Skip to main content
    European Search Awards 2026 Winner - Best PPC Agency

    Measurement Methodology

    Incrementality Testing

    Attribution tells you who gets credit. Incrementality tells you what actually worked. Most agencies only do the first. We do both - because the difference between 'attributed' and 'incremental' is where most wasted spend hides.

    The uncomfortable question every brand should ask:

    "How much of what Google claims to have driven would have happened anyway?"

    The Attribution Problem

    Google Ads claims credit for conversions using attribution models - last click, data-driven, or position-based. These models answer: "Which ad touchpoint should get credit for this sale?"

    But they never answer the more important question: "Would this sale have happened without the ad?"

    Brand searches are the clearest example. A customer who types your brand name into Google was already going to buy. The brand ad intercepts them - and the platform claims the conversion. Your agency reports a 12x ROAS on brand. Everyone celebrates. But nothing was actually created by the ad.

    This isn't fraud. It's how attribution works. But if you're making budget decisions based on attributed performance alone, you're almost certainly overspending in some areas and underspending in others.

    A Typical Discovery

    What the platform reported

    4.2x

    Blended ROAS

    £340k

    Attributed Revenue

    After incrementality testing

    1.8x

    Incremental ROAS

    £145k

    Incremental Revenue

    57% of 'attributed' revenue would have happened without ads. The brand was overspending by £32k/month on non-incremental activity.

    How We Test Incrementality

    Four methodologies, each suited to different account structures and commercial questions.

    Geographic Holdout

    Pause all paid activity in a matched region. Compare sales trends against an active region with similar demographics and seasonality.

    Best For

    Brands with national coverage and consistent regional demand

    Duration

    4-6 weeks minimum

    Measures

    What percentage of 'attributed' sales would have happened anyway

    Worked Example

    A pet food brand paused Google Ads in the North East for 5 weeks. Platform attributed 1,200 orders/month to ads. During the holdout, 840 orders still came through organically. True incrementality: 30%, not 100%.

    Campaign-Level Holdout

    Pause a specific campaign type (e.g., Brand, PMax, Generic) while keeping everything else running. Isolate the true lift of that campaign.

    Best For

    Measuring the real value of brand campaigns or Performance Max

    Duration

    3-4 weeks

    Measures

    Whether pausing a campaign reduces total sales or just shifts attribution

    Worked Example

    A home furnishings brand paused Brand campaigns for 3 weeks. CPC costs dropped £8k/month. Revenue fell by only £2k. The brand campaign was cannibalising organic traffic at a cost of £6k/month.

    Budget Scaling Test

    Increase or decrease spend by a fixed percentage in a controlled period. Measure whether the marginal spend produces marginal profit.

    Best For

    Testing whether 'scaling' actually improves outcomes

    Duration

    2-4 weeks per increment

    Measures

    Marginal POAS - the profit generated by the last £1,000 of spend

    Worked Example

    A supplements brand increased daily budget by 25%. Revenue rose 12%. But COGS-adjusted profit fell 8%. The marginal spend was buying revenue at a loss - invisible in platform reporting.

    Channel Isolation

    Run one channel in isolation while pausing others. Understand true channel contribution without cross-attribution noise.

    Best For

    Multi-channel brands unsure which platform drives genuine demand

    Duration

    4-6 weeks

    Measures

    True channel contribution vs. platform-claimed attribution

    Worked Example

    A fashion brand running Google, Meta, and TikTok paused Meta for 4 weeks. Google's reported ROAS improved by 15% - not because Google got better, but because Meta was no longer claiming credit for the same conversions.

    The 5-Phase Process

    From baseline to reallocation in 10 weeks. Every test produces a CFO-ready output.

    1

    Week 1-2

    Commercial Baseline

    • Establish true P&L baseline (not platform P&L)
    • Map current attribution claims vs. bank-reconciled revenue
    • Identify the gap between reported and actual
    • Define the commercial question the test will answer
    2

    Week 2-3

    Test Design

    • Select test methodology based on account structure
    • Define control and test groups with statistical rigour
    • Set minimum detectable effect thresholds
    • Align test duration with business cycles (avoid peak/promo periods)
    3

    Week 3-8

    Execution & Monitoring

    • Implement test with clean controls
    • Monitor for contamination (e.g., organic changes, competitor activity)
    • Track both platform metrics AND commercial metrics simultaneously
    • Weekly check-ins to ensure test integrity
    4

    Week 8-9

    Commercial Analysis

    • Reconcile platform data against actual revenue and margin
    • Calculate true incremental contribution (not platform-attributed)
    • Build diminishing returns curve for marginal spend
    • Produce CFO-ready summary with P&L impact
    5

    Week 9-10

    Reallocation

    • Redirect budget from non-incremental to genuinely incremental activity
    • Set new POAS targets based on proven incrementality
    • Establish ongoing measurement cadence (quarterly re-testing)
    • Document findings for board/finance review

    Why Most Agencies Don't Do This

    Incrementality testing is uncomfortable. It almost always reveals that a portion of 'performance' is actually platform-claimed organic activity. Brand campaigns with 12x ROAS often show 60-80% of conversions are non-incremental. Performance Max often absorbs credit from Shopping campaigns running alongside it.

    For an agency whose fee is justified by attributed ROAS, this is a structural threat. Proving that 40% of spend isn't incremental means recommending a 40% budget cut - and a corresponding fee reduction.

    We charge fixed fees specifically so this incentive conflict doesn't exist. When we find waste, we recommend cutting it. When we find incrementality, we recommend scaling it. The commercial alignment is clean.

    Common Questions

    Incrementality testing measures the true causal impact of your advertising by comparing outcomes with and without ads running. Unlike attribution modelling (which assigns credit after the fact), incrementality tests prove whether ads caused sales that would not have happened otherwise. This is the gold standard for measuring ad effectiveness.

    The primary cost is the revenue you forgo during the holdout period. For geographic holdouts, this typically means pausing spend in 10-15% of your addressable market for 4-6 weeks. The insight gained - knowing exactly which spend is profitable and which is wasted - typically pays for itself within the first reallocation cycle.

    Because the results often reveal that a significant portion of 'attributed' conversions would have happened without ads. This threatens the narrative that justifies the agency's management fee. An agency whose value depends on inflated ROAS numbers has a structural disincentive to measure true incrementality.

    We recommend quarterly re-testing for major campaign types and annual full-account incrementality audits. Market conditions, competition, and consumer behaviour change - incrementality is not a fixed number. What was incremental 6 months ago may not be today.

    Attribution tells you which touchpoints a customer interacted with before converting - it assigns credit after the fact. Incrementality tells you whether the conversion would have happened without the ad at all. Attribution answers 'who gets credit.' Incrementality answers 'did it actually work.' They are fundamentally different questions.

    We use cookies to improve your experience. Privacy Policy