Read Three Hundred Reviews Instead of Commissioning Research

Most brands are using customer reviews for half the job.

A bad review arrives. Someone forwards it to the relevant team. The issue gets fixed, a response goes out, everyone moves on. Reviews are treated as a support queue with a public-relations attachment.

That is a legitimate use. It is also the smaller half of what is sitting there. Spend an hour reading reviews differently and they start looking much more like market research, except it has already been conducted, it is free, and nobody had to be recruited for a focus group.

Why reviews beat commissioned research for a lean team

I want to be careful here, because this is not an argument that formal research is worthless. It is not. It is an argument about what a team without a research budget should do instead of nothing. Reviews have three properties that survey research usually does not.

Nobody was prompted. A survey asks about the things you already thought to ask about. A review is written about whatever the person could not stop thinking about, which is a different and often more useful set.

The language is theirs. This is worth more than most marketers realise. Customers frequently describe your value proposition better than your own website does, and those exact phrases are the ones that will work in your copy, because they are already how people talk about the category.

It includes your competitors at the same cost. You cannot commission research into how someone else’s customers feel. You can read it this afternoon.

How to actually run it

1. Pick four brands: yours and three competitors. Not the three you admire. The three you actually lose to.
2. Pull from where people naturally talk. Google reviews, Amazon, app stores, marketplaces, whichever surface fits your category. Aim for a few hundred across the four brands.
3. Read for four things, and keep four separate lists.
4. Keep the exact wording. Do not paraphrase into marketing language as you go. The value degrades the moment you translate it.

The four lists are: what people genuinely value, and note that this is often not what your marketing emphasises. What they keep complaining about, where frequency matters more than intensity. What customers praise in a competitor that nobody in the category delivers consistently, which is the positioning gap and the highest-value output of the whole exercise. And what people have simply accepted as a category-wide frustration.

That last one deserves attention. Everyone complains about it, nobody expects it to change. That is either a genuine constraint or the best opportunity on your list, and working out which is a real strategic question.

Why three-star reviews carry the most

This is the part I would fight for.

A one-star review often comes from a single catastrophic experience, and it tells you about a failure mode rather than about the product. A five-star review tells you what went right, usually in three words. Three-star reviews have nuance, because the person is genuinely conflicted and has to explain themselves.

“The product is good, but…” “I like the service, except…” “I would recommend them if…”

There is a great deal sitting after those words. It is the closest thing to an honest customer interview you will find in public, and it is the section of the review set almost everyone skips, which is exactly why it is worth your attention.

Where this method breaks

Reviews over-represent the extremes. People write when delighted or furious, and the quiet middle of your customer base is underrepresented in a way you cannot correct for. Treat frequency as directional, not statistical.

Reviews also skew towards consumer and small-business categories. In complex enterprise sales, the buyer often never writes one, and you are back to talking to people.

And in a category with visible review coercion, the star ratings are contaminated. The free text usually still holds, because a paragraph is much harder to dictate at a doorstep than a tap on five stars.

What to do with the output

You should finish this exercise with three artefacts: a list of phrases in customer language you can use verbatim, one positioning gap you can test, and one category frustration you have decided either to solve or to ignore deliberately.

If you finish with a summary deck and no decision, you did the research part and skipped the point.

Who this applies to

Founders and marketers without a research budget, in categories where customers write publicly. My challenge to you: read fifty three-star reviews in your category this week and see whether you still believe your positioning on Friday.

Frequently asked questions

Can customer reviews replace formal market research?

Not entirely, but for a team with no research budget they are the highest-value substitute available, and they cover competitors at no extra cost.

How many reviews should I read?

A few hundred across your brand and three competitors. Enough for patterns to repeat, not so many that the exercise never finishes.

Why are three-star reviews the most useful?

Because the reviewer is conflicted and has to explain themselves. One-star reviews describe a failure, five-star reviews describe delight, three-star reviews describe the trade-off.

What are the limits of using reviews as research?

They over-represent extremes, they under-represent complex enterprise buyers, and in categories with coerced ratings the stars are unreliable. The free text usually still holds.