Someone types your company name into Google. They already know you. They were always going to land on your site and buy. Then, above your own listing, an ad appears. They click it, it routes them straight to your checkout through an affiliate's tracking link, and they buy exactly what they were going to buy anyway. Your affiliate dashboard records a conversion. You pay a commission.
That is brand bidding, and it is the quietest way an affiliate program can lose money while every report says it is winning.
What brand bidding actually is
Brand bidding happens when an affiliate buys paid search ads on a merchant's own trademarked terms: the company name, product names, the brand plus a word like "discount" or "login." The affiliate's ad intercepts a searcher who was already looking for the brand specifically. The click goes through an affiliate link, the sale completes on the merchant's site, and the affiliate collects a commission on a customer the brand had already won.
The line here matters, because brand bidding gets lumped in with general affiliate fraud and it is a narrower, stranger thing. Cookie stuffing drops a tracking cookie on a browser that never saw a real ad. Click flooding fires fake clicks at scale. Brand bidding does none of that. The click is real. The ad is real. The customer is real. What is fake is the idea that the affiliate created anything.
There is also a more aggressive cousin worth naming separately. In classic brand bidding, the affiliate at least sends the searcher to their own page first. In ad hijacking, the affiliate copies the merchant's own paid ad almost line for line, same headline, same description, sometimes the same display URL, so the searcher cannot tell it apart from the brand's official listing. The click routes immediately through the affiliate's link to the brand site. To the customer it looks like the brand's ad. To accounts payable, it looks like an affiliate earned a sale. BrandVerity, a paid search monitoring firm, describes the hijacker's ad as deliberately mimicking the brand's headlines, descriptions and display URL so it blends in.
Why the numbers lie
The mechanism that makes brand bidding pay is last-click attribution. Most affiliate programs credit the conversion to whoever set the final tracking cookie before purchase. A brand bidder's ad sits at the very bottom of the funnel, one click before checkout, on a customer with the highest possible intent. That is the single most flattering position in the entire journey, and it is the position that requires the least actual marketing.
So the affiliate looks like a top performer. High click volume, an unusually high conversion rate, low cost per acquisition. A program manager scanning a dashboard sees a partner that converts and is tempted to send more budget their way. The data trains the brand to reward the behavior.
None of it is incremental. Incrementality asks one question: would this sale have happened without the partner? For a customer who searched your exact brand name, the honest answer is almost always yes. They knew you, they wanted you, they were typing your name. The affiliate did not introduce them or persuade them. It stepped in front of a door the customer was already opening.
The cost is double. First, the commission itself, paid on revenue that was already yours. Second, your own paid search team is now bidding against the affiliate for a keyword you own, which pushes up the cost per click on your brand term. mFilterIt, a fraud detection vendor, notes that affiliate bids on brand keywords move in lockstep with the cost per click on those terms. You end up paying twice to win a customer who was free.
Tapfiliate, an affiliate software vendor, illustrates the scale with a worked example: an affiliate driving 400 brand-search conversions a month at a 25 dollar commission, with 60 percent of those intercepted from traffic that was already converting, costs roughly 6,000 dollars a month in commissions for nothing. That figure is a hypothetical, not a measured average, but the shape is right. A single brand bidder, undetected, is a five-figure annual leak hiding inside a metric that looks like success.
The trademark line
Here is where most people guess wrong. The instinct is that bidding on a trademark must be illegal. It generally is not.
US courts have repeatedly held that buying a competitor's trademark as a search keyword, on its own, is not trademark infringement. In a long-running dispute, 1-800 Contacts discovered in 2005 that Lens.com ads were appearing on searches for its brand; in 2013 the Tenth Circuit found that purchasing a rival's trademark as a keyword does not by itself create trademark liability. The principle held firm in October 2024 when the Second Circuit ruled that Warby Parker bidding on 1-800 Contacts' marks in search auctions was, in the court's framing, a standard and permissible practice that could not support an infringement claim without something more.
That "something more" is the real line. Trademark trouble starts when the trademark shows up in the ad text or the visible URL, or when the ad implies an affiliation or endorsement that does not exist. Google's own advertising policy splits exactly here: a trademark can almost always be used as a keyword, but Google will restrict a trademark that appears in ad copy if the trademark owner files a complaint. Resellers and genuine informational sites get an exception for ad text, which is precisely why coupon sites can often run "BrandName coupons" in their headlines without falling foul of the policy. One regional wrinkle: in the EU and EFTA, Google may restrict a trademark even as a keyword, which is stricter than the global rule.
So the legal floor is low. An affiliate bidding on your brand name as a keyword, with clean ad copy, is usually not breaking trademark law. What they are very likely breaking is your contract. Brand bidding is governed by program terms, not the courts. Almost every serious affiliate program prohibits it, or restricts it to specifically approved partners, in the agreement an affiliate accepts when they join. The trademark question is a sideshow. The program agreement is the actual instrument of control, and it works because the merchant can simply reverse the commission and remove the partner.
How merchants police it
You cannot catch brand bidding from inside your affiliate dashboard. The dashboard shows you a partner with great numbers. It does not show you the Google ad that produced them. Detection happens on the search results page.
The crude version is manual: search your brand name, your product names, and your common brand-plus-modifier phrases, then look at the paid results. The problem is that brand bidders are built to dodge exactly that check. They use geo-targeting to show the ad only where the merchant is unlikely to be watching, and dayparting to run it only at hours when no one is looking. The Search Monitor, a paid search monitoring service, describes bidders restricting ads to off-hours or regions outside a brand's primary monitoring locations. Check from one office at 10am and you will see a clean SERP while the ad runs all evening in another country.
This is why the policing function is mostly automated. Tools such as BrandVerity, The Search Monitor and Bluepear crawl search results continuously across many locations, devices and times of day, and flag any ad that appears on the brand's protected terms. They trace the full redirect chain from the ad's display URL to the final landing page, which is how an affiliate link hiding behind a string of redirects gets unmasked.
Inside the program, a few signals corroborate what the SERP crawl finds: a sudden traffic spike from one affiliate, a conversion rate far above the program average, a high share of conversions with blank or missing referrer data. None of those prove brand bidding alone. Together with a captured screenshot of the ad, they build the case.
Then there is the policy layer. Networks bake this in. Awin, which absorbed ShareASale and fully retired that brand at the end of 2025, gives a merchant a PPC compliance tool to register protected keywords and flag affiliates who bid on them. CJ Affiliate operates a trademark policy where the merchant submits protected terms and can authorize specific publishers for trademark-plus bidding while everyone else is blocked. The enforcement ladder is consistent across networks: a commission reversal, often with no warning required, then a documented written notice with the screenshot as evidence, then removal from the program and, if the ad copy crossed the trademark line, a complaint to Google. Monitoring without that enforcement ladder is theatre. The detection only matters if the penalty is real and the affiliate knows it.
The grey areas
Brand bidding would be a simple problem if every case were a hijacker in another country running your cloned ad at midnight. Most cases are not that.
The most common one is genuinely innocent. An affiliate runs broad paid search campaigns on generic, non-brand terms, the kind of upper-funnel keywords a program might actually want. Broad match and automated campaign types then quietly expand that targeting and start matching the merchant's brand terms, because the search engine's algorithm found them relevant. The affiliate never chose to bid on the brand. They simply failed to add the brand terms as negative keywords. BrandVerity has noted that affiliates caught in monitoring are typically not deliberate brand bidders at all; they just neglected to negatively match the brand terms the program specified. The fix is a stern email and a required negative keyword list, not a ban. For a while affiliates had a real excuse here, claiming that Performance Max would not let them exclude brand terms cleanly; Google has since made brand exclusions and negative keywords easier to apply in those campaign types, which removes the excuse and makes the email less forgiving.
Typo and close-variant bidding is murkier. An affiliate does not bid on "Acme" but on "Acmee" or "Acme.cm," a common misspelling. They will argue they are catching fat-fingered searchers the brand's own exact-match campaign misses. Sometimes that is even true. But a misspelled brand term is still a brand term, the searcher still had brand intent, and a well-written program policy lists meaningful misspellings as protected terms precisely to close this door.
Brand-plus-modifier is the daily argument. "Acme coupon," "Acme review," "Acme discount code," "Acme vs." A coupon affiliate insists they should be allowed to bid on "Acme coupon" because that is their entire business. The merchant counters that the searcher already decided on Acme; the modifier just means they want a deal, and the affiliate is being paid to hand them one the brand never offered. There is no universal answer. It is a deliberate policy choice, and the better programs write it down term by term rather than leaving it to a fight after the commission has been paid.
And then the case that is not really an affiliate problem at all: a competitor bidding on your brand. That is legal in most of the world, as the Warby Parker ruling confirmed, and the only defense is to keep bidding on your own brand term to hold the top spot. The answer, though, is not always to lock everything down. The same 1-800 Contacts saga produced a Federal Trade Commission case arguing that the company's settlement agreements with rivals, which restricted bidding on each other's trademarks, were anticompetitive because they kept cheaper, truthful ads away from consumers. In 2021 the Second Circuit vacated the FTC's order, holding that the agency had used the wrong standard and had underweighted the procompetitive value of trademark enforcement. The episode is still a useful reminder: brand search restrictions can tip from sensible program hygiene into restraint on legitimate competition, and the distinction matters.
What it comes down to
Brand bidding is not loud. It does not look like fraud on a dashboard. It looks like your best affiliate. That is the whole danger. A channel measured on last-click conversions will quietly reward the partners who do the least work and intercept the most certain customers, and brand bidding is that pattern in its purest form.
The defense is not exotic. A program agreement that names the protected terms, including the misspellings and the brand-plus phrases the merchant has decided to protect. Paid search monitoring that crawls the results page from many places and times, because that is the only place the evidence lives. An enforcement ladder that actually reverses commissions and removes repeat offenders. And, underneath all of it, a measurement habit that keeps asking the harder question: not who got the last click, but who actually brought a customer who would not have come on their own. Get that question right and brand bidding stops looking like performance. It starts looking like the bill it is.
Council summary
This post argues that brand bidding is a measurement failure dressed as a star affiliate: last-click attribution rewards the partner who intercepts a customer the brand had already won, and the legal system offers little help because bidding on a trademark as a keyword is generally lawful. The fix sits in program contracts and continuous paid search monitoring, backed by an enforcement ladder that actually reverses commissions. Council checks corrected the case timeline: 1-800 Contacts discovered the Lens.com ads in 2005, the Tenth Circuit ruled in 2013, and the Second Circuit affirmed the Warby Parker decision in October 2024. The FTC passage was sharpened to state that the Second Circuit vacated the FTC order in 2021 on the wrong-standard ground, and the ShareASale reference was updated because Awin retired that brand at the end of 2025. The reader takeaway: stop scoring affiliates on the last click and start asking who actually brought a customer who would not have come on their own.
Comments