Product Research
Product research is the highest-stakes decision a seller makes. everything after it
is execution on a choice already made. This skill validates a product opportunity
across the four things that decide it, and pressure-tests the niche around it,
returning a clear go or no-go.
Two layers run here. The Opportunity Square is the single-product test. The Niche
Scorecard is the niche-level test that frames the product. Run the Square when you
have one product in mind. Add the Scorecard when the question is "is this whole
category winnable" or when comparing several niches before picking the product.
When to use this
- A seller has a product idea and wants it pressure-tested before committing.
- A seller is comparing several product ideas or several niches.
- A seller keeps picking products that look promising and then disappoint.
- Validating a product before sourcing or sampling.
- Sizing a niche before committing time to specific products inside it.
The framework. The Opportunity Square
A product opportunity stands on four corners. A weak corner is not fatal alone, but
two weak corners are a no-go, and the verdict must say which corners carry it.
Corner 1. Demand
Is there real, steady demand? Look for consistent search volume and consistent sales
across the top listings, not one spike. Check the trend. demand fading is a trap, even
if today's number looks fine.
Corner 2. Competition
Can a new entrant win? Read the top listings. Strong brands with thousands of reviews
and excellent listings make a hard square. Weak listings, tired images, and a wall of
complaints in the reviews are an open one. Count the review barrier honestly.
Corner 3. Profit
Does the money work? Selling price minus the full FBA fee stack, the landed cost,
returns, storage, and a realistic launch ad budget must leave a real net margin. Use
amz-fba-calculator. A product that sells well at a 4 percent net margin is a job, not
a business.
Corner 4. Barrier to entry
What stands in the way? Patents, brand-locked categories, gating, heavy compliance,
high minimum order quantities, oversize fee tiers, fragile or hazardous shipping. A
high barrier is sometimes good, it keeps competitors out, but only if the seller can
clear it themselves.
Sub-framework. The Niche Scorecard
When the question is whether to enter a whole niche (not just a single ASIN), score
the niche on six gates. Two of them are pass-or-fail. fail either and the niche is
avoid, regardless of the rest.
| Gate |
What to look for |
Weight |
| Demand |
Steady search volume and sales across the top listings |
Pass/fail |
| Margin headroom |
Selling price leaves real margin after fees and ads |
Pass/fail |
| Competition depth |
How many entrenched sellers, and how strong |
Scored |
| Review barrier |
Review counts of the top listings. the wall a new entrant climbs |
Scored |
| Differentiation room |
Are the top listings beatable, or already excellent |
Scored |
| Seasonality and risk |
Year-round demand, no patent, brand, or compliance trap |
Scored |
Demand and margin headroom are the pass-or-fail gates. No demand, nothing to
win. No margin, winning does not pay. Only score the rest if both pass.
Reading the scored gates
- Competition depth. A page-one of 10 strong brands is a hard niche. a page-one
with weak listings and tired images is an opening.
- Review barrier. If the top listings have 200 to 800 reviews, a new entrant can
climb in. If they have 5,000-plus, the review wall alone can take a year.
- Differentiation room. Read the top listings' reviews. recurring complaints are
the room to differentiate. If buyers are already happy, there is no opening.
- Seasonality and risk. A niche that sells only in December is a different
business. a niche full of patented designs or one dominant brand is a trap.
How the two layers relate. The Square asks "is this specific product the right bet."
The Scorecard asks "is the surrounding niche worth fighting in at all." A product can
look strong in the Square and still sit inside a brutal niche, in which case the
verdict should be Investigate at minimum.
Step by step
Collect inputs. The product idea, category, typical price, what the user can
observe about the top listings (reviews, ratings, image quality, complaints),
rough costs, any known patents, gating, or compliance issues, and any seasonality
or brand concentration signals.
If the question is niche-level, run the Scorecard first. Pass-or-fail gates
before the scored ones. A failed gate ends the analysis with Avoid.
Score each corner of the Square. Strong, mixed, or weak, each backed by
evidence.
Run the profit math properly. landed cost, full fee stack, returns, storage,
launch ads. Do not let a guess stand in for the calculation.
Find the angle. If this is a go, state the specific reason a new entrant wins:
an unmet complaint, a weak incumbent, a price or quality gap.
Verdict. Go (at least three strong corners and a real angle, niche not Avoid),
Investigate (mixed, name what data would settle it), or No-go (two weak corners,
no angle, or niche-level Avoid).
Run the quality check, then deliver.
Output format
## Product Research. [product]
### The Opportunity Square
Demand: [strong/mixed/weak] . [evidence]
Competition: [strong/mixed/weak] . [evidence]
Profit: [strong/mixed/weak] . [the net margin number]
Barrier to entry: [strong/mixed/weak] . [evidence]
### Niche Scorecard (if niche-level)
Demand gate: [pass/fail] . [evidence]
Margin headroom gate: [pass/fail] . [evidence]
Competition depth: [score] . [notes]
Review barrier: [score] . [notes]
Differentiation room: [score] . [notes]
Seasonality and risk: [score] . [notes]
### The angle
[the specific way to win, or "no clear angle"]
### Verdict: [GO / INVESTIGATE / NO-GO]
[reasoning, naming which corners carry it and which are risks]
### If GO, next step
[sourcing, sampling, or deeper validation]
Worked example
Idea: a silicone stretch lid set, around 14 USD.
Demand: strong, steady year-round volume. Competition: mixed, a crowded page-one but
the listings are generic and the images are weak. Profit: weak, at 14 USD after the
fee stack, landed cost, and launch ads the net margin is thin. Barrier: low, easy to
source, which also means easy for others.
Niche Scorecard: demand passes, margin headroom borderline-fail at this price point,
review barrier high (top listings 3k-6k reviews), differentiation room real
(recurring complaints about stretch tearing).
Verdict: No-go as-is. Two weak-leaning corners, profit and barrier, a borderline
margin gate, and a low-price product where ads eat the margin. The angle (better
images, better listing, non-tearing version) is real but cannot fix a structurally
thin margin. Investigate a higher-priced, clearly premium version that lifts the
profit corner before committing.
Quality check
- All four corners of the Square are scored with evidence, not asserted.
- If niche-level, the Scorecard pass-or-fail gates were checked first and a failed
gate produced Avoid.
- The profit corner uses a real calculation through the full fee stack, not a guess.
- A product with two weak corners is a no-go, and the verdict says so.
- A go verdict names a specific, concrete angle to win.
- The verdict is Go, Investigate, or No-go, with reasoning that names the carrying
corners and the risks.
Common mistakes
- Demand tunnel vision. Picking a product because search volume is high, ignoring
that profit or competition fails.
- Guessing the margin. Skipping the real fee-stack math and assuming it works.
- Ignoring the trend. Today's demand looks fine while the category is sliding.
- No angle. Entering a niche of excellent listings with happy buyers and nothing
to take.
- Falling for demand alone at the niche level. High search volume in a niche with
no margin or a 10,000-review wall is a trap, not an opportunity.
- Skipping the risk scan. Walking into a patented design or a single-brand-locked
category.
- Confirmation bias. Researching to justify a product the seller already chose.
Built by Jay GPT Pro
Part of Amazon Pro Skills. Production-grade skills for serious Amazon sellers.
Free and open. Built by Jay Margaliot.
I share a new AI play for Amazon sellers every week, free, in my WhatsApp group.
Join here: https://chat.whatsapp.com/ILX65p1yWcaIG3c9WGHpTY
1---2name: amz-product-research3description: Research and validate an Amazon product opportunity end to end, and evaluate whether the niche around it is winnable. Assesses demand, competition, profit potential, entry barriers, review wall, differentiation room, and seasonality, and returns a go/no-go with the reasoning. Use when a user asks to research a product, find a product to sell, validate a product idea, assess an opportunity, evaluate a niche, find a profitable niche, judge whether a category is worth entering, or compare niches. Trigger phrases: "product research", "find a product to sell", "validate this product", "is this a good product", "product opportunity", "should I sell this", "niche finder", "evaluate this niche", "is this niche worth it", "good niche", "low competition niche", "should I enter". Works with zero tools. the user describes the product and what they can observe.4---56# Product Research78Product research is the highest-stakes decision a seller makes. everything after it9is execution on a choice already made. This skill validates a product opportunity10across the four things that decide it, and pressure-tests the niche around it,11returning a clear go or no-go.1213Two layers run here. The Opportunity Square is the single-product test. The Niche14Scorecard is the niche-level test that frames the product. Run the Square when you15have one product in mind. Add the Scorecard when the question is "is this whole16category winnable" or when comparing several niches before picking the product.1718## When to use this1920- A seller has a product idea and wants it pressure-tested before committing.21- A seller is comparing several product ideas or several niches.22- A seller keeps picking products that look promising and then disappoint.23- Validating a product before sourcing or sampling.24- Sizing a niche before committing time to specific products inside it.2526## The framework. The Opportunity Square2728A product opportunity stands on four corners. A weak corner is not fatal alone, but29two weak corners are a no-go, and the verdict must say which corners carry it.3031### Corner 1. Demand3233Is there real, steady demand? Look for consistent search volume and consistent sales34across the top listings, not one spike. Check the trend. demand fading is a trap, even35if today's number looks fine.3637### Corner 2. Competition3839Can a new entrant win? Read the top listings. Strong brands with thousands of reviews40and excellent listings make a hard square. Weak listings, tired images, and a wall of41complaints in the reviews are an open one. Count the review barrier honestly.4243### Corner 3. Profit4445Does the money work? Selling price minus the full FBA fee stack, the landed cost,46returns, storage, and a realistic launch ad budget must leave a real net margin. Use47amz-fba-calculator. A product that sells well at a 4 percent net margin is a job, not48a business.4950### Corner 4. Barrier to entry5152What stands in the way? Patents, brand-locked categories, gating, heavy compliance,53high minimum order quantities, oversize fee tiers, fragile or hazardous shipping. A54high barrier is sometimes good, it keeps competitors out, but only if the seller can55clear it themselves.5657## Sub-framework. The Niche Scorecard5859When the question is whether to enter a whole niche (not just a single ASIN), score60the niche on six gates. Two of them are pass-or-fail. fail either and the niche is61avoid, regardless of the rest.6263| Gate | What to look for | Weight |64|------|------------------|--------|65| Demand | Steady search volume and sales across the top listings | Pass/fail |66| Margin headroom | Selling price leaves real margin after fees and ads | Pass/fail |67| Competition depth | How many entrenched sellers, and how strong | Scored |68| Review barrier | Review counts of the top listings. the wall a new entrant climbs | Scored |69| Differentiation room | Are the top listings beatable, or already excellent | Scored |70| Seasonality and risk | Year-round demand, no patent, brand, or compliance trap | Scored |7172**Demand** and **margin headroom** are the pass-or-fail gates. No demand, nothing to73win. No margin, winning does not pay. Only score the rest if both pass.7475### Reading the scored gates7677- **Competition depth.** A page-one of 10 strong brands is a hard niche. a page-one78 with weak listings and tired images is an opening.79- **Review barrier.** If the top listings have 200 to 800 reviews, a new entrant can80 climb in. If they have 5,000-plus, the review wall alone can take a year.81- **Differentiation room.** Read the top listings' reviews. recurring complaints are82 the room to differentiate. If buyers are already happy, there is no opening.83- **Seasonality and risk.** A niche that sells only in December is a different84 business. a niche full of patented designs or one dominant brand is a trap.8586How the two layers relate. The Square asks "is this specific product the right bet."87The Scorecard asks "is the surrounding niche worth fighting in at all." A product can88look strong in the Square and still sit inside a brutal niche, in which case the89verdict should be Investigate at minimum.9091## Step by step92931. **Collect inputs.** The product idea, category, typical price, what the user can94 observe about the top listings (reviews, ratings, image quality, complaints),95 rough costs, any known patents, gating, or compliance issues, and any seasonality96 or brand concentration signals.97982. **If the question is niche-level, run the Scorecard first.** Pass-or-fail gates99 before the scored ones. A failed gate ends the analysis with Avoid.1001013. **Score each corner of the Square.** Strong, mixed, or weak, each backed by102 evidence.1031044. **Run the profit math** properly. landed cost, full fee stack, returns, storage,105 launch ads. Do not let a guess stand in for the calculation.1061075. **Find the angle.** If this is a go, state the specific reason a new entrant wins:108 an unmet complaint, a weak incumbent, a price or quality gap.1091106. **Verdict.** Go (at least three strong corners and a real angle, niche not Avoid),111 Investigate (mixed, name what data would settle it), or No-go (two weak corners,112 no angle, or niche-level Avoid).1131147. **Run the quality check**, then deliver.115116## Output format117118```119## Product Research. [product]120121### The Opportunity Square122Demand: [strong/mixed/weak] . [evidence]123Competition: [strong/mixed/weak] . [evidence]124Profit: [strong/mixed/weak] . [the net margin number]125Barrier to entry: [strong/mixed/weak] . [evidence]126127### Niche Scorecard (if niche-level)128Demand gate: [pass/fail] . [evidence]129Margin headroom gate: [pass/fail] . [evidence]130Competition depth: [score] . [notes]131Review barrier: [score] . [notes]132Differentiation room: [score] . [notes]133Seasonality and risk: [score] . [notes]134135### The angle136[the specific way to win, or "no clear angle"]137138### Verdict: [GO / INVESTIGATE / NO-GO]139[reasoning, naming which corners carry it and which are risks]140141### If GO, next step142[sourcing, sampling, or deeper validation]143```144145## Worked example146147Idea: a silicone stretch lid set, around 14 USD.148149Demand: strong, steady year-round volume. Competition: mixed, a crowded page-one but150the listings are generic and the images are weak. Profit: weak, at 14 USD after the151fee stack, landed cost, and launch ads the net margin is thin. Barrier: low, easy to152source, which also means easy for others.153154Niche Scorecard: demand passes, margin headroom borderline-fail at this price point,155review barrier high (top listings 3k-6k reviews), differentiation room real156(recurring complaints about stretch tearing).157158Verdict: No-go as-is. Two weak-leaning corners, profit and barrier, a borderline159margin gate, and a low-price product where ads eat the margin. The angle (better160images, better listing, non-tearing version) is real but cannot fix a structurally161thin margin. Investigate a higher-priced, clearly premium version that lifts the162profit corner before committing.163164## Quality check165166- All four corners of the Square are scored with evidence, not asserted.167- If niche-level, the Scorecard pass-or-fail gates were checked first and a failed168 gate produced Avoid.169- The profit corner uses a real calculation through the full fee stack, not a guess.170- A product with two weak corners is a no-go, and the verdict says so.171- A go verdict names a specific, concrete angle to win.172- The verdict is Go, Investigate, or No-go, with reasoning that names the carrying173 corners and the risks.174175## Common mistakes176177- **Demand tunnel vision.** Picking a product because search volume is high, ignoring178 that profit or competition fails.179- **Guessing the margin.** Skipping the real fee-stack math and assuming it works.180- **Ignoring the trend.** Today's demand looks fine while the category is sliding.181- **No angle.** Entering a niche of excellent listings with happy buyers and nothing182 to take.183- **Falling for demand alone at the niche level.** High search volume in a niche with184 no margin or a 10,000-review wall is a trap, not an opportunity.185- **Skipping the risk scan.** Walking into a patented design or a single-brand-locked186 category.187- **Confirmation bias.** Researching to justify a product the seller already chose.188189---190191## Built by Jay GPT Pro192193Part of **Amazon Pro Skills**. Production-grade skills for serious Amazon sellers.194Free and open. Built by Jay Margaliot.195196I share a new AI play for Amazon sellers every week, free, in my WhatsApp group.197Join here: https://chat.whatsapp.com/ILX65p1yWcaIG3c9WGHpTY