Judge the additional customer, not the attractive campaign report
A person types your company name into Google. Your paid ad appears above the organic listing. They click, enquire or buy, and the advertising report claims the result. The report is technically correct about the clicked path. It has not proved that the ad created the customer.
This is the central brand-search problem. Brand-name advertising means paying for an ad when someone searches for your own company, product or distinctive trading name. That demand may already exist because of reputation, referrals, offline activity, email, social media, sales work or previous advertising.
My verdict is direct: separate brand demand, cap it and make it earn its place. Run it when the ad protects a valuable customer path, corrects weak organic visibility, carries a time-sensitive message or produces additional customers. Reduce or pause it when organic results recover most journeys and total qualified outcomes barely change.
This is not the same decision as advertising against competitors. That guide addresses entering another brand's demand. It is also distinct from the Performance Max brand-cannibalisation teardown, which diagnoses when an automated campaign inherits existing demand. Here, the owner is deciding whether paying for the business's own name creates commercial value at all.
The Brand Search Incrementality & Defence Gate
Use six checks before approving, renewing or expanding brand-name search spend. A cheap click and high reported return are not enough.
Who created the search?
Identify whether reputation, referrals, offline activity, email or earlier campaigns generated the brand interest.
Would buyers still arrive?
Review the real search page, business listing, sitelinks and any confusion around the company name.
Is valuable demand exposed?
Verify actual competitor participation, resellers, affiliates or misleading results rather than relying on anecdotes.
Does the ad add useful control?
Use a specific offer, location, product, booking path or correction—not a duplicate homepage promise.
What value remains after cost?
Judge qualified enquiries, new customers, contribution and total acquisition cost, not brand-campaign return alone.
Can leadership observe the difference?
Choose a safe holdout, stable window, stop condition and owner before changing coverage.
Google's current brand-settings guidance confirms that Search campaigns can use brand inclusions and that Search and Performance Max can use brand exclusions. Those controls help isolate the traffic. They do not answer the commercial question on their own.
Start with separation. Keep own-brand demand away from non-brand acquisition reporting. Where Performance Max is also eligible to serve, Google's brand-exclusion guidance recommends excluding a brand when that traffic is managed in a separate campaign. Otherwise, one easy source of demand can distort decisions about prospecting and budget.
Why the reported return can overstate business value
Search intent and advertising impact are correlated. People who already intend to visit the business are more likely to search its name, click and convert. A platform can observe the click but not automatically know what the same person would have done without the ad.
A large eBay field experiment published through the National Bureau of Economic Research found that paid-search returns were a fraction of non-experimental estimates and varied across customer groups. That does not prove every business should stop brand ads. It proves one famous result cannot replace a business-specific test.
| Matched period | Paid brand visits | Organic brand visits | Total brand-site visits | Suitable outcomes | Ad cost |
|---|---|---|---|---|---|
| Brand ads on | 420 | 710 | 1,130 | 86 | £720 |
| Brand ads off | 0 | 1,060 | 1,060 | 82 | £0 |
The campaign might report 86 outcomes while live, but this planning example observes only 70 more visits and four more suitable outcomes across the whole brand journey. It still does not prove the ad caused either difference. Seasonality, competitors, promotions and customer mix must be checked before treating £180 per observed additional outcome as incremental cost.
Use Google's Auction Insights to see which advertisers participate in the same auctions. Do not turn one competitor impression into an unlimited defence budget. The commercial question is how much suitable demand is genuinely lost or confused without the ad.
The Search-to-Proven-Value Evidence Loop
The gate decides whether a test is justified. This loop keeps the decision connected to total business outcomes.
Separate the demand
Define own-brand queries, remove obvious ambiguity and prevent other campaigns from obscuring the result.
Record the whole journey
Capture paid, organic and direct visits, competitor presence, qualified outcomes, customer status and contribution.
Create a controlled difference
Use an eligible platform experiment, matched geographies or a guarded pause designed around business risk.
Join marketing and sales truth
Compare total suitable customers and value, not the movement of clicks between paid and organic reports.
Run, cap, narrow or stop
Keep only the coverage that adds customers, protects material demand or delivers a justified message.
Recheck when conditions change
Repeat after a competitor enters, organic visibility changes, a promotion launches or brand demand grows.
Google describes Conversion Lift as a way to measure the causal impact of advertising, but also says it is not available to every account. Google also documents geography-based Conversion Lift, which separates comparable regions into ad and no-ad groups. Eligibility and scale matter. A smaller business should not imitate a large experiment with too little data and declare certainty.
Trademark policy is another boundary, not a performance conclusion. Google says it generally does not restrict using trademarks as keywords, while restricting certain confusing or misleading uses in ad copy. Legal and policy risk differs by market and wording; obtain qualified advice when the issue is material.
Match the decision to the evidence
| Situation | Verdict | Owner action | Avoid |
|---|---|---|---|
| Competitors repeatedly appear on valuable brand searches | Defend and test | Use capped coverage, clear messaging and a total-outcome holdout | Assuming every defended click was otherwise lost |
| Organic result is clear, dominant and uncontested | Test reduction | Pause or narrow safely and monitor total suitable outcomes | Keeping ads because reported return looks high |
| A promotion, urgent booking path or location needs control | Run selectively | Match ad and destination to the temporary decision | Sending every brand search to a generic homepage |
| Performance Max also captures own-brand demand | Isolate first | Separate reporting and apply the appropriate brand control | Counting the same demand as proof of prospecting |
| The supposed brand is a generic or ambiguous term | Do not label it owned demand | Inspect real queries, intent and competitors before classifying | Using low campaign cost as proof of brand ownership |
| Search volume is too low for a credible test | Cap and document uncertainty | Use a longer window, qualitative risk evidence and a review date | Claiming incrementality from a handful of conversions |
If the bigger question is overall affordability, use the guide to setting a paid-advertising budget. If one channel or campaign receives too much credit, review marketing-channel dependency and how to know whether marketing is working. ThomPerformance's Google Ads management and conversion tracking connect these decisions to qualified pipeline, purchases and customer value.
A 45-day brand-search value test
Separate and baseline
Define the brand-query set, exclude overlap, record organic coverage, auction pressure, suitable outcomes, new-customer share and value.
Run the guarded holdout
Use the safest credible design, keep unrelated changes stable and monitor agreed loss, confusion and revenue stop conditions.
Reconcile and decide
Join paid, organic, direct, CRM and financial evidence; document limitations; then run, cap, narrow or stop.
Run when additional customers or material defence justify the total cost. Cap when the value is real but narrow. Narrow when only certain locations, products or high-risk queries need coverage. Stop when organic journeys recover the demand and total suitable outcomes do not materially change.
Review ThomPerformance's paid-growth services, approved case evidence, evidence standards and practitioner background before deciding whether the working model fits your business.
Frequently asked questions
Should a business always bid on its own brand name in Google Ads?
No. Brand-name ads can protect valuable demand, control the message or support a specific offer, but they can also receive credit for customers who would have clicked the organic result. Use a separate, capped campaign and keep it only when business-level evidence shows incremental or genuinely defensive value.
Why do ads on a business name often report such strong results?
People searching for a known business usually have high intent before the ad appears. The ad platform may attribute their click and conversion to advertising even when they would have reached the website organically. Reported return therefore describes attribution, not automatically the additional customers caused by the ad.
What if competitors advertise when someone searches for my business?
Treat competitor presence as a defensive signal, not automatic proof that every brand click is valuable. Review actual auction participation, the importance of the query, organic visibility, customer confusion and downstream outcomes. A capped defensive campaign may be sensible while the business tests how much demand is truly at risk.
Can Performance Max spend money on my own brand searches?
Yes. Google provides brand controls for Search and Performance Max. If brand demand is managed separately, use the appropriate exclusions and reporting separation so an automated campaign cannot quietly inherit easy brand conversions and make prospecting performance look stronger than it is.
How should a business test whether brand-name ads are incremental?
Compare total business outcomes—not paid conversions alone—across a controlled holdout or carefully matched periods. Include paid, organic and direct visits; qualified enquiries or purchases; new-customer value; competitor presence; media cost and material changes. Larger eligible advertisers may use formal incrementality experiments; smaller tests need explicit limitations and stop conditions.
Make brand-name advertising prove what it adds
An impressive brand campaign can be a useful defence, a controlled message or an attribution illusion. Separate the demand, observe the whole customer path and test the business outcome. Then pay for the part that creates or protects value—not automatically for every customer who already knows your name.
Discuss your brand-search decisionSources and methodology
This guide combines current Google documentation, published field-experiment evidence and Thomas Ho's practitioner judgement. It does not claim one universal brand-search result. Search priority is qualitative; no search-volume claim is made. Sources were checked on 29 September 2026. All figures in the worked comparison are illustrative and are not client results or market benchmarks.