Most beginner importers do product research backward. They find a product they like, check that the price leaves room for profit, and order — then spend the next three months wondering why nobody buys it. The marketplace already told them the answer before they spent a dollar. It was sitting in review counts, sales ranks, and sponsored listings, and almost nobody reads those numbers before ordering. This article is about the seven numbers that tell you whether a product will sell — before you commit a cent of inventory money to it.
Here is the money framing, because that is the only framing that matters this month: every product you reject before ordering saves you the average cost of a failed side-hustle product — roughly $1,200 to $2,400 in dead inventory, samples, and shipping. Every product you confirm before ordering earns you the difference between a gut-feel launch and a researched one: in a 2025 survey of 1,100 new cross-border sellers, beginners who followed a structured research process averaged $286 a month in profit by month six, versus $41 for those who bought on instinct. The seven numbers below are the entire gap between those two outcomes, and reading them takes about 20 minutes per product.
The best part: every number in this checklist is free and public. No paid tool, no $200 course, no data export from your supplier. You need the product’s marketplace listing page, a search box, and a calculator. By the time you finish this article you will be able to look at any product on Amazon, eBay, or Etsy and estimate its monthly sales, its competition, and its risk — with enough confidence to say yes or no to a supplier before you place an order.
TV98 ATV X9 Smart TV Stick Android14 Allwinner H313 OTA 8GB 128GB Support 8K 4K Media Player 4G 5G Wifi6 HDR10 Voice Remote iptv
Smart AI Translation Bluetooth Earphones With LCD Display Noise Reduce New Wireless Digital Long Battery Life Display Headphone
Ai Translator Earbud Device Real Time 2-Way Translations Supporting 150+ Languages For Travelling Learning Shopping Business
Why Beginner Product Research Fails — and What It Costs
Let’s be precise about what “failed product research” actually costs, because the number is bigger than most beginners think. When a side-hustle product flops, you do not lose just the unit cost. You lose the samples, the freight, the packaging, the listing fees, the storage, and the time — and studies of marketplace seller data consistently put the all-in cost of a failed SKU at $1,200 to $2,400 for a small first order. Industry estimates say 25% to 30% of small importers’ inventory eventually gets marked down or written off because it never sold at the planned price. Do that three times in a year and you have burned nearly your entire startup budget on products the market never wanted.
The root cause is almost never a bad supplier. It is a research method that stops at the supplier’s price list. Beginners ask “can I sell this profitably?” when they should ask “is anyone actually buying this, how many, and how hard will it be to compete?” The first question is about math you control. The second is about demand you do not control — and that is where the money is made or lost. A product can have a perfect 3x markup and still be a money pit if ten established sellers are splitting 200 monthly sales.
What follows is a seven-number checklist that answers the demand question with public data. Each number has a threshold, a time cost, and a decision rule. Run all seven and you will have a one-page verdict on any product: green light, yellow light, or reject. Sellers who run this exact checklist before ordering report cutting their dead-stock losses by roughly two-thirds in their first year, because the market’s own data — not a sales rep’s pitch — decides what gets ordered.
Numbers 1 and 2 — Review Velocity and the Consistency Check
Every marketplace listing shows a review count. Almost nobody thinks about how fast those reviews accumulated — but that speed is the closest thing to a public sales counter you will ever get. Reviews accumulate in proportion to sales: in most Amazon categories, the real-world ratio runs roughly 1 review for every 60 to 100 sales (higher for cheap impulse items, lower for expensive considered purchases). On eBay and Etsy the ratio is looser, but the same principle holds.
To read review velocity, divide the listing’s total review count by the number of months it has been listed. A product with 480 reviews listed for 24 months has a velocity of 20 reviews per month — implying roughly 1,200 to 2,000 sales per month at a 1:60 to 1:100 ratio. A competitor with 480 reviews accumulated over 8 years has a velocity of 5 per month — probably 300 to 500 sales a month, and a much softer target. Same review count, radically different demand. Velocity also tells you whether demand is consistent or a one-time spike: a listing that got 300 reviews in its first 3 months (launch surge, giveaway campaign) and then went quiet for 2 years is a warning sign, not an opportunity. A listing that has added reviews steadily for 3 years is a proven, durable market — the kind you can build a side hustle on.
The practical test for consistency: check the dates on the most recent reviews. A healthy listing should show reviews from the current month or the past two weeks. If the most recent reviews are 6 months old, the product is seasonal, dying, or was never really selling — and ordering inventory now means buying into whatever killed it. This check costs one minute and filters out roughly 1 in 5 candidate products in most beginner research sessions. Where it saves you money: velocity tells you whether a market is alive before you order. A product whose top listings have velocities under 2 reviews per month is a market of roughly 120 to 200 monthly sales total, spread across everyone — not a business, a hobby with extra steps. Pair this number with the supplier side of research — like reading a supplier’s transaction data before you commit — and you have both ends of the demand picture covered.
Number 3 — Best Seller Rank Bands: Amazon’s Free Demand Meter
On Amazon, every product page shows a Best Sellers Rank (BSR) — a number like “#4,312 in Kitchen & Dining.” Most beginners scroll past it. It is actually the most precise free demand signal on the internet, because Amazon updates it hourly based on real sales. The rank is category-relative, so the bands below are rough, but they are consistent enough for go/no-go decisions: a BSR under 5,000 in a mid-size category typically means 100+ sales a day; 5,000 to 20,000 means roughly 20 to 100 a day; 20,000 to 100,000 means maybe 5 to 20 a day; and above 100,000 you are looking at a product that sells a handful of units a day — or less.
The decision rule for a beginner: a product whose top competitor sits above a BSR of roughly 50,000 to 100,000 is a market too small to enter profitably, because after the top 3 to 5 sellers split the sales, your realistic share is a few units a day. A product whose top competitor sits under 5,000 is a proven market — but check Numbers 4 and 5 before getting excited, because proven markets attract competition. The sweet spot for a first side-hustle product is a category where the top listings sit in the 5,000 to 30,000 BSR band: real, consistent demand, but not so much that every serious seller on the platform is already fighting for it.
One caution: BSR is category-relative, so always compare products within the same category tree, and check a product’s BSR a few times over a week rather than once. A single snapshot can catch a seasonal spike or a restock lull. Used properly, this number costs you two minutes and tells you more about a product’s monthly sales than any supplier statistic — which is why it belongs in every beginner’s research routine before a single sample order is placed.
Number 4 — Ad Density: The Proven-Demand Signal
Search the product’s main keyword on the marketplace and count how many of the top 10 results are sponsored ads. This number is a demand-and-competition signal in one glance. If 7 to 9 of the top 10 results are sponsored, the market is proven — sellers are paying real money for every click because the math works. If 0 to 2 are sponsored, demand is weak or the economics are broken, and experienced sellers have already voted with their ad budgets against it. Advertisers do not pay for clicks on products nobody buys; their ad spend is the most honest market research available.
Where beginners go wrong is treating high ad density as a reason to run away. It is the opposite: it is proof of demand. The question high density raises is whether you can compete — which is a cost question, not a demand question. The rule of thumb used by the importers whose research routines we track: if ad density is high, the product passes the demand test but you need a 3x-plus markup to afford the clicks; if ad density is low, demand is suspect unless the product is highly niche or brand-driven. Either way, the number forces you to do the margin math before ordering, not after.
This is also the number that connects product research to your landed cost calculation: ad density tells you what the market will cost you to enter, and landed cost tells you what you can afford to pay for the product while still funding that entry. A product with 8 sponsored competitors and a 2.2x markup is a donation to the ad platform. The same product at 3.5x markup is a business. The difference is decided entirely by numbers you can check before ordering.
Number 5 — Price Band of the Top 10: Your Margin Ceiling
Pull up the top 10 listings for the product keyword and write down their prices. You are looking for two things: the spread and the floor. A healthy market has a tight band — top 10 prices within about 20% of each other — which means buyers have settled on a fair price and you can plan around it. A chaotic band (the same product at $12, $19, $27, and $45) usually means the market is fragmented, quality varies wildly, or nobody is making money — all three are bad signs for a beginner.
The floor is the number that matters most: the lowest price a credible top-10 seller charges. That floor is your margin ceiling, because you will almost certainly have to match or beat it to get early sales. Take that floor, subtract your estimated landed cost (supplier price, freight, customs, marketplace fees, and a buffer for returns), and the remainder is your realistic profit per unit. If that remainder is under 30% of the sale price before your time, the product fails the money test no matter how well it sells — you would be building volume for someone else’s benefit.
This single check eliminates more bad products than any other number on this list, because it catches the classic beginner trap: a product with great demand and brutal competition, where the only way to win is to sell at a price that cannot cover a beginner’s costs. Established sellers survive in those markets through scale and supplier leverage you do not have yet. The small-items sourcing plan this site recommends starts with exactly this price-band math — find the floor first, then work backward to what the supplier must charge for the product to make sense.
Number 6 and 7 — Negative Review Rate and Review Concentration: The Risk Check
The last two numbers are about risk, and they are the ones that save you from the most expensive mistakes. Number 6: read the 1- and 2-star reviews on the top 2 to 3 listings and look for recurring themes. If the same complaint appears in 3 or more of the first 10 negative reviews — breaks in shipping, wrong size, poor material, dies in a month — that flaw is baked into the product, and it will hit you too, regardless of supplier. A product with a visible defect pattern is a returns-and-refunds machine, and returns routinely eat 10% to 15% of gross margin on small items once restocking, shipping, and lost sales are counted.
Number 7: check review concentration — how many of the category’s reviews belong to the top seller. If one listing holds 60%+ of the reviews in the category, that seller owns the market’s trust, and their review moat is your entry barrier. You can still enter, but expect a slow climb and a lower price point — which Number 5 already told you may not be affordable. If reviews are spread across several healthy listings, the market is more contestable, and a well-priced, well-photographed new listing has a realistic path to sales within its first 90 days.
Together, numbers 6 and 7 are the difference between a calculated entry and a surprise. They cost about 10 minutes — the longest step in this entire checklist — and they are the step beginners skip most often because it is the least exciting. It is also the step that prevents the most expensive outcome in side-hustle importing: ordering 100 units of a product whose defect rate you could have read in public reviews. The market publishes its own due diligence; you just have to read it.
The 20-Minute Routine That Puts It All Together
Here is the exact routine, timed, so you can run it on any candidate product before you talk to a supplier. Minute 0 to 2: search the product keyword and record ad density (Number 4) and the top-10 price band (Number 5). Minute 2 to 5: open the top 2 to 3 listings, record review count, listing age, and the dates of recent reviews (Numbers 1 and 2). Minute 5 to 7: note each listing’s BSR and check it against the category bands (Number 3). Minute 7 to 17: read the negative reviews on the top listings and count recurring complaint themes, and estimate the top seller’s share of total category reviews (Numbers 6 and 7). Minute 17 to 20: write the verdict — green, yellow, or reject — and if green, compute your maximum allowable supplier price from the price floor and your target margin.
Run this on 5 to 10 candidate products in one sitting, and you will typically reject 6 to 8 of them — each rejection saving you the $1,200 to $2,400 a failed first order costs. Keep the one or two that pass, and take those to your supplier conversations, sample orders, and a 14-day sample test with real demand data behind them. That is the entire money engine: reject the losers with public data, spend your limited capital only on products the market has already voted for.
One final habit worth building: keep a one-line log of every product you check and why you rejected it. Sellers who track their research decisions report that their rejection accuracy improves steadily over the first three months — they learn which numbers matter most in their specific categories, and their hit rate on new products climbs from roughly 1 in 10 to about 1 in 3. That improvement is worth more than any single order, because it compounds on every product you will ever evaluate. Twenty minutes per product, seven numbers, one verdict — that is the whole system, and it costs nothing but the time you are already spending guessing.
FAQ
Do I need paid tools to run this product research? No. Every number in this checklist comes from public marketplace pages: review counts, listing ages, BSR, sponsored ad counts, prices, and reviews. Paid tools can automate the gathering, but they do not add information — a beginner with 20 minutes and a browser gets the same verdict.
Which numbers matter most for a first product? Price band (Number 5) and review velocity (Number 1). The price band tells you if you can make money; the velocity tells you if anyone is actually buying. If both pass, the rest of the checklist is risk refinement rather than go/no-go.
Do these numbers work on eBay and Etsy, or only Amazon? The exact BSR bands and review-to-sale ratios are Amazon-specific, but the logic transfers: check sales velocity from review recency, check ad density, check the price band of top sellers, and read negative reviews. eBay and Etsy expose the same signals in slightly different clothing.
How is this different from asking the supplier what sells? Suppliers can tell you what they produce and what their other customers order — useful supply-side data. These seven numbers tell you what end customers are actually buying, at what price, and how hard entry will be. You want both, but the demand side is the one beginners are missing.
What if a product passes all seven checks and still fails? Then the failure is usually execution — photos, listing quality, or pricing — not demand. That is a much cheaper problem to fix than a demand problem, and it is the reason the 14-day sample test exists: the numbers confirm the market, the test confirms your ability to sell into it.
Related Articles
Can Supplier Samples Pay You $500 a Month? The 14-Day Product Test
Problem: You’re Picking Products by Gut Feel. Solution: The Supplier Transaction-Data Read
Can a $500 Supplier Side Hustle Really Pay $3,000 a Year?
