How to Find Counterfeit Listings at Scale

Counterfeit listings are a geography problem before they are a detection problem. Sellers frequently restrict who can see a listing, and infringing inventory that is invisible from your head-office IP is often perfectly visible from a consumer connection in the market being targeted. Finding it at scale means looking from the right places.

Why listings hide

Marketplaces localise search results, and sellers can limit shipping or visibility by region. A brand monitoring from one country therefore sees a systematically incomplete picture, and the gap is not random — it is precisely the markets where the seller expects less enforcement.

Building the sweep

Run one pass per market with a residential exit in that country, tagging every result with the exit used. Search your brand terms, common misspellings, and the model numbers that counterfeiters reuse.

Credentials are built from a separate variable below rather than pasted inline — a literal user:pass@host string in a page body can be rewritten by email-obfuscation filters.

import requests, itertools

endpoint = "geo.spyderproxy.com:12321"
creds = "USERNAME:PASSWORD"          # country selected via username params
proxy = f"http://{creds}@{endpoint}"

TERMS  = ["brandname", "brand name", "brandnarne"]   # include misspellings
MARKETS = ["us", "de", "it", "mx"]

for term, market in itertools.product(TERMS, MARKETS):
    r = requests.get(
        "https://marketplace.example/search",
        params={"q": term},
        proxies={"http": proxy, "https": proxy},
        timeout=30,
    )
    print(market, term, r.status_code)
    # store results tagged with market + term

Signals that flag a listing

Price well below authorised floor. The most reliable single signal, and easy to automate.
Reused imagery. Official product photography appearing under unauthorised sellers.
Seller newness and volume. New accounts listing deep inventory of a protected product.
Description artefacts. Machine-translated copy, wrong model numbers, mismatched packaging claims.

Treat these as triage, not proof. The output of a sweep is a queue for human review, and a false positive aimed at a legitimate reseller has real costs.

Evidence that survives a takedown

Takedown processes ask for specifics. Capture, per listing: the full URL, seller identifier, price and currency, the timestamp, the exit country you observed it from, screenshots or a full HTML capture, and the shipping destinations offered. A screenshot with no timestamp or point of observation is weak evidence, and listings change or vanish once a seller notices attention.

Which product

Budget Residential at $1.75/GB for breadth across markets, and Premium Residential at $2.75/GB where city-level visibility matters. Marketplaces are among the more aggressive at blocking datacenter ranges, so residential is usually necessary rather than optional.

Frequently Asked Questions

Why can I not see infringing listings from my own country?

Marketplaces localise search results and sellers can restrict visibility or shipping by region. Infringing inventory is often deliberately shown only in the markets being targeted.

What should I capture as evidence?

URL, seller ID, price and currency, timestamp, the exit country you observed from, a full HTML or screenshot capture, and offered shipping destinations. Listings disappear once sellers notice attention.

Which proxy type is needed?

Residential. Budget Residential at $1.75/GB gives breadth across markets; marketplaces block datacenter ranges aggressively.

Can this be fully automated?

The sweep can. The judgement cannot. Automated signals produce a review queue, because a false positive against a legitimate reseller carries real cost.