Generating Insights from Qualitative Research
Research as a discipline is a huge subject. I have a doctorate and a master’s and in both cases, I studied research methodology and undertook large-scale research projects. The type of research that I undertook then and the type of research I undertake and am involved in now as someone working in product is very different whilst maintaining some similarities. From the outset, I’d say that understanding how much time/resources you want to devote to researching depends on the type of product you’re working with. If you’re able to rapidly iterate and run experiments then you can spend less time researching but if that’s not the case then maybe spending more time researching is justified. Either way, it’s going to be different to the type of research undertaken by academics.
A standard process (not the only one!) of conducting user research may look something like this:
- Engage with users (interviews, focus groups, surveys)
- Record notes or observations (e.g. Word/Miro/Mural)
- Look for patterns (affinity mapping)
Oftentimes the research may end there. If you’re working digitally that might mean you have a Word doc/Miro/Mural with a bunch of notes/stickies grouped around some form of a common theme. The question is then what do you do next?
I’d suggest the next step is to:
- Synthesis the data and extract insights
- Put these into some form of report/tool (like Dovetail) that allows your findings to be shared with key stakeholders
Q: Why are these two aforementioned points important?
A: Oftentimes the notes from an interview or the subsequent affinity mapping may end up as the end of the process. If you want to share findings or justify to senior executives this isn’t the most optimal way. Synthesising your data, producing insights and then putting these into some form of shareable form allows you as product manager to be clear about what the research is telling you and then to pass that clarity onto others.
Research Plan
The starting point of any research is to have some form of research plan. Here’s an example that I’ve put together based on a digital asset company I used to work at:
When you’re producing your insights here’s a few things you might want to consider:
- Are there any similar themes from across your research — so for interviews are there similar ideas/problems/pain points/needs being expressed by multiple participants?
- Of these themes that you’ve identified which are the most important and why?
- How, from my research, can I evidence by insights
I’ll take each of these points individually:
Similar themes
Are there ideas/themes/concepts etc that are repeated by multiple research participants?
Are there further levels of detail/nuances etc that can help find the insight?
For example:
Initial insight: A trending theme may be that interviewees say that knowing the difference between the thousands of different digital assets is very challenging.
A furtherlevel of detail: Customers ask friends/colleagues for recommendations
It is important not to over-generalise when presenting the research findings. A small sample size may not be indicative of the entire market. This perspective needs to be balanced with the requirement to move quickly and make rapid decisions. This is where triangulating data can help through the use of other research instruments such as survey/focus group/secondary research etc.
Most important or impactful insights
This is essentially looking across all your insights and asking which are the most important ones. You could argue that to save time you could just include these insights in your report/documentation etc. The benefit of having both all the insights and then focusing on the most important one is it stops readers from asking the question ‘have you considered X?’
Evidencing Insights
I would be very judicious in evidencing findings. Again it’s worth stressing that this is not an academic research project and so what you’re looking for is strong evidence to justify the points you’re making.
Conclusion
In the final analysis, I’d say that in most cases research needs to proceed at pace and facilitate quick decisions. If resources are thin on the ground then documenting research in detail may not be the most optimal use of resources but how you proceed will depend on your own idiosyncratic context.
