Mastering user insights - From good to great
Over the years, I've discovered there's a big difference between surface-level insights and those magical ones when things just click, like a light bulb turning on!
User research has a low barrier to execution as is evident by various functional roles wanting to do their own research. How difficult can it be to ask a few questions to customers (“How will you use my product to solve your problem?”) and arrive at insights (“Users prefer using the FAQs rather than calling customer care as they can’t find the phone number.”)?
However, high-quality research is about high-quality insights. Great insights dial the impact needle all the way up. Once teams experience it, they cannot unsee it. So what is the magic sauce?
From Observations to Insights
Observations are the raw data collected during research. In cross-functional friendly language, we can call this ‘The What’.
Example: Users frequently use the search bar on an e-commerce site.
Insights are the ability to gain an accurate and deep understanding of someone or something. We can call this ‘The So What’.
A surface-level insight offers an obvious explanation for user behaviour.
Example: Users rely heavily on the search functionality to find products.
This insight is surface level as it highlights the ‘how’ but not the ‘why’.
Better insights should go an order of magnitude deeper to uncover underlying motivations and needs.
Example: Users find (evidence) or possibly find (hypothesis) the site navigation confusing and switch to using the search bar to locate products.
Why is this a better insight? It highlights a pain point/frustration. It opens the door to design ideation and action. And, it creates business impact by focusing on and solving the user problem and not a feature problem.
Great insights toy with second-order thinking(deeper consequences) and systems thinking (broader consequences).
Example: Users feel like they waste time and that the company is trying to sell them other things than what they need. So they would rather use the search bar to avoid confusion.
This is a strategic insight as it affects the company’s trust and loyalty factors. To act upon this problem, different teams need to come together and create longer term strategies.
Some obvious techniques that uncover great insights are
creating darn good research questions and interview questions,
nudging and probing, and
The richer the insights are, the more actionable and impactful will the recommendations (”The Now What”) be.
If your teams are looking to up-skill insights efforts, let’s chat.
Framework to get from good to great insights
Guiding principles for great insights
When in doubt, embrace open ended questions that encourage narratives and stories.
Lead with user’s environment and context over solution’s context.
Space your fieldwork and embrace immediate debriefing after user interactions.
Triangulate not just methods and data, but ways of asking the same question. This is a robust area to use AI in the process.
Test hypotheses for not just validation but falsification, viz. What would take for the hypothesis to be wrong?
The role of questions in getting deeper insights
Question framing significantly impacts the quality of insights. Without good questions, it is simply ‘garbage in, garbage out’.
Since examples speak louder than explanations, let’s look at one:
Bad question - Do you find this search bar useful?
There are so many things wrong with this question! A trained researcher will see that it is close-ended, leading, and feature-focused rather than need-focused amongst other issues.
Average question - Can you describe the last time you used the search bar?
This question is open-ended but feature specific. Even when the hypothetical search bar team wants to know the usability of the search bar, this question still does not approach the problem from the user’s perspective. At best, it will give surface insights leading to micro improvements in the search bar functionality.
Good question - Can you retrace your steps that last time you were on the site looking for a product? Follow-up: I saw you touched the search bar. What led you to do that?
Not only is this question open-ended and behavioural, but looks at the wider scope of problem-solving. Insights from this question could potentially help multiple teams.
Here is another example.
Bad question - What do you like about our product?
Average question - When was the last time you found it difficult to use our product?
Good question - When was the last time you had to solve problem X and how did you do it? Follow up: At what point did you consider our product and what was the outcome?
Magical user insights lead to actionable and impactful recommendations that in turn lead to transformative user experiences. By using best practices and expert reviews consistently, researchers and PwDRs can create meaningful user outcomes. The goal after all is not to simply understand what users do, but why they do it and how you can better meet their needs.
I am Soma and I collect and make sense of all the necessary user knowledge to help founders and teams make confident growth and revenue decisions. I create product strategies using research, design and systems-thinking toolkits.
What I’m inspired by recently
Carolina Perez shared some good news about women crash test dummies. She writes cool stuff about non-representative data.