Branded Search Incrementality: Is Google Brand Paid Cannibalising Organic?
A practical playbook for measuring branded search incrementality: pause Google brand-paid in matched geographies, measure organic recapture, decide what brand spend is actually buying.
Quick answer
Run a geo-split pause test: switch off branded paid search in half your regions for 2-4 weeks, then compare total branded revenue (paid + organic) against the control regions. Most stores recover 70-90% of paused paid clicks through organic — meaning only the remaining 10-30% is truly incremental.
Branded Search Incrementality
The share of branded paid-search revenue that would not have happened anyway through organic — the true incremental lift of bidding on your own brand name.
Branded search incrementality measures how much new revenue Google brand-paid actually generates versus how much it simply intercepts clicks that would have arrived through the organic blue link below the ad. When your brand sits at organic position 1 with no competitor bidding above you, paid clicks are mostly cannibalised — the user typed your name, they were coming anyway. Incrementality is established by pausing branded paid search in a matched set of geographies and measuring whether total branded revenue (paid + organic combined) actually drops, and by how much.
This is the single experiment a CFO will pay attention to in your Google Ads account. Branded search routinely shows the best ROAS in the platform — 15x, 20x, sometimes 40x — and is also the line item most likely to be wasted spend. The two facts coexist because Google Ads attributes the whole conversion to the ad even when the organic listing would have caught it.
Why branded paid cannibalises organic
When someone Googles your brand name, intent is already locked in. They are not browsing — they are navigating. Whether they click the ad at position 0 or the organic result two pixels below, they land on the same product page and convert at roughly the same rate.
Google's auction makes this worse: branded CPCs are cheap (€0.20-€0.80 for most stores) precisely because no one outbids you on your own name in most categories. That low CPC creates the illusion of efficiency. You are paying a small toll for clicks that were free a moment before you started paying.
The competitor-bidding exception
If a competitor (or a marketplace listing of your own SKUs) is bidding on your brand name, incrementality jumps. In categories where Amazon, a reseller, or a direct rival shows above your organic listing, pausing brand paid can drop branded revenue by 20-40%. Check the auction insights report before assuming you can safely pause.
How to run the pause test
Use a geo-split, not a time-split. A before/after test confounds seasonality, email sends, and PR. Split your top 10-20 markets into two matched groups by historical branded search volume and revenue, then pause brand campaigns in group A while group B keeps running.
Run for at least 2 full weeks, ideally 4, to absorb day-of-week effects and any latent visits from people who saw the ad days ago. Measure total branded revenue per region (paid + organic + direct from the branded landing pages) — not paid-only revenue, which will obviously crash to zero in the test cells.
Typical branded paid incrementality by competitive context
| Scenario | Organic recapture | True incremental % | Pause recommendation |
|---|---|---|---|
| No competitor bidding, you rank #1 organic | 85-95% | 5-15% | Pause or cap spend |
| Light competitor bidding (1 rival, intermittent) | 70-85% | 15-30% | Reduce to defensive only |
| Heavy competitor bidding (Amazon / aggressive rival) | 50-70% | 30-50% | Keep running |
| Brand misspellings / generic-modifier queries | 60-75% | 25-40% | Keep, separate campaign |
| Mobile-heavy traffic (ad pushes organic below fold) | 65-80% | 20-35% | Test by device |
Reading the result and acting on it
Calculate two numbers. First, the lift: (total branded revenue in control − total branded revenue in test) ÷ control revenue. Second, the true incremental CPA: brand paid spend ÷ incremental revenue (not total attributed revenue). If your incremental ROAS lands below your blended target, the spend is buying recycled clicks.
Most apparel and beauty stores we see land somewhere between 10% and 25% true incrementality on brand-paid when no competitor is bidding. That typically reframes a 'ROAS 30' brand campaign as a 'ROAS 4-6' campaign once you strip out the recaptured organic — still positive, but no longer untouchable.
What to do with the freed budget
Stores that confirm low brand-paid incrementality usually redirect 50-70% of that budget into non-brand prospecting (where incrementality is harder to win but the spend is genuinely additive) and keep 30-50% as a defensive cap that only activates when auction insights detect a competitor encroaching.
Experiment ideas beyond the basic pause
Once the headline pause test is done, layer in segmentation. Run a device-split test — mobile SERPs push the organic result further below the ad than desktop, so mobile cannibalisation is often lower. Test brand + generic queries separately ('yourbrand running shoes') versus pure navigational ('yourbrand'); the modifier queries are usually meaningfully incremental and the navigational ones are not.
Re-run the test quarterly. Competitor bidding behaviour shifts, your organic ranking can slip after a Google core update, and new SERP features (AI Overviews, shopping packs) change how visible your organic listing is. A result from 18 months ago is no longer load-bearing. Treat branded incrementality as part of your broader incrementality testing rhythm, not a one-off.
Frequently asked questions
Yes — that is the main practical risk. Mitigate by running the test in non-flagship markets, keeping the campaign live (just paused) so you can re-enable in hours, and monitoring auction insights daily. If a competitor's impression share on your brand spikes, end the test early in that region.
Google's brand lift studies measure ad recall, not revenue, and its built-in conversion lift studies only run on conversion actions tied to ads. A geo-split pause test you run yourself measures total branded revenue end-to-end, which is the number that matters.
Minimum 2 weeks, ideally 3-4. Shorter tests get distorted by day-of-week and email-driven traffic spikes. If your branded search volume is low (under 5,000 monthly queries), extend to 6 weeks or run a larger geo split.
Then your incrementality is structurally higher and pausing is more dangerous. If you rank #2 or #3 organically — common in categories with a Wikipedia entry or a marketplace listing above you — paid is likely doing real work and the pause test will confirm a higher incremental percentage.
The principle is identical but the math is smaller. Run the same test on Bing if it represents more than 5% of branded paid spend; otherwise just mirror the Google decision.
Your organic ranking on your own brand name is the lever that determines incrementality. Investing in a strong branded SERP (knowledge panel, sitelinks, owned-domain dominance) directly reduces the incremental value of brand-paid, which is a feature, not a bug.
These are mostly existing customers and are almost entirely cannibalised by organic. Exclude them from brand campaigns by default — they convert poorly (support intent, not purchase intent) and inflate your branded ROAS misleadingly.
The total-branded-revenue method is attribution-model agnostic — you are comparing two geo cells on the same metric. Attribution model only matters when you read paid-only ROAS, which is exactly the number you should stop relying on for this decision.
If you can't run a geo split (single-country store, low volume), do a clean before/after pause: 4 weeks on, 4 weeks off, 4 weeks on, and look at total branded sessions and revenue. Less robust than a geo split because seasonality contaminates it, but directionally useful.
It is usually the highest-ROI incrementality test to run first, because the spend in question is concentrated, the test is low-risk, and the result often releases meaningful budget. Treat it as the entry point to a quarterly incrementality testing programme covering Meta prospecting, retargeting, and affiliate.
Track CAC, channels, and funnel conversion in one place
Metricuno connects ad spend, funnel events, and revenue so you can see CAC by channel, cohort, and campaign — without stitching together five tools.