Pillar · For Ecommerce brands on GA4
Incrementality Testing
for DTC & Ecommerce
Incrementality testing measures the sales a channel actually caused, not the ones it merely touched. The classic way to do it is a holdout. There is a faster way.
Incrementality testing measures the sales a channel actually caused, not the ones it merely touched. The classic way to do it is a holdout: switch a channel off for some users or regions, wait, and compare. It works, but it is slow, it costs you real revenue while the test runs, and most ecommerce brands cannot hold a clean control group.
Causality Engine gives you the same incrementality answer from data you already have. Upload 40–90 days of Shopify and GA4 history and it estimates each channel's incremental contribution causally, in 5–10 minutes, with confidence intervals, and without turning anything off.
The three ways to measure incrementality
Geo holdout and lift tests
Suppress a channel in some regions, compare against control regions. Rigorous, but slow, expensive, and fragile when your regions are not comparable.
Conversion lift studies
Platform-run experiments. Useful, but the platform designs and grades its own test.
Causal modeling on historical data
Estimate the counterfactual from variation already in your sales history. No test to run, no revenue sacrificed, and you are not grading your own homework.
Why most brands cannot run clean holdouts
A trustworthy holdout needs comparable test and control groups, enough volume to detect an effect, and weeks of patience while you deliberately under-spend a working channel. Most DTC brands have none of those to spare. That is why incrementality testing so often gets talked about and so rarely gets done. Modeling incrementality from existing data removes every one of those blockers.
How to get incrementality from data you already have
Your spend, sales, and seasonality already contain natural experiments. Weeks you spent more, weeks you spent less, channels that paused, promotions that ran. Causal inference reads that variation to estimate what sales would have been without each channel, and reports the gap as incremental ROAS. Every estimate comes with a confidence interval so you can tell a real signal from noise.
Reading the confidence intervals
A tight interval well above 1.0x means the channel is almost certainly paying for itself. An interval that straddles 1.0x means the channel might not be incremental at all, which is exactly the situation where platform dashboards quietly mislead you. The interval is the point. It tells you how hard to lean on the number.
Frequently asked questions
What is incrementality testing?
+
Incrementality testing measures the conversions a marketing channel caused that would not have happened otherwise. It separates incremental sales from sales the channel merely received credit for.
Do I have to run a geo holdout?
+
No. Causality Engine measures incrementality from your existing Shopify and GA4 history using causal modeling, so there is no holdout to design and no revenue lost to a test.
How is this different from a platform's conversion lift study?
+
The platform designs and grades its own lift study. Causal modeling on your first-party data is independent of any single platform's incentives.
How long does it take?
+
5–10 minutes on 40–90 days of data, versus weeks for a typical holdout.
Keep reading
Related on causal attribution
Causal attribution
What it is, how it differs from last-click and multi-touch.
Marketing mix modeling
Modern MMM for SMB and DTC brands.
Holdout test
The classic incrementality experiment, explained.
Uplift modeling
Modeling incremental effect per customer.
Potential outcomes framework
The Rubin causal model: what would have happened if not for X.
Regression to the mean
Why naïve lift studies overstate the effect of an intervention.
Exogeneity
The identifying assumption every credible incrementality estimate rests on.
D-separation
How to read a causal graph to know which channels actually identify.
Real incrementality reports
See what incrementality looks like on real data.
Anonymised holdout and geo-experiment readouts from Ecommerce brands. Lift estimates with confidence intervals; the channels that paid for themselves and the ones that did not.
Browse all incrementality reports →Use cases
What you can do with this.
Each line is a workflow a marketing team runs against its own data. No adjectives, no implementation details, no product mechanics. If a line describes what you are trying to do, the rest of this page is the answer.
- Investigate a channel's missing performance
- Estimate the contribution of a launch campaign
- Identify the breaking point of a paid channel
- Audit a sudden ROAS jump
For Ecommerce brands on GA4 · Defensible in a budget meeting
Five things to know before you upload.
If you run an Ecommerce site with Google Analytics 4 installed, you are a fit. Any ecommerce platform (Shopify, WooCommerce, BigCommerce, custom) works as long as GA4 is the analytics layer.
Proprietary causal-inference model
Not an LLM. Not last-click in a trench coat. Our model is the same statistical machinery used to evaluate medicine and policy, applied to your Shopify and GA4 data.
Confidence intervals on every estimate
Honest uncertainty, not a single confident-looking number. A causal claim without an interval is a guess in a suit.
Methodology open on request
Every assumption documented: prior, functional form, covariate set, robustness checks. The methodology document ships to any customer who asks. The goal is a number you can defend in a budget meeting.
EU data residency. First-party only.
Your Shopify and GA4 exports are processed inside the EU and never sold. No pixel, no SDK, no third-party tracking. GDPR-compliant by construction.
No engineering ticket
Standard exports from Shopify and GA4 go in. Two minutes of setup, no developer needed, no 90-day onboarding, no platform migration.
Want the methodology document? Email hi@causalityengine.ai. Reply within one business day. Or jump to pricing or the interactive demo.
Causal attribution check
Find your wasted ad spend
in 5–10 minutes.
Watch the model work on a sample store first, no signup. Then upload your last 40–90 days of GA4 sessions (and Shopify orders if you have them) and get incremental ROAS with confidence intervals. No pixel, no SDK, no integration project. €99 per run. Every quarter on last-click adds to your marketing debt.
Prefer to talk it through first? Book a 20-min call, or read how it works.