The spreadsheet had 47 rows. Each one was a product idea the seller had found somewhere: a YouTube recommendation, a Reddit thread, a 'what to sell on Amazon' listicle, a niche they'd read about in a newsletter. They'd spent three weekends building the list.
The problem was that they didn't know what to do with it. All 47 products had some search volume. Several had reasonable review counts. Three or four seemed genuinely interesting. But without a method to evaluate them, the list was just organised procrastination: research theatre that looked productive and produced nothing actionable.
This is the most common product research trap. The problem isn't a lack of ideas. It's a lack of a filtering process that turns ideas into decisions: specifically, a method that can tell you, for each product, whether the demand is real, the competition is beatable, the margin is workable, and the risk is acceptable.
Top Amazon sellers don't have better instincts. They have a repeatable process. They look at the same signals, in the same order, every time. They run the same margin maths. They apply the same disqualifiers. And when something passes all of those filters, they do one more thing before ordering inventory: they test it with advertising.
Based on Hector AI's internal analysis of 1,100+ product launch campaigns (Jan–Dec 2025), products launched after structured demand validation, where a review gap and keyword opportunity were confirmed before inventory commitment, reached break-even ACoS 47% faster than products launched without those validation steps. Methodology available on request. Research quality predicts advertising performance. The two processes are not separate.
What Product Research Actually Is: And What It Isn't
Before the five steps, a clarification: product research is not using a tool to find a product. Tools surface data. Research is the process of interpreting that data to make a decision: specifically, the decision to commit capital to sourcing, listing, and advertising a product.
Most product research guides are actually tool walkthroughs. This one isn't. The process described here can be completed with Amazon's own free tools: the Product Opportunity Explorer, the Search Query Performance report, and the FBA Revenue Calculator. Third-party tools accelerate and aggregate this data. They don't replace the underlying logic.
The Difference Between a Research Process and a Tool Walkthrough
A tool walkthrough tells you how to find a BSR or export a keyword list. A research process tells you what to do with those numbers: specifically, how to evaluate them against each other and against your own business constraints to reach a clear go/no-go decision.
The difference between sellers who launch winning products and those who launch products that plateau or lose money isn't access to tools. It's the quality of the questions they ask at each stage of evaluation, and the discipline to move on when a product fails a filter, even when it's exciting.
The One Question Every Product Idea Must Answer First
Before anything else, one question: does unmet demand exist that you can serve better or differently than existing options?
'Better or differently' is the operative phrase. You don't need to be the first seller in a category: most successful products launch into categories with existing competition. You need to enter a category where there is unfulfilled demand that current sellers aren't capturing well. That unfulfilled demand shows up in review gaps, in keyword white space, in niches where search volume exists but listing quality is poor. Finding it is the core task of product research.
Step 1: Reading Demand Signals Before You Commit
Demand signals are the evidence that shoppers are actively looking for what you're thinking of selling. Three primary signals, read in combination, give you a high-confidence picture of whether genuine, consistent demand exists.
Search Volume as the Starting Filter
Monthly search volume for your target keywords tells you how many shoppers are looking for this type of product on Amazon. Search volume requirements vary significantly by category and niche. Many successful products launch in lower-volume niches with strong unit economics, differentiated positioning, and limited competition. Search volume should therefore be evaluated alongside competition, pricing, margins, and overall market opportunity rather than as a standalone threshold.
Search volume alone doesn't validate demand: it validates awareness. Shoppers are searching for something in this space. That's the minimum requirement, not the go/no-go.
BSR (Best Seller Rank) as a Demand Proxy
The Best Seller Rank (BSR) is Amazon's ranking of a product's sales performance relative to other products in the same category. In general, a lower BSR indicates stronger sales relative to competing products. However, Amazon does not publish a direct relationship between BSR and unit sales, and the same BSR can represent very different sales volumes across categories.
The most useful thing BSR tells you: how many units are multiple sellers in this category selling consistently? Look at the top 10–20 products in your target category. If multiple sellers maintain BSRs under 5,000 in a major category, that's evidence of broad category demand, not just one product's success.
Review Velocity as a Real-Time Demand Indicator
Review velocity, or how quickly products in a category accumulate new reviews, can provide an indirect indication of category activity. However, review generation rates vary considerably by category, customer participation rates, and Amazon's review systems. Categories where leading products consistently gain new reviews may indicate ongoing demand, while little or no review growth over extended periods may suggest slower category activity or seasonality. Review velocity should be interpreted alongside other demand signals rather than as a direct measure of sales.
To estimate review velocity: check a product's total reviews today and compare to a screenshot from 30–60 days ago (accessible via the Wayback Machine or a review tracking tool). The difference divided by the time period gives you a monthly velocity estimate. Review velocity helps indicate category activity. Products with stronger review volume and positive ratings often build greater shopper confidence over time.
Amazon's Product Opportunity Explorer: The Free Signal Most Sellers Miss
Step 2: Analysing the Competitive Landscape Honestly
Demand signals tell you shoppers are looking. Competition analysis tells you whether you can reach them: specifically, whether the existing sellers in this space have built a position you can compete with or whether there's a genuine opening for a new entrant.
What Competition Level Tells You: And What It Doesn't
High competition is not automatically a disqualifier. High competition in a category with strong demand means the category is large enough to support multiple sellers. The question is not whether competition exists, but whether you can differentiate enough to earn click share and convert it to sales.
Low competition requires more scrutiny than most sellers apply. A category with weak competition and apparent demand might signal: (a) a genuine untapped opportunity, or (b) a category that other sellers tried, couldn't make profitable, and abandoned. Both produce the same surface-level data. The review history tells you which one you're looking at.
The Review Gap Analysis: Finding the Opening in a Healthy Market
A review gap exists when demand is proven (strong search volume, healthy BSR among top sellers), but the average review quality, not just the count of top sellers, suggests shoppers are underserved. Specifically: a category where the top products have high review counts but low average ratings (3.8–4.1 stars) on consistent, recurring complaints that your product could address.
This is the signal most worth finding: genuine demand, clearly frustrated buyers, and a specific product improvement you can build in. Your listing's A+ Content, your bullets, and your initial advertising messaging all flow from this gap: making your research directly actionable in your product development and launch strategy.
Reading Search Query Performance Data for Keyword White Space
Amazon's Search Query Performance (SQP) report, available in Brand Analytics for Brand Registry sellers, shows which search queries drive traffic and purchases in a category, along with the click and purchase share held by the top 3 brands for each query.
Keyword white space is a query where search volume is significant, but purchase share is distributed across multiple sellers rather than concentrated among a few dominant brands. Fragmented purchase share may indicate opportunities for new entrants, but it does not guarantee unmet demand or easy market entry. It should be evaluated alongside product differentiation, customer needs, competition, and overall market dynamics before concluding the opportunity.
Step 3: Running the Margin Maths Before You Fall in Love
This step kills more products than any competitive analysis ever will. Products that seem compelling: strong demand, accessible competition, obvious differentiation, frequently fail the margin maths when the full cost stack is calculated honestly.
The Full Landed Cost Calculation: COGS to FBA to Referral Fee
The Minimum Viable Margin for an Amazon FBA Product
After accounting for all five cost components, many sellers use a 30% gross margin as a practical benchmark. However, appropriate margin targets vary by product category, advertising strategy, return rates, and business objectives. Lower margins generally provide less room for advertising and operating costs, but profitability depends on factors such as CPCs, conversion rates, repeat purchase behaviour, and overall business strategy. Your break-even ACoS should always be calculated using your actual contribution margin rather than generic benchmarks.
Step 4: The 5 Research Disqualifiers That Override Everything Else
These five disqualifiers eliminate a product regardless of how compelling its demand signals, competitive landscape, or margin look. When a product triggers any one of them, the correct decision is always the same: move to the next candidate.
Disqualifier 1: Brand Dominance With No Review Gap
If the top sellers in a category are established brands with high review counts, strong ratings, and few recurring customer complaints, competition is likely to be well established. However, high review counts and strong ratings do not automatically eliminate opportunities. Product differentiation, niche positioning, unique features, pricing strategy, or serving underserved customer segments may still create viable entry points. Rather than relying on review counts alone, evaluate whether you can offer a meaningful advantage that resonates with your target audience.
Disqualifier 2: Unavoidable Patent or IP Risk
Before any sourcing conversation, conduct appropriate intellectual property due diligence on the product mechanism, design, branding, and any distinctive features. Reviewing relevant patent and trademark databases can help identify potential issues, particularly in categories where IP disputes are more common. Potential intellectual property risks should be evaluated carefully, and sellers should obtain appropriate legal advice when necessary before proceeding with sourcing or launching a product.
Disqualifier 3: Seasonal Spikes With No Steady Baseline
A product with BSR 400 in December and BSR 15,000 in March is a seasonal product, not a business. The capital tied up in inventory, the FBA storage costs during off-peak months, and the advertising economics that only work during a narrow peak window make pure seasonal products difficult to build sustainable businesses around. If you pursue a seasonal product deliberately, ensure your margin and storage plan account for the full-year cost structure, not just the peak-month economics.
Disqualifier 4: Fragile, Hazmat, or Oversized Products With Compressed Margins
Fragile products generate returns at significantly higher rates than durable goods: returns that come back to you as unsellable inventory or direct credits to buyers. Hazmat (dangerous goods) products require special FBA storage and have carrier restrictions that add complexity and cost. Oversized products carry FBA fees that compress margins even when demand is strong. None of these categories is automatically disqualified, but all of them require margin structures that account for the specific cost headwinds: margins that many sellers don't build into their initial calculations.
Disqualifier 5: Markets Where Amazon Itself Is a Dominant Seller
When Amazon's own retail operation, "Ships from and sold by Amazon.com," holds top positions in a category, the competitive landscape can be different from categories served primarily by third-party sellers. Amazon independently sets retail prices on products it sells directly, which can create highly competitive pricing environments. Before committing to a product opportunity, evaluate whether you can compete sustainably through differentiation, branding, pricing strategy, or customer value, particularly in categories where Amazon Retail has a significant presence.
Step 5: Validating With Advertising Before You Scale Inventory
This step is the one most product research guides don't include, and it's the one that makes the biggest difference to the probability of a successful launch.
The Advertising Validation Principle: Buy Data Before You Buy Stock
The cheapest data point available to any new seller is a small advertising test on a limited quantity: enough to generate statistically meaningful click and conversion data, but not so much that a failed test represents a significant inventory write-off.
The principle: before committing to a full-scale inventory order, consider launching with a smaller initial order that allows you to validate demand while limiting risk. Initial order quantities should be based on supplier minimum order quantities (MOQs), available capital, lead times, and your overall risk tolerance. Once the product is live, use Sponsored Products campaigns over several weeks to gather meaningful conversion and search term data before deciding whether to scale inventory.
The conversion rate, ACoS, and search term data from that test are the most accurate demand signal available: because it's actual buyers, at your actual price, with your actual listing, in real market conditions.
A Small-Scale Ad Test That Confirms Real Buyer Intent
The test campaign structure:
One auto Sponsored Products campaign with a suitable daily budget based on your testing objectives.
Allow sufficient data to accumulate before making major optimizations while continuing to monitor the campaign for obvious issues such as stockouts, excessive spend, budget limitations, or tracking errors.
Review the search term report once enough meaningful data has been collected.
Answer three questions: What is my conversion rate? Which search terms are driving conversions? Is my ACoS directionally aligned with my break-even contribution margin and business objectives?
After sufficient data has been collected, you're looking for directional signals rather than fixed benchmarks:
Conversion rates that are competitive for your product category, price point, and level of competition
Search terms that consistently generate orders
ACoS trending toward a level that aligns with your break-even contribution margin and business objectives
No unexpected return, quality, or customer experience issues
Acceptable conversion rates vary significantly by category, product price, competition, listing quality, and traffic source. Rather than targeting a universal percentage, evaluate conversion performance against relevant category benchmarks and your own profitability goals.
How Validation Quality Predicts Advertising Performance After Launch
Hector AI's internal analysis suggests structured demand validation can help sellers reach break-even ACoS sooner: structured research identifies products where the keyword-to-conversion fit is already confirmed before launch advertising begins.
Structured research and launch testing can reduce the time needed to identify high-performing keywords after launch. By validating demand and gathering early advertising data before scaling inventory, sellers can make more informed optimization decisions and focus their efforts on the search terms that show the strongest potential.
Their campaigns start with a higher signal-to-noise ratio, harvest profitable keywords faster, and reach target ACoS sooner. This is the core discipline behind Hector's Amazon Ads Optimization.
Why Manual Research Breaks as Your Catalogue Grows
A detailed research process for one product takes 4–8 hours, done properly: demand analysis, competition audit, margin calculation, disqualifier checks, and an advertising test interpreted over 3 weeks. For a seller launching one or two products per year, this is entirely workable.
For a brand managing 30 active SKUs and evaluating 20 new product opportunities per quarter, the operating reality of a mid-scale Amazon business: the maths breaks. 20 evaluations × 8 hours = 160 hours per quarter just in product research, before any of the advertising, listing management, or operational work that scaling the existing catalogue requires.
The Research Scaling Problem No One Talks About
Manual product research is not just time-intensive: it's also inconsistent at scale. The quality of a research evaluation depends on the analyst's attention and energy. The same researcher, doing their seventh product evaluation in a day, applies less scrutiny than on the first. Systematic criteria: applied consistently in the same order, against the same standards: eliminate this variability.
The solution is a research brief that captures all five steps in a standardised format: demand signals (with specific thresholds), competition notes, margin calculation, disqualifier checklist, and advertising test results. Every product evaluated by anyone in the team produces the same output format, making comparison and prioritisation straightforward.
How Automation Bridges Research Insight to Advertising Execution
The advertising validation step connects product research directly to advertising strategy. A product that has passed all five research steps and generated positive conversion data in a test campaign doesn't need to start from scratch when full-scale advertising begins. The auto campaign data already shows which search terms convert. That data feeds directly into the launch campaign structure: exact-match manual campaigns for proven keywords, automatic keyword harvesting from week 3, and automated bid rules that maintain performance without requiring daily manual review. This is the same automation layer behind Hector's Amazon PPC Software.
Building a Repeatable Research System, Not a One-Off Process
The five steps in this guide are only as useful as the consistency with which you apply them. A seller who follows the process for product one but skips the margin maths on product three because it 'obviously makes sense' has a process for one product and a gamble for the rest.
The Five Inputs Every Research Cycle Needs
Turning Your Research Findings Into a Launch Brief
Once a product clears all five steps, the research document becomes the launch brief. The primary and secondary keywords from step one define your listing title structure. The review gap from step two defines your differentiation messaging in bullets and A+ content. The margin from step three defines your target ACoS for advertising. The disqualifier checks from step four define your IP clearance documentation. The advertising test data from step five defines your initial keyword targets for your exact-match launch campaigns.