I've been scanning SaaS brand terms for affiliate hijacking for the past month. Here's what I caught, and why your spot checks are lying to you.
I manage brand-protection scans as a side project — sampling brand keywords on Google across multiple countries, desktop + mobile, several times a day, and logging every ad that shows up. Over the past month I ran 6 SaaS affiliate programs through it. Sharing what came out because I see the "are my affiliates brand bidding?" question here every few weeks.
What I caught:
Catch #1 — competitor trademark abuse. A major funnel-builder's brand name, searched from one US city: top ad belongs to its biggest rival, with the victim's trademark right in the headline ("[Rival] is Much Better | [Brand]"). Here's the part that matters: the same hour, the same query from two other US cities showed nothing. If the brand's team searched from their own office, they'd see a clean page — forever.
Catch #2 — coupon-site interception. "[VPN brand] coupon", UK only: a coupon aggregator running paid ads with the brand's trademark in the title ("Daily Tested Coupons — Exclusive [Brand] Deal"). Textbook interception: the buyer already decided, the coupon site catches them at the last step, cookie gets set, brand pays commission on a sale it already owned.
The other 4 brands came back clean — and here's why that means less than you'd think.
Ad serving is probabilistic. Even when a hijack campaign is live, a single search has maybe a 10-20% chance of being served that specific ad (Google decides per-query whether to show ads at all, and which ones). Add the evasion staples — dayparting (run 2-6am in the brand's timezone), geo-targeting (every country except where the brand's team sits), mobile-only — and the math is brutal: a weekly manual check catches a competent hijacker approximately never. You need dozens of samples across geos and devices before "clean" starts to mean something.
If you find one, what actually works:
- Trademark in the ad copy → Google's trademark complaint form. This works and gets ads pulled. Pure bidding on your term with no trademark in the copy is NOT a policy violation — you can't take that down, only outbid or out-negotiate it.
- Rogue affiliate → your network needs the affiliate ID from the redirect chain plus a timestamped screenshot. "My CPCs went up" gets ignored; a capture with the click-through chain gets someone's commissions cancelled.
- Auction Insights is better than nothing but only shows domains that entered your auctions — no ad copy, no screenshots, nothing you can file.
Happy to answer questions on method. And if you want your brand terms checked, drop the brand below or DM me — I'll run it through the pipeline for a week and send you whatever it catches, free. I'm collecting war stories at this point and clean results are data too.
Submitted September 17, 2026 at 08:52AM by DrRoseLee123
via reddit https://ift.tt/5rcRC8y
