Detection guide
A counterfeit is a fake made by an unauthorized manufacturer and passed off as your product — unauthorized at the factory, not just the storefront. Of all the violation types, counterfeits carry the strongest legal remedies and the hardest detection problem: fakes surface across marketplaces, social commerce, and standalone storefronts, and migrate channels after every takedown.
This guide covers how counterfeit listings present, why single-signal detection fails in both directions, and the multi-signal scoring Pricelysis uses to rank suspect listings for review and test purchase.
Counterfeit sellers optimize for your buyers' trust signals, so listings often look plausible at a glance:
Price is the most common tell, but not a universal one — some counterfeit operations price at parity precisely to dodge discount-based screening. That's why detection has to stack signals.
Manual search can't keep pace with channel migration. A takedown on one marketplace moves the operation to another storefront or platform within days, and a team spot-checking brand searches weekly is always a cycle behind.
Single-signal tools fail in both directions. Screen on discount alone and you flag every legitimate sale while missing parity-priced fakes. Screen on image matching alone and you drown in noise — counterfeiters use your genuine photos, and so do your legitimate retailers. No one signal separates a fake from a discounter.
And lumping everything under "infringement" backfires operationally: counterfeit complaints, grey-market complaints, and unauthorized-seller complaints have different evidence bars and different platform workflows. Filing the wrong one gets rejected and burns credibility with the platform's enforcement team.
Pricelysis scores every candidate listing with a weighted multi-signal model, then routes by confidence rather than firing takedowns from a score.
A listing advertised below 70% of the expected price for that product triggers the heaviest-weighted signal. Deep discounts on premium goods are the classic counterfeit bait — and the threshold is explicit, not a black-box suspicion score.
Your brand keyword in a listing title only counts against sellers who aren't you and aren't on your authorized-dealer safe list. Your own listings and your partners' listings never fire this signal.
Two patterns from the counterfeit-seller playbook, checked with explicit rules: bulk-generated random alphanumeric handles, and "official / authentic / genuine" impersonation names.
Listings on marketplaces with documented counterfeit concentrations carry a base-rate weight, so the same evidence scores higher where fakes are structurally more common.
Scores bucket into likely counterfeit, uncertain, and likely authentic — with thresholds you can tune. Likely counterfeits go to the top of your inbox; uncertain cases queue for review and test purchase. No complaint is filed from a score alone: a human approves every action, and contested cases are settled by a test buy, not an algorithm.
$467B
global trade in fake goods
Source: OECD, 2025
≈5%
of EU cosmetics sales are counterfeit
Source: OECD-EUIPO, 2024
Counterfeit enforcement is the one category where platforms move fast — if your evidence is complete:
Customs recordation is worth the paperwork for physical-goods brands: counterfeit shipments can be seized at the border in many jurisdictions, upstream of every marketplace.
Rank by score, visibility, and blast radius. A likely-counterfeit offer attached to your best-selling ASIN — where it inherits your reviews and your traffic — outranks a low-score listing on a marketplace you barely sell on.
Budget test purchases monthly and spend them on the highest-score contested cases. A confirmed fake with a documented test buy typically clears a platform's counterfeit workflow quickly, and creates a template for the next one.
A counterfeit is fake at the manufacturing level. A grey-market good is genuine product moved through the wrong channel. The distinction decides the enforcement path: counterfeit complaints carry the strongest remedies but collapse if the goods turn out to be genuine diverted stock — which is why test purchases matter.
For most platform IP workflows, effectively yes. Amazon Brand Registry, eBay's VeRO program, and Google's counterfeit complaint paths all key on registered rights. If your registration is pending, prioritize it — it unlocks every other tool.
They usually do — it's the cheapest way to look genuine. Stolen photography strengthens your complaint (copyright on top of trademark), and it's why image matching alone can't separate fakes from legitimate retailers using the same assets.
Parity pricing dodges discount-based screening and captures full margin from deceived buyers. Multi-signal scoring still catches these through seller-name patterns and marketplace base rates — price is one signal, not the definition.
The OECD estimated global trade in counterfeit and pirated goods at $467B in 2025, and OECD-EUIPO work in 2024 put roughly 5% of EU cosmetics sales as counterfeit. Exposure varies by category — high-margin, light-to-ship, trust-dependent products are hit hardest.
Yes. They erode price perception, siphon ad clicks, and their reviews — on piggybacked offers — land on your listing. The brand damage compounds before the first chargeback.
Run a free brand audit. We scan your active SKUs against every violation type on this page and return a PDF with timestamped evidence for each match.
Pricelysis Assistant
Ask about Pricelysis, pricing, or brand protection