abundance of data, some models, such as for example the top down approach of creating personas (i.e. detailed profiles of a target customer) as a way to predict a certain customer behavior, may have lost a good degree of their relevance. Others, more statistically-based, bottom-up models have become more broadly applicable. Regardless, the process of marketing has become more like an ever shifting, continuous activity of making assumptions, testing them and tweaking assumptions again. It requires marketers to be willing to put their mental models to the test and trust data over their intuition. And their egos.
Advertisers, create elaborate stories about representative consumers, and then they build a campaign around selling to this person that they’ve created in their minds. That, to me, is deeply flawed. (7'05")
