The most common consumer-brand mistake we see is the same one we are most tempted to make ourselves: skipping the slow, uncomfortable, unglamorous work of primary customer research before writing the positioning brief. The reasons sound rational. The team feels they already know the customer. The category is "obvious." The competitive landscape can be summarized in a tab of an investor deck. Research feels like a delay tax on the launch timeline.
It is not a delay tax. It is the launch.
What 200 quotes actually unlock
When we say "research" at KEVERA, we mean a specific, mechanical thing. We mean reading two hundred or more real customer statements — full sentences, paragraphs, full forum threads — written by real people in the language they would never use if they thought a marketer was listening. Reddit comments. Quora answers. Amazon verified-purchase one-star reviews. Facebook group replies. Etsy review tails. We document each one with source, context, and a reliability tag.
Two hundred is not a magic number. It is the smallest sample size at which patterns become undeniable. Below one hundred, you can still argue with yourself about which complaints are the "real" one. Somewhere between one hundred and two hundred, the same complaint appears in three different communities, phrased in three different ways, by three different demographic slices — and now it is not a complaint. It is a category fact.
The thing research saves you from
For sofahug, our first brand, the research saved us from a specific product mistake. The category-default fabric for "cat-friendly" sofa covers is a kind of chenille — soft, plush, looks great on a render. Fifteen-plus reviews across one specific competitor showed the same outcome within fifteen minutes of unboxing: a cat's claws caught on the chenille loops and pulled them out, leaving the cover snagged and the cat's claws caught in the fabric. Without those fifteen-plus real reviews, we would have built the same product and shipped the same failure.
The cost of not doing it
We have watched brands skip this step and then spend the next eighteen months iterating on a product that was wrong at the spec. The thing about a wrong spec is that no marketing channel, no influencer partnership, no email sequence, and no creative agency will save you from it. You can spend two million dollars learning what fifty hours of forum reading would have told you for free.
Research is what a small operating company has that a venture-scale brand cannot afford to do, because at venture scale the math demands you launch first and pretend you researched later.
What this looks like in practice
For every brand we build, the first eight to twelve weeks are research — exclusively. No brand brief. No supplier conversations. No domain purchase. Just reading. By the end of the eight weeks we have a documented Desire Map, four to seven validated sub-avatars, a list of the top five purchase objections in customer language, and a competitive map of who is actually winning and why. The brief writes itself after that.
We will keep doing this for every brand. It is the cheapest insurance any operator can buy.