The word insight gets used for almost anything: a statistic, an observation, a chart from a report. That inflation made it nearly useless in meetings.
An insight is something else, scarcer and harder to come by: a truth about behavior that explains why people do what they do, and that a brand can address.
What it is and what it is not
Data describes. “62% buy online on Sundays” is data.
An observation spots a pattern. “People compare prices on their phone while standing in the store” is an observation.
An insight explains the tension behind the behavior. “I compare prices at the shelf because I don’t want to feel stupid after I buy” is an insight: it has emotion, conflict, and it suggests what the brand can do.
The simplest test: if someone on the team says “that’s just how we all are,” it is probably an insight. If they say “interesting,” it is data.
The three components
A solid insight always contains these three elements.
A behavioral truth, what people actually do, not what they say they do in a survey.
A tension, a conflict between what they want and what they can have, or between what they feel and what they show.
An opportunity for the brand, a space where the product or service can credibly resolve or ease that tension.
Without the tension you have a description. Without the opportunity you have interesting sociology that leads to no campaign.
Where to find them
Real conversations with customers
Twenty well-conducted half-hour interviews produce more material than any trends report. The key is asking about specific episodes, not general opinions.
Instead of “what do you look for in a product like this?”, ask “walk me through the last time you bought one, from when you first thought about it to when you used it.”
The people in direct contact
Salespeople, floor staff, customer service, and community managers hear objections every day. It is the cheapest source inside any company and the most wasted.
Reviews and comments
One and two-star reviews are especially valuable: that is where people explain in detail what they expected and did not get. Competitor reviews count as much as your own.
Search queries
What people type into a search engine when nobody is watching is a source of honesty that is hard to match. Long queries reveal real doubts and fears.
Observation in context
Watching someone use the product in their environment reveals things no interview captures, because much behavior is automatic and goes unremembered.
How to run an interview that works
The difference between a useful interview and a useless one is the kind of question, not the duration.
Ask about facts, not intentions. “Would you buy this?” produces polite, false answers; “what did you buy last time and why that one?” produces real material.
Ask for the full story of the last episode: what prompted the search, where they looked, what made them hesitate, what decided it. The hesitation moments are where tensions live.
Stay quiet after each answer. The three or four seconds of discomfort usually bring the second half, which is the honest one.
And record verbatim phrases, in the interviewee’s words. An insight rewritten in corporate language loses exactly what made it useful.
From raw material to insight
The process has four steps and none of them is a brainstorm.
First, gather the raw material: direct quotes, observations, comments. Unedited and uninterpreted for now.
Second, group by tension, not by topic. “Price” is a topic; “I want to take care of myself without looking like I’m trying” is a tension.
Third, write each candidate in first person, the way a real person would say it. If it sounds like a deck headline, it is not there yet.
Fourth, put it through two tests: recognition, that someone in the audience identifies with it; and ownership, that the brand can address it credibly.
Worked example
Seeing the contrast across levels on the same behavior makes it concrete.
Data: “70% of insurance buyers compare at least three options.”
Observation: “people request quotes from several insurers but do not compare coverage in depth.”
Insight: “I get several quotes so I feel like I did my homework, even though I know I don’t fully understand what I’m comparing.”
The difference shows in what each one enables. From the data, no campaign follows. From the insight, several do, all built around removing the feeling of deciding blind.
Validating before you build on it
An insight that only convinces the room that produced it is a liability. Two cheap checks protect against that.
The first is the stranger test. Show the written insight to people in your audience who were not interviewed and ask whether it describes them. Agreement should be immediate, not reasoned.
The second is the competitor test. Ask whether a rival could address the same tension just as credibly. If yes, it is a category truth and will not differentiate anything you build on it.
Run both before the creative team starts work. Validating afterward turns into defending the decision rather than testing it.
The common traps
The convenience insight: the one that appears right after the campaign was already decided, to justify it backwards.
The category insight: true for every competitor and therefore useless as a differentiator.
The demographic insight: “young people value authenticity.” It explains nothing and could be said of any group.
The borrowed insight: imported from another market, works in the deck and collapses against the local customer.
And the tensionless insight, which is simply a polite description of the target audience.
How it gets used in the brief
The insight enters the brief as a short first-person sentence and connects directly to what the brand will offer.
That link is what allows creative proposals to be evaluated afterward: a route that does not address the insight can be dismissed without debating aesthetics.
If the brief lacks one, evaluation happens on taste. It is the origin of most long meetings about colors. We cover the full structure in how to write a creative brief.
How many a campaign needs
One. A campaign built on three insights ends up saying three different things and fixes none of them.
You can have several identified and save them for later. Choosing which to address this time is a strategic decision, and it is worth making with the business objective in front of you.
At Conqr we work the insight before opening a single design file, because it is what makes an idea arguable on evidence.
Frequently asked questions
Do I need market research to find insights?
Not necessarily. Twenty well-conducted conversations, your category’s reviews, and a session with your sales team cover most projects.
How many interviews are enough?
Fifteen to twenty-five. Beyond that, themes start repeating and additional material adds little.
Can an insight be unflattering to the brand?
It can be uncomfortable, and the best ones usually are. What it cannot be is false or impossible to address with what the brand actually offers.
How often should they be refreshed?
An annual review is reasonable, and a mandatory one whenever the product, the price, or the market context changes.
Who should be in the insight session?
Someone from marketing, someone from sales, and someone who has spoken to a customer this week. That mix is what keeps it from staying theoretical.
Want to find the insight that holds your next campaign together? Let’s talk.