Get help with the data depth and do the judgment yourself. The right question is whether this
is a growing market you can profitably capture, not whether the product looks good. A tool
answers the second question. Answering the first takes market size, growth trajectory, return
rate, and a traffic plan.
The short version
- Wrong question: is this a good product. Right question: is this a growing market I can capture.
- Minimum market size $2 million a year. Below that the economics rarely work.
- Review count is a snapshot, not a trend. It says nothing about direction.
- Differentiate from negative reviews, not from imagination.
- Buy the data depth. Own the decision.
What research should actually examine
I run Flapen with 50 operators managing about 70 brands, and product research is the
core of everything we do. We analyze 90 plus data points, and these are the ones that decide
outcomes.
| Factor | What it tells you | What most beginners use instead |
|---|---|---|
| Market size | Whether there is enough revenue to capture | Search volume of one keyword |
| Growth trajectory | Whether it is rising or dying | A snapshot of today |
| Return rate | Whether the category destroys margin quietly | Ignored entirely |
| Rating gap | Where the market wants something better | Star rating alone |
| Segment dynamics | Where inside the category to enter | Whole-category averages |
| Traffic capturability | Whether you can reach buyers profitably | Assumed |
The gap between the two columns is the whole problem. Helium 10 and Jungle Scout can show
review count, search volume, and a snapshot of today, so frameworks built on them inherit
that limit and teach a filter rather than a method.
The market size floor
We use $2 million a year as a minimum. Below that there is not enough revenue to capture
profitably once you account for cost of customer acquisition.
That number is not magic and it is not universal. What matters is having one at all. A
beginner without a floor evaluates every market as potentially viable, which means the
decision gets made on enthusiasm rather than arithmetic.
Ask any research provider what their floor is and how they arrived at it. A specific number
with reasoning means they have a method. "It depends" means they will validate whatever you
bring them.
Differentiation without guessing
The standard advice is to differentiate through creativity, bundling, or better branding. All
three are guesses dressed as strategy.
The reliable method is feedback-driven. Read the negative reviews on the top competing
products, and measure the rating gap between what buyers expected and what they received.
The market tells you where it is asking for something better, and innovating only there is
far more likely to work than invention.
Those reviews do double duty. They give you the vocabulary buyers actually use for your
listing, and the objections your images need to answer.
What to buy and what to keep
Buy the data depth. Analyzing dozens of factors across a category is time-consuming and
mechanical, and it is faster with help.
Keep the decision. You will live with this market for years, and the judgment about whether
to enter it is not one you can meaningfully delegate. An agency that makes the choice for you
has taken on a decision it will not bear the consequences of.
The middle ground most beginners miss: get a free market sizing before committing capital.
Most agencies offer one. Ours returns a written report within 48 hours. Sometimes it concludes
that the market is too small, which is the cheapest useful information available to a first
time seller.
What most agencies will not tell you
Research is the easiest deliverable in this industry to make look impressive. A spreadsheet
with hundreds of rows and color-coded scores takes little time and rarely produces a
decision.
Judge the reasoning, not the volume. Ask why this market rather than the four adjacent ones,
what the growth trajectory looks like over three years, what the return rate implies about
the category, and where the rating gap says you should innovate. Specific answers are hard to
fabricate and take about ten minutes to test.
The second thing: no research provider is incentivized to conclude that you should not
proceed, because that ends the engagement. Ask directly what would make them tell you to walk
away. A specific threshold is a good sign. An assurance that every category has opportunity is
not.
Related answers
- What products to use for Amazon keyword research
- New seller audit checklist from an agency
- How to launch first product on Amazon
- Amazon beginner budget: agency vs DIY tools
- Hiring an Amazon agency: the complete guide
Ask us what would make us tell you to walk away. There is a number, at Flapen.
