A Surefire Method to Relevant and Useful Insights

If you went ahead and opened this, I bet you expected to learn about a new application of AI to automatically create amazing insights.


Sorry to disappoint you. This is a totally non-tech method. Besides, I believe AI or any other tech tool will never replace people. I stand firm on the belief that people are integral to discovering and articulating insights. People make great cheeseburgers. And people make great insights. (And if you don’t get the parallel between great insights and great cheeseburgers, you obviously have never had a great cheeseburger. I can help you with that.)


There are a number of ways to improve the insights you get from research. I want to focus on one that I have found to make a huge difference.


Before I get into it, defining what I mean by relevant and useful insights. There are all sorts of insights. Some insights are interesting but somewhat academic. Some insights directly impact the business decisions you face. It’s the latter, insights that impact decisions, that I consider relevant and useful.


The key to discovering relevant and useful insights is to focus the research from the start on the business decision(s) at hand.


That's one way to make business decisions. Not highly recommended nor effective, but it is efficient. Photo by Marcus Wallis on Unsplash

Yes, I hear you. That is sooo obvious.


My experience tells me otherwise. Since my rookie season at Hewlett-Packard many years ago, I would say well over half the research requests that have come to me are cloaked in the method someone thinks is best for their needs or seeking learnings in the hopes that the consumer will actually do the requestor’s job for them. The former have helped methods turn into verbs--let’s focus group this out, we need to conjoint this through…


Seldom do requests begin with the decisions at hand that are being made.


While that may sound obvious and easy to do, it is one of those things that is much more difficult to do than to talk about.


Often research studies are begun with a vague idea of what needs to be learned, and a vague idea of how it might affect the business. Somewhere inside the messy jumble of sought after learnings are decisions that need to be made by someone in the business in order to move forward. Unless those decisions are brought into the light, the results will resemble the objectives and be a jumble of insights that, while interesting, are neither relevant nor useful.


Sometimes in research it is easy to nail down the decisions. Pricing comes to mind as an obvious candidate. Or which ad campaign to run with.


One of the culprits towards the movement away from the decisions is that often the person requesting the research is several steps away from the decision-maker. Also, with research, once an organization decides to do research, several stakeholders will all want to get in on the action and thus dilute the original purpose.


The end result is often a discussion guide or survey that way too long and meanders like a Hobbit. Start with a mess, and the outputs will be a mess. And that makes coming up with meaningful, relevant insights from a mess can get messy and difficult.


So how do we get back to the decisions to give us the focus we need?


One way is to use the scientific method. We all learn about this in school, but forget about it once faced with business problems. We seem to relegate the method entirely to the sciences, when it is applicable across a much broader spectrum of fields. (Just ask Billy Beane about this.)


The steps in the scientific method are simple enough:

  1. Ask a question

  2. Do some background research

  3. Construct a hypothesis

  4. Test

  5. Analyze the data and draw conclusions

  6. Communicate the results

Apply that to your business issues and it might look something like this:

  1. Face a business problem -- i.e., have a question about the business you need answered

  2. Do some background research - get thoughts from colleagues, stakeholders, look at the competition

  3. Construct a hypothesis-- “Based on what I have learned so far, I think the answer is to do X.”

  4. Test (this means do the research)

  5. Analyze the data and draw conclusions

  6. Communicate the results


Just because you are using the scientific method, that doesn’t mean you have to conduct quantitative research to get your answer. Look at the litany of behavioral economics experiments. They use a qualitative sample frame to test hypotheses all the time.


All you are doing is helping yourself narrow down to and focus on the real questions on hand--the decisions you need to make.


There are other ways to focus in on the business decision, but I have found using the scientific method is the one that works best for me because it provides a proven process and guardrails.


The important thing is focusing on the business decision(s). With that focus and that perspective you will design research that is more relevant and useful and thus, the insights that spring from the research will also be relevant and useful. I highly recommend this.




PS: As I was writing this, I was listening the 1974 Raspberries album Starting Over. In my opinion, a vastly underrated group and album specifically--showcasing an Eric Carmen before he got all mellowed out. I highly recommend it if you are a fan of 70s bands like the Small Faces or Badfinger. And the hairstyles on the album cover are iconic.