Skip to main content

Incrementality basics

Get a high-level overview of incrementality testing before using the Funnel interface.

S
Written by Sowjenya Parthasarathy
Updated this week

Incrementality testing involves running experiments, such as holdout testing, A/B testing, or geo-lift tests, to find the impact of your marketing efforts on sales. Incrementality testing in Funnel allows you to measure the true business impact of your marketing. Rather than attributing value based on customer touchpoints alone, incrementality testing helps you understand whether a campaign actually caused a lift in conversions or revenue.

This method gives you better insights when optimizing spend. By identifying what marketing is driving value, and what is not, you can reallocate budgets with measurable, statistical backing.

Incrementality is the difference between outcomes that would have happened without marketing and those that occurred because of it.

For example, if you run ads in Region A, which is your test group, and pause them in Region B, which is your control group, any increase in performance in Region A compared to Region B can be considered incremental. This lift represents the results caused by your campaign.

Incrementality testing helps you understand if a campaign actually made a difference.

Advantages of Funnel’s incrementality testing

Marketers use incrementality testing in Funnel to:

  • Validate impact by quantifying what campaigns truly contribute to business outcomes

  • Challenge or support what platform-based attribution models report

  • Optimize budget by identifying which campaigns drive incremental performance

  • Build trust with stakeholders using statistically validated results

This approach is especially valuable when you:

  • Test new channels or campaigns

  • Evaluate media saturation or diminishing returns

  • Justify performance to finance, leadership, or other stakeholders

Incrementality testing workflow in Funnel

Incrementality testing in Funnel follows a structured workflow.

  1. Select test and control groups: Choose geographic regions, such as states or countries. One group receives the marketing treatment, which is the test group, and the other does not, which is the control group. Groups must be separate and similar in behavior.

  2. Export your geo-level metrics you want to test: Use Funnel’s export feature to send your performance data to Funnel Measurement. This allows the incrementality module to access the required metrics.

  3. Launch a test: In the incrementality module, define the test name, test type, metrics, time range, and planned spend. Assign your test and control groups.

  4. Analyze the results: After the test completes and data is available, Funnel creates a synthetic control group and predicts its outcomes without the marketing effect and compares it with the actual outcome. The output includes lift, total and average effects, and key metrics.

Types of incrementality tests

Funnel supports two types of incrementality tests:

  • Uplift test: Use this test to measure the effect of increasing or introducing spend. For example, run a campaign in Region A only.

  • Inverse test: Use this test to measure the effect of removing or pausing spend. For example, stop the media in Region B while continuing in Region A.

Both test types estimate how performance changes when you either add or remove media investment.

Comparison between incrementality and attribution

Incrementality testing focuses on causal impact, while attribution models focus on exposure and touchpoint contribution. Both are useful and complementary.

Features

Incrementality

Attribution

What it measures

If marketing caused an outcome

Which channel a user interacted with

Method

Based on test and control group comparison

Based on rule-based or algorithmic models

Accuracy

Designed to reduce platform bias

May over-credit marketing platforms

Use case

Strategic budget planning and validation

Tactical performance reporting

Did this answer your question?