A spare part is the perfect counterfeit product. The buyer cannot judge quality by looking, the listing borrows all its credibility from a part number, and the proof of failure arrives weeks later, inside a machine, where a frustrated owner concludes the original brand makes junk.
Marketplaces know the pattern and genuine manufacturers know it better. What follows is how these listings are built, the tells that survive a quick scroll, and the enforcement routine that keeps them down instead of playing whack a mole with the same seller under new names.
Counterfeits return after takedowns because the listing is the cheapest part of the operation to replace. The seller accounts, image files, supplier, and fulfilment route all survive, so a new listing costs minutes. Enforcement that only files takedowns wins each page and loses the pattern; breaking the cycle means targeting accounts, payment routes, and repeat infrastructure.
A visible watermark stops the laziest and most common image theft: a copycat lifting your listing photos wholesale to sell a lookalike. It does not stop an operation that buys one genuine unit and photographs it. Choose watermarking based on which theft your reports actually show, and consider invisible fingerprinting and monitoring for the rest.
Image theft splits into kinds. Listing scrapers copy your photos to sell a product they may not even ship. Competitor adjacent sellers lift lifestyle shots for their own genuine but competing goods. Counterfeiters selling a physical fake usually need only one genuine purchase to generate unlimited original photography, which is why watermarked brands still find flawless counterfeit listings with unmarked photos.
Pull your last quarter of infringement reports and sort by image source. If most offending listings use your actual photos, watermarks and fingerprinting address the bulk of your problem. If most use original photos of fake goods, image tools are a side issue and the money belongs in test purchases and account level reporting.
Watermarks cost conversion at the margin: they sit on the product in the customer's eye, complicate marketplace image compliance, and some platforms reject or rank down visibly marked images. Heavy marks also punish your own channels, wholesale partners, and press, who all need clean assets. The usual compromise is placement and weight: a light mark across a non-critical area, or marks only on the hero images scrapers favour most.
Invisible fingerprinting flips the trade: embedded identifiers survive crops and edits, prove ownership in a dispute, and cost the customer nothing visually. They deter nobody, because nobody can see them, but they make takedown evidence faster and harder to argue with. Many brands run both: light visible marks on marketplace feeds, fingerprints everywhere.
Neither mark matters without detection. Reverse image search monitoring on a schedule, weighted to launch periods when scraping spikes, turns marks and fingerprints into reports. New product launches are the exposure window: the first weeks of clean, high resolution imagery are what scrapers harvest for the season.
Keep the original files, unmarked masters, and the fingerprint registry organised by product and date. When a dispute asks you to prove an image is yours and when you published it, the answer should be a file pull, not a memory. That registry also protects you the day a platform asks the same question about a listing of your own.
Keep reading: Why Counterfeits Keep Coming Back After a Takedown covers a nearby case in this series.
It stops listings that reuse your photos, which is a large share of casual copycat activity. It does not stop sellers who photograph a genuine unit themselves; those cases are fought with test purchases and account level evidence.
Rules vary and change: some restrict promotional text and logos on main images specifically. Check each platform's current image policy per placement; a common setup is clean main images with marked secondary or lifestyle images where allowed.
An identifier embedded in the image file or pixel data that survives ordinary editing, letting you prove a copied image came from your asset. It supports takedowns and disputes rather than deterring theft visibly.
On a regular cycle and always around launches and promotions, when scraping peaks. Automated monitoring beats occasional manual checks once a catalogue passes a size one person can hold in memory.