Brand search incrementality test: incremental ROAS 8.17
Commentary byROASArchiveOur analysis, not theirs@runtraffic tested whether Google brand search ads bring in sales that wouldn't happen anyway, and found an incremental ROAS of 8.17. New-customer ROAS was 1.17. The post doesn't break down where the rest of the incremental value came from.
- Incremental ROAS: 8.17. Brand search drove real extra revenue, not just sales that would have come in through organic results.
- New-customer ROAS: 1.17. Much lower than the total, but the post doesn't break down where the rest of the value came from.
- Spend, test length and sample size: not shared.
The change
Before
A control where brand search ads were held back, used as the baseline. The test design details are not shared.
After
Brand search ads ran, and the extra revenue over the control was measured.
Our reading: why it likely worked
Our interpretation, not the advertiser's claim.
- People searching your brand name are already close to buying, but without your ad a competitor or reseller can take the click.
- The gap between 8.17 and 1.17 suggests much of the incremental value was not new-customer sales, though the source doesn't break it down.
- Many brand search tests find low incrementality, so this result may depend on how much competitors bid on the brand.
Caveats: Self-reported split or holdout test on one ecommerce account. Spend, duration, conversions and how the holdout was set up are not shared.
How to run this in your account
- Hypothesis
- If we hold out brand search ads in some regions or periods, then we'll see how much revenue they add because the holdout shows what happens without them.
- Primary metric
- Incremental ROAS
- Guardrails
- New-customer ROAS
- Organic brand clicks
- Competitor share of brand searches
- Who should try it
- Brands spending real money on brand search, especially if competitors bid on their name. Small brands with no competitor bidding may see little lift.
Questions
Is brand search incremental?
In this test, yes: incremental ROAS was 8.17. Results differ a lot by brand, so run your own holdout before cutting or raising it.
Does brand search bring new customers?
Less than total sales here: new-customer ROAS was 1.17, against an incremental ROAS of 8.17. The post doesn't say where the rest came from.