Researching Products Before You Commit
Demand, competition and the margin maths people skip — done before money is committed rather than after.
This is an implementation example, not a client case study. It describes how this work is actually carried out — the method, the sequence and the reasoning. No client is named and no result is claimed, because inventing either would make it worthless as evidence.
Most product research is confirmation bias with a spreadsheet. Someone likes a product, finds data suggesting it sells, and skips the part where they work out what is left after fees, postage, returns and advertising.
- Project type
- Marketplace product and niche research
- Sector
- Online retail
The challenge
Demand data is easy to find and easy to misread. A category with high sales volume is usually one with correspondingly high competition, and a category with no competition frequently has none for a reason.
The objective
A decision based on evidence of demand, an honest read on competition, and a margin calculation that includes every cost — before inventory is bought.
Approach
Work through demand, competition and margin as separate gates. A product must clear all three. Most fail on the third, and the third is the one people calculate last.
Implementation
1. Demand evidence
Use sold listings, not active ones. Active listings tell you what people are hoping to sell; sold listings tell you what people bought.
Look at: sell-through rate, price achieved versus asking, and seasonality over at least twelve months. A category that looks strong in November may be dead for the other eleven.
2. Competition
- How many sellers, and how concentrated?
- Are the top listings held by established sellers with long feedback histories?
- Is there a brand that dominates, and can you compete with it at all?
- Are the existing listings actually good, or is the category poorly served?
A poorly-served category with moderate demand is more interesting than a well-served one with high demand.
3. Margin — the gate most products fail
Calculate the whole thing:
| Cost | Commonly forgotten |
|---|---|
| Unit cost | — |
| Inbound shipping and duty | Frequently |
| Marketplace fees | Rarely |
| Payment processing | Often |
| Outbound postage and packaging | Often underestimated |
| Returns — rate × full cost | Almost always |
| Advertising per unit sold | Almost always |
| Storage | For FBA, often |
The last three are where apparently profitable products turn out not to be. A 12% return rate on a low-margin item can remove the entire margin.
4. Validate before scaling
Small quantity first. Real listings, real customers, real returns.
The data from twenty units sold is worth more than any amount of tool output, because it includes the costs the tools do not model.
5. Be honest about the answer
Sometimes the research says no. That is the research working. The cost of a "no" is the research time; the cost of ignoring it is the inventory.
Technology
- Terapeak
- Marketplace sold data
- Margin modelling
- Keyword research tools
Decisions worth explaining
Sold listings as the demand source, never active listings
Active listings measure seller optimism. Sold listings measure buyer behaviour, which is the thing being forecast.
Returns and advertising included in the margin calculation
These are the two costs most often omitted, and they are frequently the difference between a viable product and a loss-making one.
Small-quantity validation before committing
Twenty real sales surface the costs no research tool models. It is the cheapest information available.
What it produces
A go/no-go decision supported by sold-listing demand data, an honest competition assessment, and a full-cost margin model — with small-quantity validation before any significant commitment. No revenue figures are claimed.
What it teaches
- Active listings are seller hope; sold listings are buyer behaviour.
- Returns and advertising are the costs that quietly remove the margin.
- An underserved category beats a busy one at the same demand level.
- A well-evidenced 'no' is a successful piece of research.
Related services
Related guides
eCommerce & Marketplaces
eBay Product Research: What Should Sellers Analyse?
Demand, competition, margin. The third one is where most products fail, and it is the one calculated last.
eCommerce & Marketplaces
eBay vs Amazon: Which Marketplace Is Better for Sellers?
They suit different products and different risk appetites. Here is how to tell which one fits yours.
Dropshipping & TikTok Shop
How to Start Dropshipping in the UK
The overview: how dropshipping works, what you are legally responsible for, and the order to set it up in, with a detailed guide behind every step.
