If you’re running both PMax and Standard Shopping, this is probably happening:

PMax shows 4–6× ROAS.

Standard Shopping looks flat.

Branded Search is strong.

Blended revenue looks “healthy.”

But here’s the uncomfortable question:

Would that revenue have happened anyway?

Most brands never test this properly. And that’s where scaling breaks.

The Core Problem

Performance Max blends:

  • Shopping

  • Search

  • Display

  • YouTube

  • Discovery

  • Remarketing

When it reports 5× ROAS, you don’t know:

  • How much came from branded search

  • How much came from remarketing

  • How much was pure demand capture

PMax is very good at harvesting existing demand.

It is not automatically good at creating new demand.

And if you’re scaling budgets based purely on platform ROAS, you may just be paying more to capture what you already owned.

What “Incremental” Actually Means?

Incremental revenue = revenue that would not have happened without that campaign.

In DTC, true incrementality usually comes from:

  • Non-brand Standard Shopping

  • Competitor conquest Search

  • YouTube prospecting

  • Cold Demand Gen traffic

Not from branded capture.

Branded search is defense.

Remarketing is recovery.

Incrementality is expansion.

If your PMax campaign is leaning heavily on brand + remarketing inventory, you’re not scaling the pie you’re slicing it differently.

Why This Matters at $30K–$500K+/mo

At smaller budgets, inefficiency hides.

At scale:

  • Branded traffic caps out

  • Remarketing pools saturate

  • CPCs rise

  • Blended ROAS drifts

And suddenly “stable” performance collapses.

When brands hit instability at scale, it’s rarely creative fatigue.

It’s usually demand dependency.

How We Test It Properly

If you want the real answer, you need isolation.

1️⃣ Geo Split Test (Most Reliable)

  • Turn off PMax in 30–50% of states

  • Keep Brand + Standard Shopping running everywhere

  • Measure blended revenue across test vs control

Watch total revenue not just campaign-level ROAS.

If revenue drops meaningfully in PMax-off regions → PMax was incremental.

If revenue barely moves → you were harvesting demand.

Most brands are surprised by the result.

2️⃣ Brand Exclusion Audit

Inside PMax:

  • Apply brand exclusions

  • Separate branded Search into its own campaign

  • Compare performance before vs after

If ROAS collapses when brand is removed, that tells you exactly what was doing the work.

3️⃣ Non-Brand Shopping Isolation

Run:

  • Standard Shopping (non-brand only)

  • Phrase + Exact first

  • Tight query mining

  • Aggressive negatives

Measure:

  • New customer rate

  • Assisted conversions

  • Blended MER

Non-brand Shopping often drives more incremental lift than PMax when structured correctly.

Structure beats automation.

The Big Idea

High ROAS ≠ High Incrementality.

You can have:

  • 6× platform ROAS

  • Flat total revenue

  • No new demand entering the funnel

And still feel like you’re “scaling.”

True scale comes from:

YouTube to create demand.

Search & Shopping to capture it.

Clean brand separation.

And testing incrementality instead of trusting dashboard metrics.

What To Do This Week

  • Separate brand and non-brand everywhere

  • Add brand exclusions in PMax

  • Compare “Conversions” vs “Conversions (by conv. time)” before reacting

  • Model a geo split test plan before your next budget increase

Don’t increase budget until you know what’s actually driving net-new revenue.

If you’re a DTC brand spending $30K–$500K+/month on Google & YouTube and want clarity on what’s truly incremental in your account, we’ll map it out with you. 30-minute deep audit.

Patrick

CEO, Ad-Lab

Keep Reading