What you can still test without an attribution pixel
Pixel-free does not mean test-free. Four experiments that run perfectly well on aggregate data, and one honest case where you really do need the session detail.
Three platforms, three answers, one budget to defend. Your GA4 export already holds the answer, and the first finding is free and never leaves your browser. Get your first finding free
By Joris van Huët, Founder & CEOUpdated 3 min read
Most of the experiments worth running to a DTC brand do not need session-level tracking, because they compare groups rather than following individuals. The exception is real and worth naming, but it is one case rather than the general rule.
The assumption that experimentation requires a pixel comes from ad platform tooling, where the pixel is the measurement layer. Take the platform out of the loop and the requirement goes with it.
Four that work on aggregate data
| Test | What it compares | What you need |
|---|---|---|
| Geo holdout | Regions with the channel on against regions with it off | Regional order data |
| Time-based pause | The channel running against the channel dark | A clean before and after window |
| Budget step change | Spend at one level against another | A large enough step to clear the noise |
| Creative rotation at fixed spend | Two creatives, same budget, same window | Order counts by period |
None of these observe a person. They observe groups over time, which is what makes them robust to consent rates, ad blockers and browser policy in a way session-based measurement is not.
The geo testing guide covers the first properly, including the pre-period you need before the switch. How to measure incremental lift covers the design questions common to all four.
The one that genuinely needs session detail
Sequencing. If your question is whether people who saw the display ad first convert differently from people who saw the email first, you need to observe the individual sequence, and aggregate data cannot recover it. That is a real limitation and no amount of clever modelling fixes it.
Worth asking, though, whether the sequencing answer would change what you do.
Where the observational read fits
Between "no test" and "a geo holdout" sits the observational read: a causal estimate built from variation already present in your data. It costs a fraction of a holdout and delivers in minutes rather than weeks, and it is weaker evidence, which is why it should be labelled as observational rather than presented as a test.
That labelling matters. An observational estimate ranks below a randomised design and above a rules-based model, and a report that says which one it is lets you weigh it correctly. The comparison is worked through in incremental ROAS from GA4 without a geo test.
A sequence worth running
Start with the observational read to find the channels worth investigating, because it is cheap and it narrows the field. Then spend the expensive instrument, a holdout, on the one channel carrying the most budget. Running a holdout on every channel is neither affordable nor necessary.
Causality Engine sits at the first step: a €99 one-time read on a Google Analytics export, no pixel, no code, refundable within 30 days, no questions asked.
The reframe
The question is not "what can I still measure without a pixel". It is "which of my questions actually needed one".
Related answers
Frequently asked questions
Can I run incrementality tests without a tracking pixel?
Yes. Geo holdouts, time-based pauses, budget step changes and fixed-spend creative rotations all compare groups rather than following individuals, so they work on aggregate order data and are unaffected by consent rates or ad blockers.What experiment genuinely requires session-level tracking?
Anything about sequencing, such as whether people who saw a display ad before an email convert differently from those who saw them in the other order. Aggregate data cannot recover the individual sequence, and no modelling approach can reconstruct it.Should I run a holdout on every channel?
No. Holdouts cost the revenue forgone in the held-out group and take weeks to read. Use a cheap observational read to narrow the field, then spend the holdout on the single channel carrying the most budget, where a wrong answer is most expensive.
Go deeper: Causal attribution, explained.
Sixty-second versions of these ideas: Causality Engine on YouTube Shorts.
Have an audience of ecommerce operators? Earn 20% recurring commission by recommending the €99 read.
Keep reading
Terms in this article
- AnalyticsAnalytics is the systematic computational analysis of data. It reveals customer behavior and measures campaign performance.
- AttributionAttribution identifies user actions that contribute to a desired outcome and assigns value to each. It reveals which marketing touchpoints drive conversions.
- CausalityCausality is the relationship where one event directly causes another, essential for identifying specific actions that drive desired outcomes in marketing.
- ExperimentationExperimentation in marketing conducts controlled tests to determine the causal impact of specific actions. This includes A/B testing and other controlled experiments to establish causality.
- ExperimentsExperiments are scientific procedures that test hypotheses or demonstrate facts. In marketing, experiments like A/B tests determine the causal effect of campaign changes, enabling data-driven decisions.
- Google AnalyticsGoogle Analytics is a web analytics service that tracks and reports website traffic.
- Lift TestA lift test is an experiment that measures lift, the extra conversions or sales a campaign causes, by comparing a group that could see it with a control group that could not. Lift is reported as a difference or as a ratio to the control group's result.
- SessionA Session is a group of user interactions with your website within a given timeframe. It can include multiple page views, events, and transactions.