I’ve recently completed a piece of research for a client where I used a ‘learn as you go’ approach

We were exploring new behaviour (because of a change in the law), so we were entering uncharted territory. We also had a small number of list respondents to recruit from. It was one of those projects where we didn’t know what we didn’t know.

We opted for a staggered approach, which meant right from the off we had built in time to change course if the project required it.

We (myself and Streetmate Stephanie Rowley) started the project with a number of shorter telephone ‘pre’ interviews in order to test the water. This was really useful because we were able to get a feel for what the key topics emerging from the project would be, we could change our recruitment criteria for the next batch of interviews coming, and we cherry-picked the best respondents from this telephone stage for a more in-depth stage of in-home, filmed interviews. (Which meant we already knew we’d be getting great stories for our debrief before we stepped across the threshold.)

This all worked really well. My client was so adaptable that as we neared the end of the fieldwork, she agreed to us doing a small piece of quant research to confirm that the respondents we were speaking too were typical of the market and not extraordinary because of self-selection. (In this case people were much more tuned in and sensible than we were expecting them to be, which made us doubt their normality!)

If you have the luxury of time on a project, I recommend this learn as you go approach. It also means that the research design can be nimble – you can actually stop when you feel you have got enough insight rather than designing a bloated project, just to be on the safe side.

Learn as you go only works when you’ve got great clients who trust you and want to work in partnership. You have to be able to build in time for fieldwork that might, or might not happen too. However, for the quality of the project, I think it’s the way to go!