Should you bid on your own brand name? The evidence, and a way to check
TL;DR
Experiments on brand ads range from no measurable benefit to a strongly positive return, and the one study that separated the cases found competitors on the page made the difference. What to look at first, and what a pause test can and can't show.
Search for a business by name and its own website is usually the first unpaid result. Many accounts also run an ad on that search, and in Google Ads it tends to look like the best campaign there is: cheap clicks and a low cost per conversion.
The open question is how many of those people would have clicked the unpaid result just below. Clicks the ad adds on top of what would have happened anyway are called incremental, and the reports in an ad account can't show them, because a paid click is counted the same whether or not it replaced a free one. We covered the short version before. This is the evidence, and a way to check an individual account.
What the experiments found
The best-known test is eBay's. Blake, Nosko and Tadelis (NBER working paper 20171, later published in Econometrica, 2015) stopped eBay's brand ads on two search engines and found that "almost all of the forgone click traffic and attributed sales were captured by natural search". Their summary was that "brand-keyword ads have no measurable short-term benefits". eBay was one of the most recognised names on the web, and the data is from 2012.
Simonov, Nosko and Rao (Marketing Science, 2018) ran experiments across the 2,500 most searched brands on Bing and split the result by who else was advertising. With no competitor on the page they found "a positive, statistically significant impact of brand ads of 1%–4%, with larger brands having a smaller causal effect", and worked out that "for each incremental paid click, the firm must pay for about 16 clicks that would have been free". Where competitors advertised on the name and the brand itself didn't, the competitors took 18% to 42% of the clicks, and the authors concluded that the return on defending "is strongly positive".
Two wider experiments are a caution against reading eBay's result as general. They paused all search ads, so they say nothing about brand terms on their own. At Edmunds, "more than half of the paid traffic is lost when we shut off paid-links search" (Coviello, Gneezy and Goette, 2017). In Golden and Horton's 2019 experiment a firm that paused its ads in randomly chosen markets had "about 23% fewer customer registrations", which was "only 63% as many" as "a non-experimental estimate would have suggested".
All of this is at least seven years old, comes from large advertisers, and predates AI answers at the top of the results page. It shows the range. It can't say where a given account sits in it.
Google's own figure
Google published its own analysis in 2012. Across 390 studies in which advertisers paused their ads, "50% of the ad clicks that occur with a top rank organic result are incremental", rising to 82% when the unpaid result ranked second to fourth and 96% when it ranked lower. Three limits apply. It covers all keywords, of which brand terms are one part. It measures clicks, and a click isn't a sale. And it is Google's research on the value of Google's ads. The paper behind it is careful on the last point: it recommends that advertisers run "randomized experiments" of their own.
First, see who else appears on your name
The one experiment that separated the cases found that the presence of competitors made the difference, so that is the first thing to look at, and it costs nothing.
Other businesses are allowed to advertise on a brand name. Google's trademark policy lists "Using trademarks as keywords" among the things it "will not restrict". Whether anyone does can be seen in the auction insights report for the brand campaign. Google defines its overlap rate as "how often another advertiser's ad received an impression when your ad also received an impression". The report needs the brand ad to be running, and it "doesn't show insights when impression share is less than 10%", so it has to be read before any pause. A search for the name in a private browser window shows roughly the page a stranger sees.
If other advertisers do appear, the one study of that situation found the brand ad earning its place, which makes cutting it the riskier move. If none appears, the same study found a small lift at a high price per added click, and its authors' advice was to "conduct 'ad pause' experiments".
A pause, and what it can and can't show
Google provides a free report for reading the two sides together. The paid and organic report appears in Google Ads once Search Console is linked, and Google lists one of its uses as "Measure changes holistically": seeing the effect of a change "across text ads, organic results and combined traffic". It "counts performance of text ads only".
The test is to pause the brand campaign for a few weeks and compare combined clicks on brand searches before and during. Three things decide whether the answer can be trusted.
- Whether brand ads really stopped. A Performance Max campaign may also show on searches for the brand. Google offers brand exclusions that "keep your campaigns from serving for branded queries that you want to avoid", and without one a pause of the Search campaign may not be a pause.
- What else changed. The published experiments switched ads off in randomly chosen regions and left them on in others. A before-and-after comparison in one account has no such control, so a promotion, a seasonal dip or a press mention in the same weeks will be mixed into the result.
- Whether there is enough to read. A brand with a few dozen brand clicks a week won't show a clear difference in a fortnight. At that volume the spend at stake is small as well, which is a reason to rely on the first check.
Clicks are half of it. Search Console holds no sales or enquiries, so the business's own count of them over the same weeks is the figure to set beside the clicks.
Reading the result
The figures here are invented to show the arithmetic.
A brand campaign gets 400 clicks a week at 30p each, £120 a week. Unpaid results for the same searches get 900 clicks a week. During a three-week pause the unpaid clicks average 1,240 a week, and enquiries go from 31 a week to 30.
Of the 400 paid clicks, 340 came back through the unpaid result and 60 didn't. The £120 a week was buying 60 clicks the business wouldn't otherwise have had, which is £2 for each added click against the 30p the account reported. Simonov and colleagues call this the "cost per incremental click". Whether £2 is worth paying is the same question as for any other click: what a click is worth to the business.
Had the unpaid clicks risen by only 100, the ad would have been adding 300 clicks a week at 40p each, and the pause would be an argument for switching it back on. One account's version of this overlap is described in a case study.
What we still don't know
The experiments cited here all use data more than five years old, and none includes a results page with an AI-generated answer above the ads. The brands studied were far larger than most advertisers, so a small local name falls outside every sample. The studies also disagree on what competitors gain: one found they take a large share of clicks when a brand stops advertising, another found the nearest rival "did not measurably benefit". And a pause shows a few weeks. It doesn't show what a competitor does once the space has been empty for a year.
Sources
- Thomas Blake, Chris Nosko and Steven Tadelis, "Consumer Heterogeneity and Paid Search Effectiveness: A Large-Scale Field Experiment", NBER working paper 20171; published in Econometrica, 2015.
- Andrey Simonov, Chris Nosko and Justin Rao, "Competition and Crowd-Out for Brand Keywords in Sponsored Search", Marketing Science, 2018. Bing data.
- Lorenzo Coviello, Uri Gneezy and Lorenz Goette, "A Large-Scale Field Experiment to Evaluate the Effectiveness of Paid Search Advertising", CEPR discussion paper 12333, 2017.
- Joseph Golden and John Horton, "The Effects of Search Advertising on Competitors: An Experiment Before a Merger", working paper, October 2019.
- David Chan and Lizzy Van Alstine, "Impact of Organic Ranking on Ad Click Incrementality", Google Research, 27 March 2012.
- Google: Trademarks policy; Use auction insights to compare performance; About measuring paid and organic search results; Apply brand exclusions to Performance Max or search campaigns. Read on 15 November 2026.