The Challenge
You manage a consumer brand. Your reseller policy says nobody sells the flagship product below $179. Then a customer emails: someone on Amazon listed it at $144 last weekend. By the time you check, the listing's gone. The seller relisted at the right price the moment they hit their target sales velocity. The damage is done, and you've still got Walmart, eBay, TikTok Shop, and Google Shopping to check.
This is the rhythm of MAP enforcement in 2026. Counterfeit and pirated goods alone hit $467 billion globally (OECD and EUIPO data, via Red Points), and that's before you count gray-market arbitrage and reseller policy violations. The marketplaces don't police price for you. And 79% of counterfeit seizures now come from small parcel shipments, meaning individual sellers slipping past customs at scale.
The brands catching it fast aren't running better spreadsheets. They're running monitoring infrastructure that hits every product page on every marketplace, every hour, in every region they sell into.
The FourA Approach
The naive setup is one scraper per marketplace, one cron job, one alert. Then your scraper breaks because Amazon shipped a layout change, your IPs get blocked because you're hammering the same product from one ASN, or your price extraction silently returns the strikethrough MSRP instead of the actual offer.
A working pipeline has four parts.
Discovery. Start with your SKU list and find every active offer across marketplaces. Amazon ASINs, Walmart item IDs, eBay listing IDs, TikTok Shop product URLs. This is mostly catalog scraping: hitting category pages, search result pages, and PDP variants.
Page collection. For each offer, pull the live page and grab the visible price, the seller name, the buy box owner, any coupon or Subscribe & Save discount, and a timestamp. Some marketplaces serve clean HTML. TikTok Shop and Google Shopping render most of the price logic in JavaScript, so you need a real browser, not a request library.
Price reconstruction. This is where most teams get burned. Amazon shows a list price, a Subscribe & Save discount of 5-15%, sometimes a coupon, sometimes a multi-pack split. The actual price the customer pays is rarely the number in the headline. A monitor that flags only the headline misses the violations hiding in the discount stack.
Evidence and alerts. When you flag a violation, your enforcement team needs proof: URL, screenshot, exact price, timestamp, seller, and the offer's lifespan. Without that, the marketplace dispute process goes nowhere.
A platform like FourA gives you one API for the messy parts. HTTP requests for marketplaces that serve clean HTML, browser sessions for the JavaScript-heavy ones, and proxy routing you don't have to manage. You point a request at an Amazon product URL with the unblocker flag set, and Chrome-equivalent headers go out so the page renders cleanly. You point one at Amazon US through a US-region proxy, then the same SKU at Amazon DE through a DE-region proxy, and the prices come back in local context.
curl -X POST "https://api.foura.ai/api/proxy" \
-H "x-api-key: YOUR_KEY" \
-H "Content-Type: application/json" \
-d '{
"maxTries": 5,
"timeout_ms": 30000,
"request": {
"method": "GET",
"url": "https://www.amazon.com/dp/B0EXAMPLE",
"unblocker": true,
"validate": {
"status": { "accept": [200] },
"data": { "fail": ["captcha", "Robot Check"] }
}
}
}'
That's the whole call. No proxy rotation logic in your code. No "is this site running JavaScript today" check. The infrastructure handles routing and retries; your pipeline handles price logic. For a deeper read on what the unblocker flag actually does on the wire, see our breakdown of Web Unblocker.
The geographic piece matters more than people think. A brand selling into the US, EU, and UK needs price visibility from each region's IP space, because marketplaces sometimes serve regional pricing or hide the offer entirely from out-of-region IPs. Per-region routing means you can verify that a seller advertising "ships worldwide" actually breaks MAP in the markets that count.
Results
A mid-size brand with 500 SKUs, six marketplaces, and three regions ends up with roughly 9,000 product pages to monitor. Hourly coverage is around 216,000 requests per day (illustrative scenario based on a typical mid-market brand-protection scope). That's nothing for an API designed for it, and it's a full-time engineering team if you build it yourself. But the interesting number isn't requests-per-day; it's how often a violation goes live and gets caught the same hour.
In the field, the shape of the improvement looks like this:
- Detection latency drops from days (manual or weekly scrape) to under an hour with hourly polling
- Coverage moves from "the three marketplaces I can keep up with" to all six the brand actually sells on
- False positives drop sharply once price reconstruction handles coupons, S&S, and multi-packs correctly
- Evidence quality improves when each detection arrives with a screenshot and timestamp pinned to the request
If you've already read our breakdown of aggregating real estate listings at scale, the pattern's the same: lots of pages, mixed rendering, geographic variation. The product changes; the infrastructure shape doesn't.
Key Takeaway
MAP violations aren't a data quality problem. They're a clock problem. Whoever sees the violation first has the edge: the brand who catches it before the listing converts, or the reseller who pockets the margin and rotates the price by the time anyone notices. Every layer your monitoring stack adds (geographic precision, JavaScript rendering, price-stack reconstruction) is really just buying back minutes from that clock.
The brands winning this in 2026 stopped treating MAP enforcement as a quarterly audit. They started treating it as live infrastructure. The cheapest violation to fix is the one that surfaces in the same hour it appears.