Last-Click Lies: Why Branded Search Flatters Your Numbers
Your branded search campaigns probably show incredible ROAS. They are also probably the most expensive way to capture sales that were already yours.
The Attribution Illusion
Someone searches your brand name. They click your ad. They buy. Last-click attribution gives Google Ads 100% of the credit. But that customer already knew who you were. They were coming regardless.
The Branded Search Economics
Branded search typically delivers the best metrics in any Google Ads account. ROAS of 10:1, 15:1, sometimes 20:1 are not unusual. The numbers look exceptional because the intent is exceptional: these customers have already decided to buy from you.
The problem is not the performance. The problem is the assumption that performance equals incremental value. A customer typing your brand name into Google was going to find you. Your organic listing was right there. Your direct traffic was a bookmark away.
"Paying for branded search is like paying someone to hold a sign at your own front door. The people walking in were already looking for you."
The Competitor Argument
The standard justification for branded search is competitor defence. "If we do not bid on our brand, competitors will take our traffic." This sounds logical but rarely survives scrutiny.
Studies consistently show that branded searchers have strong purchase intent. They are not casually browsing alternatives. A competitor ad might appear above your organic listing, but click-through rates on competitor ads for branded searches are typically under 5%. Most customers scroll past and click your listing.
The real question is whether the 5% you might lose is worth the cost of bidding on 100% of branded traffic. Usually, it is not.
What You Pay
CPC on branded terms ranges from 20p to £2+ depending on category. Multiply by click volume and you are spending thousands monthly on traffic that would have found you organically.
What You Lose
Without branded ads, perhaps 5-10% of branded searchers click a competitor. But many of those would have returned anyway. True leakage is often 2-3%.
The Multi-Touch Confusion
Multi-touch attribution models attempt to distribute credit across the customer journey. But they still give credit to branded search, just less of it. A position-based model might give branded search 40% credit instead of 100%.
This misses the point entirely. The question is not how much credit branded search deserves relative to other touchpoints. The question is whether branded search caused incremental purchases at all. Attribution cannot answer that.
Testing the True Value
The only honest way to understand branded search value is to test it. Turn off branded campaigns for two weeks. Measure total revenue, not just Google Ads revenue. Compare to the previous period.
In our experience, brands typically see 85-95% of revenue maintain through organic and direct traffic. The 5-15% that disappears represents true branded search incrementality. That is a very different picture from what last-click would suggest.
Branded Search Test Protocol
- 1.Document current branded search spend and attributed revenue
- 2.Pause all branded campaigns for 14-21 days
- 3.Monitor organic and direct traffic closely
- 4.Measure total revenue, not just channel revenue
- 5.Calculate true incrementality as the revenue difference
When Branded Search Makes Sense
Branded search is not universally wasteful. It makes sense when competitors actively bid on your terms and your testing shows meaningful leakage. It makes sense when you want to control messaging during a product launch or crisis. It makes sense when branded CPCs are negligible relative to customer value.
But it should never be an unexamined default. And its performance should never be blended with generic campaigns when evaluating channel effectiveness.
Separating the Numbers
At minimum, branded and non-branded performance should always be reported separately. Blending them creates a flattering average that obscures the real work happening in generic campaigns.
A 6:1 blended ROAS might actually be 15:1 branded and 3:1 generic. Those require very different conversations about investment and optimisation. Hiding behind branded performance is one of the oldest tricks in agency reporting.
The Honest Question
If you turned off branded search tomorrow, how much revenue would you actually lose? If you have never tested it, you are making decisions based on hope rather than evidence. That is expensive hope.
Ready to measure honestly?
We separate branded and generic performance from day one.
Get our insights in your inbox
Plain-English thinking about Google Ads. No spam, unsubscribe anytime.