This blog has been updated in September, 2024.
‘User Insight’ is one of the most frequently (and often loosely) used word by UX Researchers, Market Researchers, UX Designers, Data Analytics teams, Management stakeholders and almost everyone else in the organisation. However, the usage of the word “Insight” without a shared understanding can cause notable expectation mismatch and negative business impact. If you are a researcher reading this, Insights management is a relevant and juicy topic, but that is for another day!
A good “User Insight” needs to be actionable. It takes some practice to properly write insights and only well written insights provide value of UX research to the business. I clarify the meaning and definition of ‘Insight’ from a UX Research perspective and the way we at UXArmy, use it in communication – internally as well as with our existing clients, business prospects, etc.
To begin with, User immersions are the source of insights. Simply put – no user engagement, no real Insights.
Let me describe my little journey in pursuit of clarity on the word ‘Insight’. I tried to Google the definition of ‘Insight’, I first hit Wikipedia. From a definition perspective, it seems to have no clear, factual mistake:
“Insight is the understanding of a specific cause and effect in a specific context.”
— Wikipedia
However, it is the ‘related meanings’ of the definitions that are more commonly put to convenient misinterpretations.
I encounter vague and sometime clear sounding definitions. For example, ‘Good Insights must be compelling, without being preachy. They must be truthful, without being too obvious. They must be empathetic, without being presumptuous.’ If some of these definitions do not convince you how bad Insight definitions can get, here are some that actually come closer in meaning and credibility:
“A big discovery that changes market conversation.”
— John Griffiths
“An Insight is something you didn’t know before. Something that may change the way you think about a problem”
— DaveTrott
Here is an example by Kevin Drew Davis, Chief Creative Officer, DDB San Francisco.
FACT: People tend to feed their pets twice a day.
OBSERVATION: They tend to feed them at breakfast and dinner time.
INSIGHT: People feel guilty eating in front of their pets.
Not confused yet? Hence, every time you hear someone using the word ‘Insight’ in a pitch to entice you, take a step back, think and clarify what does he/she is actually referring to.
In reference to UX Research, here is some useful information about User Insights.
a. Quantitative Research—Insights, and Data and Information
Insights are often confused with Data and Information. Data is the raw and unprocessed numbers and text collected from a quantitative research or Usability testing. Information is the processed version of the data that is often presented in a form where viewers can understand from it for instance Heatmaps, Mouse Movement paths, Navigation journeys, etc. Insights are derived from analysing the information and data.
One good example is how a fitness-tracking tool provides data and information, and how the users can carry out the action with the Insight derived from the information.
b. Qualitative Research—Insight and Finding
Another common confusion is between Insight and Findings.
Insight is not a Finding! A Finding indicates the WHAT of user behaviour, whereas insights indicate WHY the users behave in that way.
Finding tells us what people say and why they do it. In a Qualitative research context, get them to do what they said and observe for the truth about what they said. Add to it the imaginary real context or the situation in which those people would do certain things, you are moving towards an Insight.
The difference between Data, Observations and Insights is also illustrated through several videos in the article “The insight: the most important part of the brief”. An actionable Insight helps you learn something about people’s behaviour and able to do something about people’s problem.
The most easily and often forgotten part of an Insight is the Context. Our response and behaviour varies between one situation and another.
For instance, while using a printer at home I do not bother to collect the printouts until I need them. However, at work I collect them as soon as they are printed. That presents an opportunity for a Printer manufacturing business to avoid printed papers from falling on the home floor and getting lost.
Therefore, our User Research team suggests to our clients that Ethnographic Research is one of the most appropriate research methods to build product strategies or finding new opportunities. It could reveal people’s problems about which stakeholders haven’t thought about earlier.
Badly written insights do exist! How does one figure out that an insight is not delivering value? Here are some ways to diagnose that:
i) Insight has nothing to action on or states something that is already known. For instance “The Users associate the brand with something Premium, but the pricing is slightly higher than competitors.”
It is already established that price sensitive customers want lower pricing. So this insight is not new and therefore not useful.
ii) Insight does not support user behaviour with adequate reasoning. For instance, “Users shared mixed feeling on the new visual design of the app – some said it is Clean, while some found it cluttered.”
The insight does not present clear user motivations or persona specific classification in response to the app’s visual design, making it unusable by the business.
iii) Insight without any context is just a fact / finding. That finding although correct cannot be made use of by the product teams. For example “In the past month, mobile traffic to the website grew by 10% whereas Computer traffic reduced by 8%.”
While the wording is informational in nature, it does not tell what might have caused the traffic to shift from Computer to mobile. So this is also not counted as an “Insight”.
In conclusion, it does not matter so much what is the correct definition of ‘Insight’ is. For our clients, we always differentiate between UX Research Findings and Insights. While Findings from a Research can be several, user insights tend to be far fewer. These are the hidden opportunities which bring in fresh perspectives.
If you can’t remember a lot of what you read, at least remember this. User insights must be:
i) non-obvious
ii) relevant, in-context
iii) actionable
Insights must be thoroughly validated and detailed before setting sail to build expensive software. Implementing a Lean prototype based on the Insight to further learn and validate the Insight can save a lot of time and budget in expensive software rework. The end goal? A User Insight must add significant value to the business and to its’ end Consumers.