Should I Bid on My Own Brand Name in Google Ads?

There is no universal yes, and the honest default is "measure it, because the ads dashboard cannot answer this question." The largest published randomized test of brand keyword advertising found that when eBay switched off paid search on brand keywords, almost all of the lost paid clicks and attributed sales reappeared in organic search: substitution was close to complete, and brand keyword ads showed no measurable short-term benefit (Blake, Nosko and Tadelis, "Consumer Heterogeneity and Paid Search Effectiveness: A Large Scale Field Experiment," Econometrica 83(1), 2015, pp. 155-174). That is a real finding about a very strong brand. It is not automatically your finding, and the difference is the whole article.

Why brand campaigns always look good in the ads interface

Start with the measurement problem, because it explains why this debate never resolves.

A brand campaign buys clicks from people who typed your company name. Those people already know who you are. They were going to arrive by some route. When they convert, the conversion is credited to the ad they clicked, because that is how click-based attribution works.

The result is structural, not suspicious: brand campaigns produce the best cost per conversion and the best return in almost every account, by construction. Nothing about that number tells you whether the campaign caused anything.

This has a specific consequence for anyone reporting upward. If your dashboard blends brand and non-brand into one account-level cost per acquisition, then shifting budget into brand improves the headline number while telling you nothing about incremental demand. It is the easiest way in paid search to make a report look better without making the business better, and it is usually done innocently. If your reporting cannot separate the two, that is the first thing to fix, ahead of the bidding decision itself.

What the eBay experiment actually showed, and its boundaries

The study is worth describing precisely, because it gets both overstated and dismissed.

eBay stopped paying for brand keyword search ads on Yahoo and Microsoft while continuing to run them on Google as a control, then measured what happened to traffic and sales. The finding on brand terms was that almost all of the forgone click traffic and attributed sales were captured by natural search results instead. The researchers also examined non-brand keywords and found that new and infrequent users were positively influenced by ads, while frequent users, who accounted for most of the ad spend, were largely unaffected, producing negative average returns for that population. Their broader conclusion was that returns to paid search were a fraction of what non-experimental estimates suggested.

Now the boundaries, which the study's own framing supports and which most people quoting it skip:

  • eBay had a dominant organic presence for its own name. Substitution can be near-complete only if there is something to substitute into. A brand whose own name does not return it as the first organic result is in a different situation.
  • It measured short-term effects. It is evidence about immediate sales, not about long-term brand effects.
  • It is one advertiser, at a particular time, in a particular category. It is the best available evidence of its kind, and it is still one case.

The correct reading is not "brand bidding is a waste." It is: the burden of proof sits with the brand campaign, not against it, and click-attributed conversions do not discharge that burden.

The three situations where paying for your own name has a defensible case

1. Someone else is showing on your name. This is the argument with the most force, and it is grounded in a fact about Google's policy rather than in speculation. Google's trademark policy states plainly that it "will not restrict" the use of trademarks as keywords. The restriction applies to ad text: Google restricts "using trademarks in an ad from a direct competitor" and ads that use the trademark "in a confusing, deceptive, or misleading way," subject to exceptions such as resellers and informational pages. So a competitor can generally target your brand as a keyword even if they cannot write your name into the ad.

Check before you assume. Look at whether anyone actually appears on your brand terms, and use the Ad Preview and Diagnosis tool rather than repeatedly searching your own name. The mirror image of this question, from the other side, is bidding on competitor keywords, and it is worth reading whichever side you are currently on.

2. Your organic result does not say what you need it to say. An ad lets you choose the destination and the message: a current offer, a specific product page, a support page during an incident. If your organic listing lands on a homepage that buries the thing brand searchers want, an ad is a control mechanism rather than a traffic purchase.

3. Your brand name is ambiguous. If your name is also a common word, another company, or a product in a different category, the organic page for your name may not be dominated by you at all. The substitution logic weakens exactly in proportion to how much of that page you already own.

Notice that none of the three is "because brand clicks are cheap." Cheap clicks that would have arrived free are not cheap.

The situations where it is probably not worth it

Nobody else is bidding, and you own the top organic result outright. This is the closest real-world match to the eBay conditions, and it is the case where the study's finding should worry you.

Your brand is genuinely unknown. If almost nobody searches your name, a brand campaign is a rounding error at best and a distraction at worst. New businesses often build a brand campaign because it is the easiest campaign to build. The money belongs on terms describing what you sell.

Budget is the binding constraint. If your non-brand campaigns are already limited by spend, moving money to brand terms means buying clicks from people who found you anyway, funded by the ones who have not heard of you yet. Impression share lost to budget will tell you whether that constraint is live in your account before you have the argument.

⚠ A note on spend risk. Brand campaigns are usually inexpensive per click and easy to leave running for years without review. That combination is how a line item survives every budget conversation without ever being tested. Whichever way you decide, put a review date on it.

How to actually test it in your own account

You do not need a research budget. You need a test that survives being questioned.

The pause test. Turn the brand campaign off for a defined period, and measure total brand-driven outcomes rather than paid ones. That means paid conversions plus organic conversions on branded queries plus direct traffic, over the same window, compared against a similar window before. If total brand-driven conversions hold roughly steady while paid conversions fall, you have observed substitution. If total conversions fall by roughly the amount the paid campaign was delivering, the campaign was doing work.

Conditions that make the result trustworthy:

  • Run it long enough to cover a full sales cycle, not a week.
  • Do not change anything else in the account during the window. If bid strategy, budget or landing pages move at the same time, the test is unreadable.
  • Watch for competitor activity starting or stopping mid-test, because that changes the counterfactual.
  • Choose a period without unusual seasonality or a marketing campaign running elsewhere.
  • Agree in advance, in writing, what result leads to what decision. A test whose outcome is negotiated afterwards is not a test.

What makes this hard, said plainly: you are pausing a campaign that looks excellent in the dashboard, and for the duration of the test your reported numbers will get worse. That is a political cost, not an analytical one, and it is the actual reason most brand campaigns are never tested. Framing it as a fixed-length, reversible experiment with a pre-agreed decision rule is what makes it survivable.

What we are not going to claim

We have not run these accounts and we are not going to invent a result. The mechanism above is drawn from Google's own policy documentation and from a named, published, peer-reviewed field experiment you can look up and read for yourself. What we cannot tell you is what happens in your category, at your brand strength, against your competitors.

The version to take to your manager

Three sentences that hold up under questioning:

Brand campaigns always show the best return in the ads dashboard because they buy clicks from people who already know us, so that number cannot tell us whether the spend is incremental. The one large randomized study on this, at eBay, found the paid clicks were almost entirely replaced by organic ones when the ads were switched off, though that brand had a stronger organic position than ours. I propose a fixed-length pause test measuring total brand-driven conversions, with a decision rule agreed before we start.

That framing concedes the uncertainty, which is precisely why it survives the meeting. It also puts a boundary around the ask: a defined window, a reversible change, and a number that decides it. The same instinct applies to most of the account, which is what the wider look at where a small budget goes keeps circling back to.

FAQ

Can competitors legally bid on my brand name? Google's trademark policy states it will not restrict trademarks used as keywords. What it does restrict is trademark use in ad text by a direct competitor, and ads that use a trademark in a confusing, deceptive or misleading way, with exceptions such as resellers and informational sites. Trademark law itself varies by country, so the platform rule and the legal position are separate questions.

Are brand keywords always cheap? Relevance is one of the inputs to Ad Rank, and you generally pay only what is needed to clear the thresholds and beat the advertiser below you, so brand terms are often inexpensive per click. That is a statement about the mechanism, not a promise about your account, and price is not the same as value when the click may have been free.

If I stop bidding on my brand, will I lose that traffic? The eBay experiment found the great majority of it moved to organic search. How much of it moves for you depends on how completely you own the organic results for your own name, which is the variable the study cannot supply.

Does a brand campaign protect me from competitors on my name? It gives you a presence there. It does not remove theirs, since Google does not restrict trademarks as keywords. Whether outbidding a competitor on your own name is worth the cost is an economic decision, and one worth checking against how much of that traffic you would keep anyway.

How long should a brand pause test run? Long enough to cover a full sales cycle and a normal weekly pattern, with nothing else changing in the account. Short tests on brand terms mostly measure noise.

The eBay finding cited above is from Blake, Nosko and Tadelis, "Consumer Heterogeneity and Paid Search Effectiveness: A Large Scale Field Experiment," Econometrica, vol. 83, no. 1 (2015), pp. 155-174. Trademark policy wording is quoted from Google's advertising policies documentation, checked 2026-08-07. No first-party account performance figures are presented as evidence in this article.

Similar Posts