Unauthorized Sellers: How to Find Who Undercuts Your Brand

The Report Amazon Rejected Was Probably the Wrong Report

You found an unauthorized seller offering your product at 30% below your dealer price. You do not supply them. You filed a report through Brand Registry, and Amazon closed it without action.

That is not the process failing. That is the process working as designed. Brand Registry, Project Zero, Walmart Brand Portal and eBay VeRO are intellectual property channels. They act on infringement: counterfeits, trademark misuse, copyright, patents. None of them removes a seller because that seller is not on your approved list, and none of them acts on price. If your only claim is "we did not authorize this", you have described a distribution problem to a system built for an IP problem.

Finding out who is selling your product is the easy half. The hard half is building a theory of harm a platform, or a court, can act on. This guide covers both, in the order they have to happen: monitor, triage, test-buy, trace lot codes to the leak, then enforce.

Two Problems Wearing the Same Label

A brand saying it has an unauthorized seller problem usually means one of two very different things, and the difference decides everything downstream.

The first is counterfeiting: someone else's product wearing your mark. Trade in counterfeit and pirated goods reached USD 467 billion, or 2.3% of world imports, with the EU alone accounting for USD 117 billion and 4.7% of EU imports (OECD, May 2025). That study appeared in 2025 but measures 2021 trade, so read it as a structural estimate rather than a current-year measurement. EUIPO puts the cost to EU clothing, cosmetics and toy makers at EUR 16 billion in lost sales and roughly 200,000 jobs a year, running at 5.2% of turnover in clothing, 4.8% in cosmetics and 8.7% in toys (EUIPO, 2025, modelled on 2018-2021 data).

Marketplaces have built real machinery against that half of the problem. Amazon reports identifying, seizing and disposing of more than 15 million counterfeit products in 2025, blocking over 99.9% of suspected infringing listings before any brand reported them, and seeing valid brand infringement notices fall roughly 35% since 2020 (Amazon, April 2026). Its Counterfeit Crimes Unit has pursued more than 32,000 bad actors through litigation or criminal referral since 2020, and more than 2.7 billion units have been verified as genuine through Transparency serialization.

The second problem is diversion: your genuine product, sold by someone you never supplied. None of that machinery applies. The goods are authentic, the mark is yours, and the platform will correctly classify the listing as non-infringing. There is no programme for this, only the evidence you build yourself.

Detection: Monitor First, Enforce Last

The recurring mistake is sending notices before any evidence exists. The sequence runs the other way: monitor continuously, shortlist, test-buy, trace lot codes to the channel leak, then act.

Continuous monitoring means three things running at once on your own SKUs. Enumerate every seller name and seller ID offering them. Watch which sellers rotate through the buy box, because an unfamiliar name winning a sale is the earliest signal available. Record listing-level price history, the artefact every later step depends on.

Each method answers one narrow question. Treating what a method proves as what you wish it proved is how brands file reports they cannot support.

Method	Proves	Does not prove
Seller-name and storefront enumeration	A specific legal entity and seller ID is offering your SKU, giving you a roster to diff against your authorized list	Where the stock came from, whether it is genuine, or whether this is a known offender under a new name
Buy-box rotation monitoring	Who is winning the sale right now and at what price, including sellers who surface only intermittently	That the winner is cheapest or the only offender. The buy box weighs fulfilment and performance too
Listing-level price history	A timestamped record of the advertised price, and whether this is a weekend dip or sustained undercutting	Why the price is low. Deep discounts fit diversion, liquidation and an authorized promo equally
Batch and lot-code tracking	Which of your own accounts or distributors the unit shipped from, identifying the leak rather than the symptom	Anything, if you never recorded lot-to-customer mapping at shipment. Codes scrubbed by the seller are themselves a material-difference fact
Test buys	Ground truth: authenticity, packaging, warranty inserts, expiry, condition, return address, invoicing entity	Scale. A test buy costs money and days. It validates a hypothesis, it does not find sellers

On Amazon, an unfamiliar seller taking buy-box rotation from your authorized partners changes the economics for everyone on the listing; the Amazon repricing and Buy Box guide covers those mechanics.

Marketplaces are also not the whole picture. Diverted stock lands on independent Shopify and BigCommerce storefronts, where no seller directory exists and enumeration means finding the store first, then monitoring its product pages directly. Most brand-protection coverage skips that case. If you are weighing building this against buying it, the build versus buy comparison sets out the trade-offs, and brand-side monitoring covers the watch list.

Authorized but Selling Badly Is a Different File

Before anything reaches a report, split your flagged sellers into two piles, because they need different responses and mixing them is expensive.

Four questions do the sorting. Is the entity on your reseller roster under that exact legal name? Does the fulfilment method match your distribution, for example a merchant-fulfilled offer on an FBA-only SKU? Does the test-buy invoice trace to an account you supply? Does the lot code map to a shipment you can identify? Only the last two are conclusive; the first two are triage.

An authorized reseller pricing below your policy is a compliance matter, not an identity one. Minimum advertised price is the floor you allow a reseller to advertise at, distinct from what they charge at checkout, and it is enforced through your agreements rather than a platform. Once you know who a seller is, MAP monitoring is how you prove the pattern is sustained rather than a one-off. Promotional peaks deserve extra attention, since diverted stock stops standing out when everyone is discounting; the peak-season playbook explains how to hold a baseline through November.

Getting this split wrong carries a compounding cost that almost nobody mentions. Rejected reports pull down your Report a Violation acceptance rate. Amazon's Project Zero requires an active Brand Registry as registered trademark owner, use of Report a Violation within the last six months, and at least a 90% acceptance rate on those reports (Amazon, 2026). File a run of unsupported reports and you can lose eligibility for the one programme that would let you remove genuinely infringing listings yourself, without opening a case.

The Test Buy Is Where the Case Starts

The test buy is the only step that produces physical evidence, and brands skip it because it costs money.

Buy from each shortlisted seller and record what arrives, in writing and with photographs: packaging and print quality, warranty card and registration insert, expiry dates, serial or batch marks, the return address on the parcel, and the legal entity on the invoice. That record is what a material-difference or quality-control claim is built from, and it also tells you whether counterfeits are involved at all.

Then read the lot code on the unit and map it back to the shipment it left your warehouse on. This is the only method that points inside your own channel rather than at the symptom, and it turns "someone is undercutting us" into "distributor number four is selling into the grey market". With no lot-to-customer mapping at shipment, the code tells you nothing, so start recording that mapping now: it only becomes useful months later.

What the Four Platform Programmes Actually Do

Programme	What it gives you	Removes a merely unauthorized seller?
Amazon Brand Registry	Prerequisite for everything else. Requires registered trademark ownership. Gives listing control and the Report a Violation tool	No
Amazon Project Zero	Automated protections plus self-service counterfeit removal, so you pull listings directly without submitting a case. Needs Brand Registry and a 90% report acceptance rate	No
Walmart Brand Portal	Copyright, trademark, patent, publicity and counterfeit claims, filed under penalty of perjury, with status tracked in-portal	No
eBay VeRO	Free reporting for IP rights owners via a Notice of Claimed Infringement or the VeRO Portal	No

The right-hand column is the whole point. Each channel asks you to assert an intellectual property right and to stand behind that assertion, in Walmart's case under penalty of perjury. None has a field for "this seller is not in our dealer network".

The one Amazon lever that acts on the unit rather than the listing is Transparency, which puts a unique scannable code on each item so non-enrolled units cannot be listed or sold. Project Zero's automated protections and self-service removal are free; Transparency codes are paid, with volume discounts. Serialisation changes what an unauthorized seller is physically able to list, which is the closest thing to a structural answer.

The Legal Ground Under the Word "Unauthorized"

What follows is general information, not legal advice. Distribution law is jurisdiction-specific and fact-specific, and every point below turns on your own contracts and procedures. Confirm all of it with counsel qualified where you sell before acting.

First sale, and the two exceptions that matter

In the United States, a lawful purchaser may generally resell genuine trademarked goods. The doctrine runs back through Prestonettes, Inc. v. Coty, 264 U.S. 359 (1924) and Champion Spark Plug Co. v. Sanders, 331 U.S. 125 (1947), where reconditioned plugs could still be resold under the mark provided they were clearly labelled as repaired or used. Disclosure, not prohibition, was the remedy.

The first exception is quality control. In El Greco Leather Products Co. v. Shoe World, Inc., 806 F.2d 392 (2d Cir. 1986), goods made under licence but distributed without the mark holder's authorization and without its contractual pre-shipment inspection were held not to be genuine. What is protected is your right to control quality. Courts look for procedures that are legitimate, substantial and non-pretextual, actually followed by you, and not followed by the seller. A policy written for the argument and never applied is worth nothing.

The second is material difference. Société des Produits Nestlé, S.A. v. Casa Helvetia, Inc., 982 F.2d 633 (1st Cir. 1992) concerned grey-market Venezuelan-made Perugina chocolates sold in Puerto Rico, and turned on whether material differences between the authorized and unauthorized product create a likelihood of confusion. Warranty coverage, customer support and post-sale benefits have all counted. Claims run under Lanham Act §32 and §43(a), 15 U.S.C. §1114 and §1125(a).

Why US brands keep MAP policies unilateral

United States v. Colgate & Co., 250 U.S. 300 (1919) established that a manufacturer may announce prices and unilaterally refuse to deal with those who do not observe them, because Sherman Act §1 reaches agreements rather than unilateral conduct. Monsanto Co. v. Spray-Rite Service Corp., 465 U.S. 752 (1984) marks the line: a §1 claim needs evidence of a meeting of the minds, that the distributor communicated acquiescence and that the manufacturer sought it. Leegin Creative Leather Products, Inc. v. PSKS, Inc., 551 U.S. 877 (2007) moved resale-price-maintenance agreements from per se illegality to the rule of reason federally, but several states still treat them harshly under state law. That is why national brands publish policies instead of negotiating them.

The EU lever most English-language guides miss

For a European brand, the strongest tool is selective distribution. Regulation (EU) 2022/720 (EUR-Lex, June 2022), in force since 1 June 2022 and running to 31 May 2034, makes restricting a buyer's ability to set its own resale price a hardcore restriction under Article 4(a). Article 4(c) does something else entirely: a supplier operating a selective distribution system in a territory may restrict its members, and their customers, from active and passive sales to unauthorised distributors inside that territory.

That provision is what makes "unauthorized" a legally meaningful status rather than a commercial preference. It does not arrive by default. It requires an actual selective distribution system with criteria applied consistently, and whether yours qualifies is a question for your counsel.

Running the Monitoring Layer Without a Brand-Protection Budget

The workflow above rests on one thing being in place first: a continuous, timestamped record of who sells your SKUs and at what price. That is a monitoring problem, and starting does not require an enterprise contract.

Respot tracks product URLs you paste in and records price and stock changes per variation rather than at parent-product level, which matters when only one colour or size is being diverted. Extraction is browser-free across Shopify, WooCommerce, BigCommerce, Magento and marketplaces, so one watch list covers an Amazon listing and the independent storefronts selling the same SKU. Checks run on an adaptive cadence with a floor that tightens on paid plans, and price alerts fire on both price and stock changes.

The free plan covers 5 trackers with a 7-day price history and no credit card, enough to instrument your most-diverted SKU and see whether the pattern is real. Paid plans run from 100 trackers up to 2,000, with 90-day history on Pro and unlimited on Agency. History length matters here: a price series is evidence only if it reaches back far enough to show a pattern.

Do the Work in This Order

The brands that clear their listings arrive with a file, not with volume.

Start monitoring before you are angry about anything, because the record you need in month six can only be created in month one. Diff the roster against your authorized list, and be honest about which flagged sellers are your own partners misbehaving. Spend money on test buys only for the handful that survive triage. Read the lot codes and find out whether the leak is a distributor you already have a contract with, because it usually is.

Then pick a channel. With a counterfeit, a material difference or a bypassed quality-control procedure, the platform programmes exist for you. With a genuine product sold outside your network at a price you dislike, the remedy lives in your distribution agreements and, in Europe, in your selective distribution system. Knowing which of the two you hold is the whole decision, and monitoring is what tells you. Start tracking who is selling your products before you need the record.