Where are market research and analytics falling short?
Through a special arrangement, presented here for discussion is an excerpt of a current article from the Joel Rubinson on Marketing Research Consulting blog.
For 10 years or more, big research providers have been forced to accept meager organic growth. Yet, growth is out there.
Companies like Marketing Evolution (media optimization, surveys linked to digital and ad serving data), Disqo (first party data access to online “walled garden” shopper, search and ad serving data) and Ace Metrix (automated and syndicated ratings of video commercials) are addressing emerging gaps. While retail moves to omnichannel, advertising is also transforming to people-based ad serving — personalized and selective — driven by data at scale.
The new media challenge is to find the right customer, whatever website or app they happen to be on, at exactly the right moment. Buying whole audiences is wasteful; that is why (addressable) digital advertising has now passed linear TV for ad revenues. Yet, walled gardens and GDPR prevent access to what marketers need to know, so serious gaps and challenges exist.
If you are in the research, insights and analytics business, here is your growth checklist. Do you offer:
- Studies at large scale for the cost of small-scale studies of yesteryear;
- Linkage to digital ad serving and profiling data at scale;
- Technology to harvest the exhaust of digital behaviors at little cost;
- Services addressing the need for media optimization in a digital age;
- Technology that supports omnichannel retailing and integrated marketing;
- Technology that provides a disruptive cost model via survey automation and self-serve access.
Where are marketing research and analytics falling short? I bet that 90 percent of sales of the large research houses fail to check off two or more items on this checklist. That is why they have little organic growth.
The acid test for no growth research is this: If your research could have been designed the same way 15 years ago, it is a no-growth business.
Market researchers on the provider or client side — you must align to a digital future and tell marketers something important they do not know. You must light a new path forward into the addressable age or risk becoming redundant.
Discussion Questions
DISCUSSION QUESTIONS: Where do you see market research falling short in providing insights into digital behaviors? What disruptive technologies or approaches do you expect to see more of?
Poll
BrainTrust
Joanna Rutter
Marketing, Dor
Ryan Mathews
Founder, CEO, Black Monk Consulting
Gathering data is the easy part. If you don’t know what you’re trying to solve for or how to interpret the data you’ll never find a pragmatic, viable answer — regardless of the amount of data. While “Big Data” has been a popular buzzword for a number of years, I believe valuable answers can be found with “little data” — data that is focused around a specific workflow or process. Start with what Jim Collins calls a BHAG (big hairy audacious goal): the promise of zero out-of-stocks. Using little data enabled by integrating new technologies and transforming the workflow and business process status quo can deliver this very real challenge for retailers and brands. Having a BHAG and designing an experiment and its associated measurement campaign will provide a data-driven solution.
I work for a foot traffic analytics provider (Dor) and couldn’t agree more with your conclusion, Joel: It’s crucial for marketing researchers to report on reality. Right now, that reality is a bit of a tangled omnichannel mess. It’s less tidy and predictable, and the way research is conducted must reflect that.
Digital “behavior” is folly. Digital behavior is not a representation of what is going on in the mind of a shopper. Solving for digital behavior is like fishing the ocean for human emotion. The evidence of this monumental disconnect is clearly stated in Joel’s article. The future now is new technologies using the exact same data, in a new way, and that is to solve for human emotion which taps into individual sensory preferences. The #1 reason a human makes the decision (falls in love with a product) to purchase. The future of digital is the human element. Technology standing right before our eyes, yet most are blind to it.
digital behavior is not folly. it is the basis of what has become over half of ad spend. TV is much smaller now. How will you understand the media channel with over 50% of media spend unless you study it? And BTW, I was not only talking about digital data. I also mentioned disruptive cost models for surveys. They are not dead, but brand tracking is bloated and segmentation research is hopelessly unactionable the way most do it.
Hi Joel. Thanks so much for your comments. I thought your article was right on target, which is why I choose your article to comment on today.
You said “Market researchers on the provider or client side — you must align to a digital future and tell marketers something important they do not know. You must light a new path forward into the addressable age or risk becoming redundant”. You also said, “The acid test for no growth research is this: If your research could have been designed the same way 15 years ago, it is a no-growth business.”
I understand digital behavior has been and is the standard for ad spend. I interpreted your article as a call to action to comment on new ways to tell marketers, “something they do not know.”
At the risk of setting off a war — or perhaps because I think a war ought to be set off — let me advance a provocative hypothesis: Maybe it’s the case that market research, as we know it, is just no longer as relevant as it used to be. Why would I suggest that? Simply because lifestyles continue to change faster than the tools we have to measure them. We continue to use the same old, Industrial Age tools and force “the facts” into neat little 20th century intellectual boxes. By its very nature a good deal of market research is linear and historical, but the shopper of 2018 bears little resemblance to the consumer of say 1998. Let’s take a concrete example. Many BrainTrusters burned many brain cells describing “Gen Y.” Remember them? They were what we now call “Millennials.” Why did we need a second label for a single cohort, besides the opportunity to take a second billable bite at the demographic apple? Because cohorts — defined in 18 years increments — aren’t that useful a measuring tool anymore. If we were honest we would all concede that traditional market research tools like focus groups and Internet intercepts have outlived their usefulness, defined as an ability to yield “actionable” — sorry, couldn’t resist — insights. Big Data? Great … but not if the algorithms are filtering for old patterns, which explains why Amazon can never get my book recommendations right, among other more critical problems.
Joel’s first and third checklist items have to do with cost — critical in the real world, but not the first consideration if you are seeing quality results. And then we have an additional checklist item referring to omnichannel — great if you think consumer shopping behaviors are limited to mobile, physical and online, but not so good when you remember that mobile is just a way to access online, and that we have lots of new access points like voice activation, subscription, etc. And we also conveniently overlook the fact that we aren’t even measuring shopping patterns at farmers markets, independent stores, Etsy, eBay, etc. No, the mistake market researchers make is trying to freeze a moving target and project the image forward in time. Now let the “thumbs downs” fly.
Ryan — there are good researchers and bad researchers, those focused on practical issues and those who don’t understand that, those who think cohorts matter and those who think we researchers have better things to do with our time. Joel thinks the universe consists of one type of shopper who is digitally driven. His points are great, but only for that small group.
Stephen, thanks. As I said at the outset, I was being a bit deliberately provocative. I think we may share a common opinion that tools need to be appropriate to tasks.
Wow, Steve, no! Did you actually read the discussion starter or just the mistaken interpretations of some of the comments?
No thumbs down from me. Your last sentence is great…market researchers …try to freeze a moving target and project the image forward in time. Excellent!
Thank you Ryan for sharing your excellent insights and comments in our conversation today. I appreciate it!
Long way to say history doesn’t matter, but I’ll buy that. Partially, it just matters less. Despite the dynamic nature of customer motivations, the tools we have today are actually quite complex and in some cases predictive and prescriptive.
Despite all the data we collect, it doesn’t add up to enough needed to make a good decision about a specific individual or a point in time — individualized data. However, market research is about the aggregate. Data from yesterday can yield insights about the present — as a matter of fact, the closer to the present it is, the more relevant and accurate the data should be. There are exceptions of course.
For part 2 of the discussion topic: The ability to make human-like inferences at scale in the future through ML/AI will be an interesting disruptor — if it can actually do it.
Joel’s article is focused on one aspect of marketing research that means a lot to some research companies and their clients. There’s still some almost traditional research going on that fills the needs of many clients that want to make informed decisions as they introduce new products and try to maintain their competitive edge with effective marketing strategies. Thank goodness!
We’ve said it before and it bears repeating. Before committing to research, whether it’s a long-term initiative or a short-term project, buyers of research should know what decisions they want to make and how the data provided will help them. Research dollars well spent are those that provide results which in a timely manner inform, guide and are actually used.