Skip to content
Takedown guides·2026-07-11·8 min read

What goes in a counterfeit takedown evidence pack (and why platforms reject weak ones)

Marketplaces reject far more counterfeit reports than they act on, and almost always for the same reason: the evidence was generic. Here's the universal checklist a takedown needs, where Amazon, eBay and Walmart diverge, and how to package it so a reviewer can act in one pass.

Brand Protector teamOperational research

You can be completely right that a listing is counterfeit and still get the report bounced. Marketplace IP reviewers work at volume, against a checklist, and they reject anything that makes them do investigative work you should have done. The pattern behind rejected reports is boringly consistent: the evidence was generic. This is the anatomy of a report that isn’t — the elements every platform expects, where the big three diverge, and how to package it so one reviewer can act in one pass.

One note before the checklist: this is operational guidance, not legal advice. A counterfeit takedown is a sworn legal assertion, and § 512(f) and its equivalents attach real liability to filing in bad faith — read what § 512(f) means for your takedown program before you build a filing habit, and involve counsel on anything fact-specific.

Why do platforms reject counterfeit takedowns?

Almost every rejection traces to one of a small set of avoidable defects. None of them are about whether you’re right — they’re about whether the reviewer can act on what you sent:

  • No ownership basis.You asserted the brand but didn’t cite a registered right. “This is our product” isn’t a claim a reviewer can verify; a trademark registration number is.
  • Wrong claim type.Counterfeiting, an unauthorized (but genuine) reseller, and a MAP-pricing dispute are three different problems with three different intake paths. Filing a distribution dispute as counterfeiting gets rejected — and can put your own account at risk. If the goods are real but the seller isn’t approved, that’s an unauthorized-seller problem, not a counterfeit one.
  • Missing or ambiguous identifier.A URL alone is fragile — listings move and get edited. Reviewers want the platform’s own handle for the item (ASIN, eBay item number, Walmart product ID) so there’s no doubt which listing you mean.
  • Bundling. Twenty listings in one report forces the reviewer to adjudicate twenty things at once; many intake forms are built around one identifier per claim. One listing, one report is the safe default.
  • Un-anchored screenshots. A cropped product photo proves nothing. A capture that shows the URL and the date proves the listing existed in the infringing state when you filed.

What’s the universal evidence checklist?

Regardless of platform, a counterfeit report is built from the same elements. Assemble all of these before you open any intake form:

  1. Proof of the right you’re enforcing. A registered trademark or copyright, cited by registration number and jurisdiction (USPTO, EUIPO, and so on). For image-theft claims, the copyright registration or clear authorship of the original product photography.
  2. The exact listing URL. The canonical link to the infringing listing — not a search results page, not a storefront.
  3. The platform’s own identifier. ASIN, eBay item number, Walmart product ID, or equivalent. This is the field reviewers key on; the URL is corroboration.
  4. A specific infringement statement.What is counterfeit and how you know — your mark used without authorization, your registered product images copied, a logo reproduced. Name the element; don’t just assert “fake.”
  5. Dated screenshots of the live listing. The product page with the URL visible and a capture date, so the infringing state is preserved even if the seller edits or pulls the listing after you file.
  6. Seller and offer details. The seller handle or storefront name and the offer specifics (price, condition, shipping origin where shown) — useful for both the reviewer and your own reappearance tracking.
  7. A good-faith declaration. The sworn statement, made by the rights holder or an authorized agent, that the report is accurate and made in good faith. Most platforms bake specific wording into their form; the liability behind it is real.
  8. Rights-holder contact information. A working contact for the rights owner so the platform (and, on a counter-notice, the seller) can route correspondence.

A side-by-side of the genuine item against the counterfeit — packaging, logo placement, image reuse — isn’t always required, but it’s the single most persuasive optional addition when a reviewer is deciding a close call.

Where do Amazon, eBay and Walmart differ?

The core elements are constant; the intake path and its required fields are not. A report assembled for one platform is not drop-in valid for another. The table below sketches the divergences as of verified 2026-07-11— always confirm the current requirement against each platform’s own IP-policy page before filing, since intake flows change without notice.

PlatformIntake pathIdentifier / entry gate
AmazonBrand Registry “Report a Violation,” or the public reporting formASIN; the streamlined path requires Brand Registry enrollment against a registered mark
eBayVeRO Notice of Claimed Infringement (NOCI)eBay item number(s); VeRO member enrollment establishes the rights-owner relationship
WalmartWalmart Marketplace IP infringement form (Brand Portal where enrolled)Walmart item / product ID; rights-ownership details on the claim

Each has a deeper guide of its own — removing counterfeit Amazon listings, eBay’s VeRO program, and Walmart Marketplace takedowns — and Amazon’s enrollment path has its own failure modes, covered in reading a Brand Registry rejection email. Treat the table as a map of where the packs diverge, then use the platform-specific guide for the exact fields.

Screenshots, timestamps and the reappearance record

Screenshots are where most homemade evidence packs are weakest, and where the difference between “acted on” and “rejected” often lives. Two rules do most of the work.

First, capture context, not crops. The screenshot should show the full listing with its URL bar and a capture date. A product image lifted out of the page could have come from anywhere; a dated full-page capture is a record that this listing existed in this state at this time. That matters because listings are mutable — a seller who senses a report coming can swap images, edit the title, or pull the listing, and if your only evidence was a crop, you’ve got nothing to point at.

Second, some platforms require the capture as proof of active commerce. TikTok Shop’s IP intake, for example, wants a full-page screenshot of the live listing. Because origin CDNs frequently block automated fetches, capturing a clean full-page image reliably is genuinely hard to do by hand at volume.

The capture is also what makes reappearance tracking possible. Counterfeit sellers relist — same product, new listing ID — and a dated evidence record of the original is what lets you recognize the relist as a repeat rather than treating it as a fresh, unrelated detection.

Packaging it: the evidence pack and PDF export

Brand Protector builds this pack per detection so you’re never assembling one by hand under time pressure. For a confirmed detection it generates a per-platform package: the listing facts and identifier, the pre-filled complaint text stamped with your legal entity and registered-rights context, the seller and offer details, and — where the platform requires active-listing proof — a screenshot.

Screenshot capture runs asynchronously. Rather than blocking the draft on a slow, block-prone fetch, the capture is kicked off and polled in the background, so drafting a takedown on a platform like TikTok Shop doesn’t hang while a proxied screenshot resolves. Capture is best-effort: if it fails, the draft still completes with the product thumbnail as a fallback and the failure is logged rather than swallowed. A captured screenshot that’s gone stale is re-captured when you draft again.

The whole pack exports as a brand-styled PDF evidence pack— a cover page with the takedown subject and identifier, the detection facts, the complaint body, and an attestation footer recording what was confirmed and by whom. It’s the artifact you attach when a platform prefers a formal document, or hand to counsel. (Listing imagery is delivered as separate downloads rather than embedded, because origin CDNs block server-side image fetches often enough that embedding produced broken evidence packs — so the reviewer attaches the thumbnail or full-page screenshot directly to the platform’s own form, which is what every IP intake expects anyway.)

One thing the pack does notdo is file itself. Every takedown clears the triple-validated gate — AI confirmation, then a human reviewer who must actually read the evidence, then an admin attestation that confirms the exact listing identifier and swears the good-faith statement — before anything is submitted. The evidence pack makes that final step fast and well-founded; it never replaces it. And a complete pack, however good, doesn’t guarantee removal — the platform decides, and sellers can counter-notice. What a strong pack guarantees is that the decision turns on the merits, not on a missing field.

Frequently asked questions

What evidence do you need to report a counterfeit listing?

At minimum: proof you hold the right (a registered trademark or copyright, cited by registration number), the exact listing URL and the platform's own identifier for it (ASIN, item number, product ID), a specific statement of what infringes and why, dated screenshots of the live listing showing its URL, and a good-faith statement from the rights holder. Generic 'this is fake' reports without an identifier and an ownership basis are the ones that get rejected.

Do you need a test purchase to report a counterfeit?

Usually no. Most marketplace IP intake forms accept a listing-based report — URL, identifier, ownership proof, and a good-faith statement — without a physical buy. A test purchase becomes relevant when your claim rests on physical authentication (the item in hand is materially different from genuine) rather than on the listing itself using your mark or images. Treat test-purchase decisions as fact-specific and get counsel involved before any evidence-gathering that could carry legal risk. This is not legal advice.

Why do platforms reject counterfeit reports?

The common rejections are structural, not subjective: no rights-ownership basis stated, the wrong claim type (reporting a pricing or authorized-reseller dispute as counterfeiting), a missing or ambiguous listing identifier, one report bundling many listings so the reviewer can't act cleanly, or screenshots with no visible URL or date. Each is fixable before you file.

What proof of trademark ownership do platforms require?

Typically a registered mark — the registration number and jurisdiction (USPTO, EUIPO, etc.) — or a copyright registration for image-theft claims. Programs like Amazon Brand Registry require enrollment against a registered mark before you can use their streamlined report path at all. Cite the registration, don't just assert the brand. Verify each platform's current requirement against its own IP-policy page before filing.

Do screenshots need timestamps, and do they matter?

Yes. A screenshot that shows the listing URL and the capture date is far stronger than a cropped product image, because listings change and disappear — the counterfeit you reported may be edited by the time a reviewer looks. A dated, full-context capture proves the listing existed in the infringing state when you filed, and it anchors reappearance tracking if the same seller relists.

Is the evidence the same on Amazon, eBay and Walmart?

The core is the same — ownership, identifier, infringement statement, good-faith declaration. The packaging differs: Amazon routes rights owners through Brand Registry's Report a Violation flow, eBay uses the VeRO Notice of Claimed Infringement, and Walmart uses its IP infringement form. Each has its own identifier field and claim taxonomy, so a report built for one isn't drop-in valid for another. Verify each platform's current intake against its own policy page before filing.

Does a strong evidence pack guarantee the listing gets removed?

No. A complete, well-formed pack removes the avoidable reasons for rejection and lets a reviewer act quickly — but the platform makes the decision, sellers can file counter-notices, and outcomes vary by platform and claim. Anyone promising guaranteed removal is overselling. What you control is the quality and completeness of what you submit.

Can Brand Protector build the evidence pack for me?

Yes. For each confirmed detection it assembles a per-platform pack — the listing facts, identifier, the pre-filled complaint text stamped with your legal entity, and an asynchronously captured screenshot where the platform requires active-listing proof — and exports the whole thing as a brand-styled PDF. Every takedown still clears the triple-validated human-and-attestation gate before anything is filed; nothing auto-submits.

Stop filing takedowns by hand.

Daily scans find the listings, evidence packs are built for each platform's intake, and a triple-validated gate reviews every notice before it's filed.

7-day free trial · card required, no charge until day 8 · cancel in-app