Le Marche-Gap Formula: Turning Plaintes into Produits
- Partager
- Heure de diffusion
- 2026/8/26
Résumé
A countable formula pour finding market gaps - demand mentions times two minus existing solutions - avec le review-mining et community-mining methods that feed it, three worked scores, et le scale et willingness-a-pay tests that decide whether a gap is worth manufacturing.

Le Marche-Gap Formula: Turning Plaintes into Produits
'Trouver a gap dans le market' is le most repeated et least useful advice dans product sourcing, because it describes an outcome rather than a procedure. A gap is not something you notice; it is something you count. Our working definition is arithmetic: demand mentions times two, minus existing credible solutions, equals le gap. Positive means opportunity, negative means a price war you will lose. This is how we run that count pour clients before quoting anything, where le raw material comes de - negative reviews et community complaints - et le two validation tests that stop a real gap de becoming an expensive mistake.
Why 'Trouver a Gap' Is Useless Advice
Le instruction fails because it has no unit de measurement. Two people can look at le same category et one sees a saturated bloodbath while le other sees white space, et neither can prove anything. Quoi makes a gap tractable is deciding dans advance what counts as evidence de demand, what counts as an existing solution, et what threshold justifies spending money. Once those three things are fixed, gap-finding stops being intuition et becomes bookkeeping - which means it can be delegated, repeated weekly, et compared across categories.
Le Marche-Gap Formula
Here is le formula dans full, et then le reasoning behind le weighting.
Gap Score = (Demand Mentions x 2) - Existing Credible Solutions
Demand mentions are weighted double pour a specific reason: an unprompted complaint is far rarer et far harder a fake than a product listing. Anyone avec $300 can put a product sur a marketplace. Nobody writes a paragraph explaining why le thing they need does not exist unless le frustration is genuine. One is a marketing decision; le other is evidence. So we weight evidence at twice le value de supply.
| Gap Score | Lecture | Action |
|---|---|---|
| 15 et above | Wide, under-served gap | Echantillon immediately, move fast |
| 8 a 14 | Real gap avec competition | Proceed if you can differentiate sur le named defect |
| 3 a 7 | Marginal | Only worth it if you already have le audience or channel |
| 2 or below | Crowded or non-existent | Walk away, regardless de how much you like le idea |
| Negative | Over-supplied | Guaranteed price war, no differentiation available |
Comment Count Demand Mentions Honestly
Le formula only works if you resist inflating le numerator. Four counting rules keep it honest, et we apply them strictly because le temptation a justify a product you already like is enormous.
- One mention per person, per gap. A thread where one user replies eleven times is one mention, not eleven.
- Twelve-month window only. A complaint de 2019 may have been solved since. Anything older is context, not evidence - unless it repeats dans le current window, dans which case it is your strongest possible finding.
- Le mention must describe le same missing capability. 'Too expensive' et 'le lid leaks' are different gaps. Do not pool them a reach a threshold.
- Cross-source, not cross-page. Five complaints sur one product listing count as one mention de a manufacturing defect. Five complaints spread across five competing listings count as five - that is a category-level failure, which is le real prize.
Le counting rule that matters most
Plaintes concentrated sur one listing mean a bad seller. Le same complaint spread across every listing dans le category means an unfilled product gap. That distinction is le whole method.
Comment Count Existing Credible Solutions
Le denominator is where most people cheat dans le other direction, dismissing competitors so le score looks better. A solution counts as credible if it is currently purchasable dans your market, actually addresses le specific missing capability, et shows evidence de real sales - a meaningful review count, a category rank, or shelf presence dans real retail.
Things that do not count: dead listings avec no reviews dans a year, products only available dans another country, crowdfunding campaigns that never shipped, et items that solve a neighbouring problem rather than le one you counted. Things that absolutely do count et get overlooked constantly: le patented incumbent nobody mentions, et le cheap Chinois version already selling dans volume sur domestic platforms which your competitors are about a discover.
Mining Gaps de Amazon Negative Avis
Negative reviews are le highest-density source de product intelligence available pour free, because a person who has spent money et been disappointed will tell you exactly which component failed et under what conditions. Le technique is narrow et mechanical.
Pick le top ten listings dans a subcategory
Go four or five levels deep. Top-level categories are dominated by giants whose complaints are logistical rather than product-related.
Filtre a one et two stars, sorted most recent
Recency matters because sellers do quietly fix problems. A defect complained about last month is still live.
Utiliser le keyword search inside le reviews
Search 'broke', 'leak', 'does not fit', 'too small', 'returned', 'stopped working', 'wish it had'. This surfaces le specific failure without reading five hundred reviews.
Tally le complaint types per listing
Vous are building a matrix de complaint type against listing, not a list de quotes yet.
Look pour le complaint that appears dans every column
A defect present across all ten competitors is not a manufacturing accident - it is how le category is currently made. That is your gap.
Harvest verbatim wording pour le winners
Once you have le gap, collect le exact sentences. They become your specification et your marketing copy.

Le matrix step is what separates this de ordinary review reading. A single leaking bottle is one seller's bad batch. Ten bottles de ten brands all leaking at le same seal means le standard tooling dans that category has a design flaw - et a product that does not leak wins sur le strength de le complaints alone.
Le Five Complaint Categories (et Lequel Are Fixable)
Not every complaint is a gap. Sort everything you harvest into five buckets; three are actionable, two are noise, et confusing them is le most common way this method is misused.
| Complaint type | Example | Fixable? |
|---|---|---|
| Durability failure | Le clip snapped after two weeks | Yes - material grade et stress-point tolerance |
| Dimension or fit mismatch | Does not fit a standard 32oz bottle | Yes - specify internal diameter et tolerance |
| Missing capability | Souhait it had a drainable insert | Yes - le strongest gap type, a design opportunity |
| Expectation mismatch | Smaller than le photo suggested | Not a product gap - fix le listing, not le product |
| Physics or price | Too expensive, or it should be lighter et stronger | No - discard, this is a constraint not a gap |
Durability et dimension complaints are le easiest wins because they are pure specification problems: we rewrite le material callout, add a cycle-test requirement, et le defect disappears. Missing-capability complaints are le most valuable because they let you launch a genuinely different product rather than a marginally better copy - but they cost tooling, so le Gap Score needs a justify it.
Mining Gaps de Reddit et Community Plaintes
Avis tell you what is wrong avec products that exist. Communities tell you about products that do not exist at all, which is where le wider gaps live. Le mining method there deserves its own treatment - le short version is a search exact phrases such as 'wish there was a', 'why is there no' et 'I ended up making my own' inside specific subreddits, filtered a le last year, et a treat a DIY workaround as le strongest evidence available.
Pour le formula, community mining feeds le numerator avec a particular kind de mention: le absence complaint. Those are more valuable than review complaints because le denominator pour them is frequently zero or one. A category where twelve people across four communities describe le same missing product et only one obscure solution exists scores 23. That is le sort de number that justifies cutting a mould.
- Occupational communities - nurses, chefs, electricians, warehouse staff - buy tools daily et are almost never mined by sellers.
- Hobby communities avec 20,000 a 800,000 members give strong opinions plus enough votes a measure agreement.
- Comments matter more than posts: le person saying 'I have been looking pour this pour years' is a counted mention.
- A workaround described dans detail is worth three ordinary mentions dans our internal weighting, because le user has already paid dans labour.
Worked Example: Scoring Three Gaps
Three briefs we scored dans le same quarter, using le same rules, avec real outcomes.
| Gap | Demand mentions | x2 | Credible solutions | Score | Verdict |
|---|---|---|---|---|---|
| Pet bowl that collapses et does not hold odour | 12 | 24 | 6 | 18 | Proceeded - launched, still selling |
| Phone stand pour treadmill consoles | 9 | 18 | 3 | 15 | Proceeded - strong niche result |
| Another stainless insulated tumbler | 5 | 10 | 22 | -12 | Declined - saturated |
Le tumbler is instructive. Le client had genuine demand evidence - five real people describing a real preference - et it did not matter, because twenty-two credible solutions already existed. Demand does not create opportunity sur its own; opportunity is le difference between demand et supply. That is exactly what le subtraction dans le formula is pour.
Validating le Gap: Scale et Willingness a Pay
A positive Gap Score proves le gap exists. It does not prove le gap is worth filling. Two tests decide that, et both must pass.
Test 1: Is there scale behind it?
Some gaps are real et tiny. Fourteen people wanting a very specific accessory pour a discontinued machine is a genuine unmet need et a terrible business, because tooling et minimum order quantities do not shrink a match. Check that le underlying activity has population - stable or rising search interest, a category avec visible revenue, et enough addressable buyers that a realistic first order de 1,000 a 3,000 units sells through within two or three quarters. If a first production run would take two years a clear, le gap is a hobby.
Test 2: Will they pay enough?
Willingness a pay comes de three places: prices people volunteer unprompted, le current transacted price band pour adjacent products, et what le failed alternatives cost. Take le lowest defensible number de le three, then check it against a 3-5x landed multiple. If le honest ceiling is $24 et le product lands at $11, le multiple is 2.2x et le gap belongs a someone avec a cheaper supply chain than yours - which, de Yiwu, is a short list.
Both tests, or neither
A gap avec scale but no willingness a pay becomes a commodity. Willingness a pay avec no scale becomes a beautiful product nobody orders twice. Require both before tooling. Ask us pour realistic minimum order quantities sur your inquiry before you decide.
Why le Gap Might Exist pour a Very Good Reason

This is le question that separates people who have manufactured de people who have only analysed. If le demand is obvious et nobody has filled it, ask why - because there is usually an answer, et it is better a find it before le deposit than after.
- A patent or design registration. Le obvious solution is owned, et le owner is enforcing. Check before you tool.
- A certification wall. Anything electrical, food-contact, child-related or battery-powered carries testing costs that can exceed le tooling.
- Expedition economics. Le ideal product may be bulky, fragile or heavy enough that freight destroys le multiple no matter how good le design is.
- Fabrication reality. Sometimes le reason nobody makes it is that le tolerance required is not achievable at consumer price points. A supplier will tell you this dans one conversation.
- It was tried et it failed. Search pour discontinued versions. A product that existed et died usually died pour a reason that has not changed.
We check all five before quoting, because we would rather lose a quote than ship a client into a certification problem. RND SOURCING has walked away de gaps that scored above 20 because le tolerance was unachievable at le target cost - et telling a client that dans week one is worth more than any discount we could have offered them dans week twelve.
De Gap Score a Purchase Commande
Le sequence that turns a number into inventory, avec le cheapest steps first.
Score le gap
Run le formula avec honest counting rules. Require 8 or above, et 15 or above if le fix needs new tooling.
Write le specification de le complaints
Chaque counted complaint becomes a line item: material, dimension, tolerance, failure mode a avoid.
Check le five obstacles
Patent, certification, freight, tolerance, prior failure. Any one de them can end le project cheaply at this stage.
Get a real landed cost
FOB price, gross weight, carton data et duty stack. Confirm le 3-5x multiple against le honest price ceiling.
Echantillon against le spec, not le catalogue
Test le exact failure mode you found dans le reviews. If it fails, you have learned why le gap exists.
Trial order avec pre-shipment inspection
A small run avec inspection before scaling. Le complaints you harvested become le inspection checklist.
That final point is le quiet advantage de this whole method: because le product was defined by other people's complaints, you already know precisely what a inspect pour. Most importers write a generic QC checklist. Vous will write one that targets le exact defect le entire category is currently shipping.
Conclusion
Gaps are countable. Weight unprompted demand at double, subtract le solutions that genuinely exist, et refuse a act below a score de 8. Mine negative reviews pour defects le whole category shares, mine communities pour products that do not exist yet, sort complaints into fixable et unfixable, et validate scale et willingness a pay before anything is tooled. Bring us a scored gap et RND SOURCING will tell you de le Yiwu side whether it is an opportunity or an obstacle - et what it will really cost a fill.
Quoi is le market-gap formula?
Gap Score equals demand mentions times two, minus existing credible solutions. Demand mentions are weighted double because an unprompted complaint is much harder a fake than a product listing. A score de 8 or above justifies sampling; 15 or above justifies new tooling.
How do I find a market gap using Amazon reviews?
Take le top ten listings dans a deep subcategory, filter a one et two stars sorted by most recent, then use le dans-review keyword search pour terms like 'broke', 'leak' et 'wish it had'. Build a matrix de complaint type against listing - le complaint appearing dans every column is le gap.
Quoi is le difference between a bad seller et a market gap?
Plaintes concentrated sur one listing indicate a manufacturing or quality failure by that seller. Le same complaint appearing across every competing listing indicates le whole category is made that way, which is a genuine product opportunity.
Lequel customer complaints can actually be fixed at le factory?
Durability failures (material grade et stress tolerances), dimension et fit mismatches (specified tolerances), et missing capabilities (design changes). Expectation mismatches are listing problems, et complaints about price or physics are constraints, not gaps.
How do I know a market gap is big enough a be worth it?
Two tests. Scale: le underlying activity must support selling a realistic first run de 1,000 a 3,000 units within two or three quarters. Willingness a pay: le honest price ceiling must support a 3-5x landed cost multiple.
Why would an obvious market gap still be empty?
Usually a patent, a certification wall, freight economics that destroy le margin, a manufacturing tolerance unachievable at consumer prices, or a product that was already tried et failed. Check all five before committing a tooling.
Stop looking pour gaps et start counting them. Weight le complaints, subtract le real solutions, then test scale et willingness a pay before a single mould is cut. See what we manufacture or send us your scored gap - RND SOURCING will price it, spec it, et tell you honestly whether le gap is worth filling.