5 Ways To Make Sure Your Insights Are Real

I use research all of the time - usually research that other people have conducted but sometimes research we've conducted. All our strategies at 36ns are informed by the way that real people think, behave and react. That's how we help our clients. But as someone who often uses research done by others, I have 5 pointers. Ok, really, you can call them requests. Or just picture me making a begging face.


1. There is no such thing as a consumer. Nor are there voters: If you are conducting primary research - whether an online survey or a focus group or really anything in between - please keep this in mind. You are asking questions of real people. And they don't think of themselves as simply the sum of their consumer or political choices. They are complicated human beings who wear different hats everyday: consumer, voter, information gatherer, family member, employee, etc. Please treat them as such.

2. Use the right language for your audience: Often, our clients get stuck in their own bubbles and (bless them) can only really communicate in a stream of jargon. Their brains have been re-wired by it, and we often adopt their terms to communicate with them. And across all of our clients, we develop what seems like an endless clutter of business speak. Real people don't speak like that. They don't say brand. They don't think in the terms that the clients do or that you do. Get back to basics. Use the right language for the audience you are engaging with in your research, or they will check out and give you unconsidered answers. When in doubt? Pilot it. Literally take the time to sit with a stranger and ask them your questions. When they make a face, you know you're doing the wrong thing.

3. Sometimes, it helps to start from scratch: We've all been there, on a tight deadline with a bunch of different projects on the go that need attention. So when it comes time to write a focus group discussion guide or a survey, you start cutting and pasting. Sometimes it's not even that you're our of time, it's that you know that a series of questions worked really well before - so why re-invent the wheel? Well, here's why: language changes. All of the time. So does the context in which our research takes place. Before the 2016 US election, I wrote about how I wasn't convinced that the polling was right because the questions themselves hadn't changed. Pollsters were using the same language in a contest between two hated celebrities that they've used for the last many elections. Sometimes, the best way you can help your clients is to approach their research question from a blank sheet of paper. It changes the way you approach the design of research. And that's what the client needs: research imagined just for them, just for the problem they have right then.

4. Do you really need to ask that: We have a problem with overtaxed panels and the real people taking part in our research are getting bored of answering the same old questions in the same old way. So here's my challenge to you: do you really need to ask those same old tracking questions? Is there another way your client could that data? How about helping your client to gather up data from across their business - web traffic, ad spend, sales, etc - and analyse it to see what correlates with what in their tracker. I'd bet that of the series of 20 questions you keep asking, only 3-4 of them actually correlate with anything in their business. So stop asking the questions that don't matter. Re-purpose that bit of the tracker to test new ideas or see if your clients' perceived competitors are actually the right ones.

5. Consider that bias might not be a bad thing: Our clients do not exist in a vacuum. They won't be selling, servicing or communicating in a vacuum either. So why do we test products, packaging and communications in an objective setting? Sometimes, the best way to test a message or a competitive position is to rile people up and then see that settles them down. The perfect line in a vacuum isn't going to work in the real world. 

By reconsidering how you approach research - when quantitative or qualitative research - you can be sure that the insights you generate are real, new, and truly relevant to your client and the problem they have right now. And if your research isn't geared towards solving a problem your client has (or may have in the future), the real question to ask is: why are you doing it? Isn't there something more meaningful that you could be doing for your client?

[Originally published on LinkedIn]

What is "New School Planning"?

Why marketers may be getting consideration sets completely wrong...