The argument is rarely about brand. It's about measurement.
Performance work is measurable and brand work mostly isn't, so budget flows to what can be defended in a meeting, not to what works.
Our bias, declared
We're a performance-led agency and this is where we're most likely to be biased, so treat it accordingly. Our honest position: the two aren't alternatives, and the reason budget skews to performance is that performance can be defended with a number. That's a measurement asymmetry, not evidence of superiority. Brand work makes performance cheaper. One client's paid CPMs fell 26% over eight months of organic and brand growth with no media change. And pure performance eventually hits a ceiling where you're re-buying the same in-market audience at rising cost.
Side by side
| Factor | Performance | Brand |
|---|---|---|
| Measurability | Direct and near-immediate | Indirect, lagging, often modeled |
| Speed to revenue | Days to weeks | Months to years |
| Effect on the other's cost | Little effect on brand strength | Lowers CPMs and lifts conversion on the same ads |
| Defensibility in a board meeting | High, there's a number | Low, requires trust or modeling |
| Behavior at scale | Diminishing returns as in-market audience saturates | Compounds and expands the addressable pool |
| What happens when you stop | Revenue stops quickly | Decays slowly over quarters |
| Suits early-stage companies | Strongly, validates demand fast | Rarely, too slow for limited runway |
| Risk of wasted spend | Lower, visible quickly | Higher, takes long enough to hide failure |
Choose performance when
You have limited runway
Performance pays back inside a quarter. Brand investment with nine months of cash is a bet you probably can't afford to lose.
You haven't validated the offer
Building awareness for something that doesn't convert is an expensive way to learn. Prove the funnel works before amplifying it.
In-market demand is still unsaturated
If you're reaching a fraction of people already searching or shopping in your category, there's cheaper growth available before you need to create new demand.
Your board needs attributable numbers
Sometimes the constraint is organizational, not economic, and that's a legitimate reason. Just name it as the reason.
Choose brand when
Performance costs are rising with no efficiency left
Climbing CPAs at flat conversion usually means you've saturated the in-market audience. More spend into the same pool gets worse, not better.
You're undifferentiated in a crowded category
When buyers can't articulate why you're different, performance ads compete on price. That's a positioning and brand problem that no bidding strategy fixes.
Your category has a long consideration cycle
If people decide months before they search, being known at decision time matters more than being present at search time.
Branded search volume is flat or falling
It's the most accessible proxy for brand health. Falling branded search is an early warning that the demand you're harvesting is shrinking.
How to split it defensibly
There's a widely-cited 60/40 brand-to-performance split from the Binet and Field research on long-term effectiveness. It's a useful reference point and a poor rule to apply blindly. It comes largely from established consumer brands with substantial budgets, and a seed-stage B2B company applying it would likely run out of money.
A more practical approach is to treat brand as a proportion you increase as performance efficiency declines. While performance is still producing efficient growth, keep the split heavily weighted toward it. When CPAs climb at constant conversion rates, that's the signal that the in-market audience is saturating and creating new demand has become the cheaper option.
Whatever split you choose, measure the brand side instead of treating it as an act of faith. Branded search volume, direct traffic, aided and unaided recall in a simple survey, and paid CPM trends all move when brand work is landing. None are as clean as a conversion, and all are better than nothing.
And run a geo holdout if the budget is large enough to justify it. Brand spend is the hardest thing in marketing to justify with attribution, which makes it exactly the thing worth testing for cause and effect instead of arguing about.
Related questions
It's harder to measure, not unmeasurable. Branded search volume, direct traffic, recall surveys and CPM trends all respond. The reason it gets treated as unmeasurable is that nobody set up the measurement before spending. That's a process failure, not a property of brand work.
Other comparisons
Services referenced
Get the audit before you decide anything.
Thirty minutes, free, no obligation. We'll tell you honestly whether you need an agency at all, including when the answer is that you don't.