Amazon product opportunity scorecard: buy, test, revise or reject

1. The short answer

Score eight gates from 0 to 3. Then apply hard stops before you look at the total.

• 0: failed, contradictory or unknown at a decision-critical point.
• 1: weak signal or mostly assumption-based.
• 2: plausible and partly evidenced.
• 3: evidence-backed for the stated store, product configuration and date.

The eight gates are:

1. Demand.
2. Competition.
3. Supply.
4. Economics.
5. Cash.
6. Product risk.
7. Evidence quality.
8. Operating readiness.

A high score cannot override a hard stop. If the exact product cannot be identified, the supplier route cannot be documented, the economics are negative or the product responsibility is unresolved, the candidate is blocked until the missing evidence is addressed.

For the evidence behind each row, use the Amazon product research checklist. For the complete research sequence, use how to find products to sell on Amazon UK and Europe.

2. 1. Define the candidate before scoring it

Do not score a category or a vague product name.

Write down:

• target Amazon store;
• exact product configuration;
• brand or generic status;
• dimensions, weight, material and pack count;
• intended fulfilment route;
• proposed customer price and currency;
• likely supplier type;
• research date; and
• the decision the scorecard is meant to support.

The unit of analysis matters. “Reusable water bottle” is too broad. A particular branded 750 ml bottle, a generic insulated two-pack and a replacement lid are different products with different supply, listing, fulfilment and compliance questions.

If the candidate changes, version the scorecard. Do not change the dimensions or pack count while keeping the old score and date.

3. 2. The eight gates

Gate one: demand

Ask whether there is a repeatable customer need supported by appropriately scoped evidence.

Evidence can include Amazon's signed-in product or niche data, public category observations, review and return themes, and relative search-interest data. Amazon describes Product Opportunity Explorer as a tool for exploring customer needs, search and purchasing behaviour, competition, reviews and returns. The exact metrics must be verified for the target account and store.

Use the Amazon UK Best Sellers pages as a dated public snapshot only. Use Google Trends for relative interest and seasonality. It is not absolute search volume or Amazon sales data.

Do not award a 3 because a tool shows high potential. Award it only when the evidence is clear enough for the decision you are making.

Gate two: competition

Ask whether you can understand the competitive offer without assuming an unrealistic price, review position or conversion rate.

Record visible brands, seller structure, price spread, variation differences, listing quality and recurring review themes. Keep product-niche competition separate from wholesale offer-level competition.

There is no universal “good” review count. A low-review listing may be weak demand. A high-review listing may be a strong established competitor. Score the evidence, not the number you hoped to see.

Gate three: supply

Ask whether the exact product has a credible, documentable route from supplier to you.

Record the supplier's legal entity, actual role, product identifier, specification, sample path, quotation basis, payment beneficiary and lead time. For branded goods, add the supply-chain and authorisation questions in the invoice and brand-authorisation guide.

A supplier catalogue is not supply proof. It is an invitation to verify.

Gate four: economics

Ask whether the product survives realistic costs and a downside case.

Use a named store, selling price, fulfilment route and date. Include supplier cost, landed cost, Amazon fees, preparation, inbound movement, returns, advertising and any other per-unit cost that the decision depends on.

The Amazon UK fee and revenue estimator can help estimate fees for specified product and fulfilment inputs. The landed-cost guide owns the import calculation. The Amazon contribution-margin guide owns the sale-price-to-contribution handoff.

Do not call the result profit when fixed overhead, financing, tax treatment or owner time is outside the model.

Gate five: cash

Ask whether the business can fund the order and survive the timing.

Record MOQ, case pack, deposits, balance payment, production, inspection, freight, receiving, stock cover, reorder timing and the downside demand case.

A high per-unit margin does not make a cash-heavy MOQ safe.

Gate six: product risk

Ask whether restrictions, compliance, intellectual property, fulfilment and return risks are understood for the target market.

The Amazon restricted-products guide covers platform approval questions. The UK and EU product-compliance guide covers the wider product responsibility question. Neither is a reason to assume that a product is safe because another seller lists it.

If the candidate is regulated, high-risk or outside your knowledge, a scorecard is a screening tool. It is not professional sign-off.

Gate seven: evidence quality

Ask whether the important inputs can be inspected and reconciled.

Evidence quality is separate from demand. A beautiful dashboard screenshot can still be the wrong market or the wrong product configuration.

Gate eight: operating readiness

Ask whether you can list, fulfil, support and monitor the product in the target store.

Record listing identity, images and specifications, prep, packaging, customer service, returns, stock monitoring and the route for handling defects.

4. 3. Weights and the illustrative total

The scorecard weights economics and product risk more heavily because a demand signal is not useful if the product cannot be sold lawfully or the contribution case fails.

Multiply each score by its weight and add the rows. The maximum illustrative total is 51.

That number has no external meaning. It is a prioritisation aid. Do not write “this product has a 78 percent chance of success” or “anything over 40 is a winner”. The model does not support those claims.

5. 4. Hard stops come before the total

Mark the candidate blocked if any of these is unresolved:

• the exact product configuration or identifier is uncertain;
• the supply route or invoice path cannot be documented;
• the contribution case is negative or relies on an unexplained price assumption;
• a category, product, IP or compliance question has not been screened;
• the available cash cannot cover the stated MOQ and lead time; or
• a third-party estimate is being presented as actual Amazon data.

These are Source and Stock editorial controls. They are not Amazon policies and they do not guarantee a safe or profitable purchase.

The practical rule is simple: a hard stop overrides a high total.

6. 5. Decision bands after the hard-stop screen

Use a small number of decisions:

Reject or pause

Use this when a hard stop exists or the missing evidence would cost more to obtain than the candidate justifies.

Write the reason. “No” is useful when you know what made it no.

Research next

Use this when the candidate is plausible but needs a defined evidence action. Examples include a supplier response, sample, fee calculation, product restriction check or clearer trend comparison.

Test carefully

Use this when the hard stops are cleared, the downside case is tolerable and a bounded sample or small order can reduce the remaining uncertainty.

Define the test limit before placing it. Include the maximum cash, units, time and conditions that would stop the test.

Buy under written controls

Use this only when the exact product, supply route, economics, cash case, risk screen and operating route are documented for the intended store. The decision still needs an order, receiving and monitoring plan.

7. 6. Worked illustrative example

The following product and numbers are fictional.

Suppose you are comparing a generic two-pack of adjustable drawer dividers for Amazon UK. You have one public category observation, a relative trend comparison, three possible suppliers and an early fee estimate.

The scorecard might look like this:

The total is not the decision. The supply score is 1 because the exact product and material are not reconciled. If that uncertainty affects the listing, invoice or compliance position, it is a hard stop.

The correct next decision is probably “request a specification and sample”, not “buy”. If the supplier cannot answer the specification question, the candidate should be paused or rejected even though the total is not low.

If the sample matches, the supplier documents agree and the downside case remains tolerable, update the scorecard with a new version and date. Do not simply change the score in place.

8. 7. Make the worksheet auditable

The on-page and downloadable scorecard should include:

Add a separate downside section for selling price, demand, returns, fulfilment, advertising and lead time. Add a final decision field with owner and review date.

Keep the formula visible. A reader should be able to inspect which rows create the total and which hard stop changes the outcome.

9. 8. Recalibrate after real evidence

The scorecard should learn from the work.

After a sample, quote, listing, test or first order, record:

• which assumption changed;
• which source was stronger or weaker than expected;
• what the supplier or marketplace actually confirmed;
• which cost was missing;
• whether the downside case was realistic; and
• whether the scoring rule needs to be clarified.

Do not quietly change the weights to make a failed product look good. Keep the old version, explain the change and use the new framework for future candidates.

10. What the scorecard cannot tell you

It cannot guarantee:

• demand or sales;
• Amazon approval or a Featured Offer;
• a stable price;
• supplier quality or authenticity;
• compliance;
• a particular return rate; or
• profit.

It can tell you whether your evidence is strong enough for the next decision. That is a more useful promise.

11. What to do next

Start with the product research checklist and build one record for each exact candidate. Then use this scorecard to prioritise the work.

For a candidate that survives the hard stops:

1. Build a supplier longlist with European supplier discovery.
2. Verify the counterparty with supplier vetting.
3. Compare offers using supplier quotation comparison.
4. Calculate the purchase-to-warehouse cost with landed cost.
5. Recheck sale-price-to-contribution economics with Amazon contribution margin.
6. Run the product risk screen before ordering.

If the candidate fails, record why and move on. The scorecard has done its job.

12. Sources and review note

This article uses Amazon's public explanation of Product Opportunity Explorer, Amazon UK public marketplace and fee pages, and Google's official Trends methodology. Amazon tools, fees and programmes can change by store, account and product. Recheck the current first-party source before using a live scorecard.

This is an editorial prioritisation framework, not a statistical forecast, professional advice or a promise of sales, approval, ranking or profit.