Budget allocation across channels

How do you split budget between Meta, Google and the rest when every platform lies about its own share? Here's how we steer by the total.

Explore the process

The concept in practice.

Consideration / 01

The problem with allocating on silo ROAS

The intuitive method — move budget toward the channel with the highest ROAS — is a trap.

Read this section
Select a point to explore the explanation. The full article remains available below.

The problem with allocating on silo ROAS

The intuitive method — move budget toward the channel with the highest ROAS — is a trap. Each platform credits itself for the same sales with its own generous windows, so the sum of their reported revenue exceeds what actually came in. Allocate by silo ROAS, and you move budget toward the channel that's best at taking credit, not the one that creates the most value.

The result is often that retargeting and brand searches get fed at the expense of the demand-generating top of funnel — after which the top of funnel gets cut, demand dries up, and the whole machine seizes up while the dashboard still looks fine.

Explore the numbers

Several channels can claim the same purchase.

Meta70
Claimed by both40
Google70
Sum of channel reports140
≠
Unique purchases in store100

All 100 purchases in this example are claimed by at least one channel. Adding channel reports counts shared purchases twice. MER uses store revenue and total marketing spend; it does not determine causal lift.

Illustration of overlapping channel reports. Shared conversion claims are different from technical event deduplication.

Steer by the total, test with lift

The honest yardstick is MER: total revenue against total spend, without attribution. Allocate the budget, and watch whether MER holds or rises as you move money around or scale. If MER falls, you bought revenue you'd have gotten anyway, no matter how flattering the silo ROAS looks.

For the bigger decisions, you supplement with incrementality tests: turn a channel off in a geo group and measure the real effect. The combination (MER continuously, lift tests occasionally) gives an allocation basis no single platform can inflate.

A practical order of operations

In practice: set MER as the top-level steering metric, use POAS and contribution margin to optimize within each channel, and reserve lift tests for the expensive questions (is brand incremental? is the top of funnel pulling its weight?). Scale as long as the total holds, and don't chase the prettiest number in a single account.

Frequently asked questions

Why can't I just move budget to the channel with the highest ROAS?

Because the platforms' ROAS double-counts the same sales. Allocate by silo ROAS and you reward the channel that's best at taking credit, not the one that creates the most real value — often at the expense of the top of funnel that drives demand.

What should I allocate budget by instead?

By MER (total revenue against total spend) as the top-level steering metric, optimized with POAS within each channel, and validated with occasional incrementality tests. Scale as long as the total holds.

From insight to action

See how it applies in practice.

Matraws

Growth also needed a better purchasing process

We built Matraws' trade-in system with submission, assessment, offers and handling in one backend. The work addressed supply as a constraint.

Read the case

Priorities and collaboration

Discuss your next step

Tell us where your business stands. We clarify the work needed and whether we are the right fit.

Choose a time

30 minutes · No obligation

About the author

Growth hacker and fractional CMO with 10+ years' experience and hundreds of millions in managed ad spend behind him. Background from larger Danish and international scale-ups, and from the agency world.

Meet the team

Keep reading.

All articles