Most first products fail for one of three reasons: the seller picked something with no real demand, chose a product where competition had already consolidated, or misjudged the economics after shipping and fees. All three are avoidable with an honest screening process before any money is spent.
This guide walks through how to screen a product idea the way an experienced seller would. The framework is deliberately conservative — it filters out ideas that look attractive at first but collapse under real numbers.
Step 1 — Confirm real demand, not imagined demand
The most common mistake is falling in love with an idea because it feels clever. Demand is not how interesting a product is. It is whether people are already searching for it and buying it right now.
Three signals confirm real demand:
Search volume
Tools like Google Keyword Planner, Google Trends, and Amazon's own search bar show what people are actively searching for. A search term with steady volume over 12 months is a demand signal. A term that spikes and drops is a trend, not a market.
What to check: search the product term in Google Trends over five years. If the line is flat, steady, or mildly up, demand exists. If the line is trending down or has never been stable, the market may be shrinking or too niche.
Active sellers and review counts
Search the product term on Amazon or Etsy. Look at the top three results. If they have hundreds or thousands of reviews, the market is real. That volume represents hundreds or thousands of individual purchasing decisions.
Do not mistake high reviews for a saturated market. High reviews mean proven demand. The correct question is not "is there competition?" but "can I differentiate enough to win a share of it?"
Communities talking about it
Search the product category on Reddit, Quora, or niche forums. If people are asking questions about it, complaining about existing options, or comparing products, that's a live market. If the conversation is silent, demand is likely absent or too small to build a business on.
Example
Idea: personalized pet portraits for dog owners.
Google Trends: steady search volume over five years, with a mild seasonal bump at Christmas. Amazon search: hundreds of sellers, top listings with 500+ reviews. Etsy: same pattern, thousands of active sellers.
Verdict: proven demand. The question is no longer whether people want this — it is how to win a share of it.
Step 2 — Understand what competition looks like
Competition is not a problem to avoid. It is a map of the market. The goal is to find a category where competition is fragmented enough that a new seller can win a niche, without being so crowded that the race is already over.
Three competition profiles:
| Profile | What it looks like | Signal |
|---|---|---|
| Zero competition | No listings, no search results, no reviews anywhere | Usually a red flag — no demand or no viable market |
| Fragmented competition | Many small sellers, no dominant brand, top listings have 50–500 reviews | Healthy — room for a focused new entrant |
| Consolidated competition | A few large sellers with thousands of reviews, Amazon Basics, or one brand dominating | Difficult — differentiation and ad spend are required to enter |
Zero competition is the most misleading. It often means one of four things: the market is too small to sustain a business, the product is illegal or restricted, the product has negative margins that every seller has already discovered, or the product category is so new that demand has not arrived yet. None of these are good starting points.
Fragmented competition is the target. If the top five listings have between 50 and 500 reviews and no single seller controls more than 20% of the category, there is room for a new seller to enter with a specific angle — a bundle, a niche, a colorway, a size, or a price point that the current sellers are not covering.
Step 3 — Run the economics before the excitement
This is where most ideas die, and it is the step beginners skip. A product that looks attractive at $30 often collapses once every line item is included.
The full stack to model:
- Product cost — what the supplier charges per unit
- Freight and duties — if importing, the landed cost per unit
- Platform fees — referral, transaction, payment processing
- Fulfillment — FBA, 3PL, or self-fulfilled shipping
- Returns — a realistic return rate for your category
- Advertising — customer acquisition cost per sale
Run the math on a $30 product with realistic assumptions:
| Line | Amount |
|---|---|
| Sale price | $30.00 |
| Platform fee (15%) | –$4.50 |
| Payment processing | –$1.20 |
| Fulfillment | –$4.50 |
| Product cost | –$8.00 |
| Freight and duties | –$1.50 |
| Returns (5% amortized) | –$1.50 |
| Advertising (20% of revenue) | –$6.00 |
| Net profit | $2.80 |
A $30 product with a $2.80 margin is not a viable business. A single bad month of returns or a rising ad cost wipes out the profit entirely.
The healthy target: net profit of at least 20% of the sale price. On a $30 product, that means $6.00 or more. If your math lands below that, the product is not ready — either renegotiate cost, find a different supplier, or pick a different product.
The rule of thirds
As a rough screening tool: about one-third of the sale price goes to product cost and freight, one-third to platform, fulfillment, and returns, and one-third remains as gross margin before ads.
If ads consume more than half of the remaining third, the product is fragile. If they consume less than half, there is room to grow.
Step 4 — Score the idea before you commit
Before you order inventory, score the product across six dimensions. This is not a perfect predictor. It is a filter that stops obviously weak ideas from reaching the ordering stage.
| Dimension | Question to answer |
|---|---|
| Demand | Are people actively searching for this and buying it today? |
| Competition | Is the category fragmented enough to enter with a specific angle? |
| Margin | Does the math leave at least 20% net profit after all costs? |
| Shipping | Is the product small, light, and cheap to ship without damage? |
| Repeat purchase | Will customers buy again, or is this a one-time purchase? |
| Risk | Is the category free from heavy regulation, seasonal collapse, or platform restrictions? |
Score each from 1 to 5. Total out of 30. Above 22 is worth a small test. Between 15 and 22 needs improvement in the weak dimensions before committing. Below 15 is usually a pass.
Step 5 — Order small, test, and let the market decide
Even after scoring well, the first order should be small. Not six months of inventory. Not one month. Roughly two to three months of expected sales, plus a small safety buffer.
This gives you enough inventory to run a real test — to see whether the product converts, whether it returns, whether the ad cost holds, whether reviews come in — without tying up significant cash in inventory you might not be able to move.
If the test works, reorder and scale. If the test fails, you have lost a small amount on one product, not your entire starting capital. Either way, you keep the option to try again.
Score an idea Check demand Estimate profit Calculate startup cost Plan your first order