Analytics & Measurement
Meridian GeoX
Also known as: GeoX
Meridian GeoX is the term practitioners use for the geo experimentation side of working with Google Meridian: designing matched test and control regions, running a media holdout or spend change, and using the measured lift to calibrate the model. Geography rather than the individual user is the unit of randomisation, so it works with aggregate data. The aim is to anchor mix model estimates in experimental evidence.
What it is
In a geo experiment, regions are split into comparable groups, one group keeps media as usual and the other is switched off or changed, and the difference in outcomes is the measured lift. GeoX describes this design and analysis workflow used alongside Meridian, where the experiment result becomes a prior or calibration point in the model. Naming varies between teams and tools, so it is worth confirming which design method or library someone means.
Why it matters
Mix models can be sensitive to correlated spend patterns and to the priors chosen, so an independent causal estimate is valuable. Geo tests supply that estimate from aggregated data, which keeps them usable under consent and tracking restrictions. Calibrating a model with a test shortens the argument about whether the numbers are believable.
How it works
You select candidate regions, match them on historical outcomes and size, work out the detectable effect for a given budget and duration, then apply the media change cleanly at region level. Lift is estimated with difference in differences, matched market or synthetic control methods, and the resulting incremental ROI or cost per acquisition is passed into the model as a calibration input.
When it applies
It applies when a channel can be switched or scaled by geography, regional outcome data is available, and the expected effect is large enough to detect inside a realistic test window.
Examples
- Pausing video advertising in twenty matched cities for six weeks and comparing sales with the remaining regions.
- Raising paid social budgets by half in a randomised set of regions to estimate marginal return.
- Feeding a measured incremental cost per acquisition from a radio geo test into a mix model as the prior for radio.
How it is measured
- Measured incremental lift with its confidence interval
- Minimum detectable effect and statistical power at the planned duration
- Pre-period match quality between test and control regions
- Change in the model's posterior channel ROI after calibration
Insights on Meridian GeoX
Related terms in Analytics & Measurement
- A/B testingA/B testing is a controlled experiment that shows two or more versions of a page, email or feature to randomly split groups of users and compares how each performs against a chosen goal. It isolates the effect of a single change so improvement can be attributed rather than assumed. It is also called split testing.
- AI cost per taskAI cost per task is the total cost of completing one unit of useful work with an AI system, divided by the number of tasks completed to an acceptable standard. It includes model tokens, tool and retrieval calls, retries, and any human review needed before the output can be used. It is the unit economics measure behind an AI workflow.
- AttributionAttribution is the practice of assigning credit for a conversion to the marketing touchpoints that preceded it. It covers the models, rules and data joins used to decide which channels, campaigns or content get counted. In AI search, attribution is harder because many assistant-led journeys leave little or no referral data.
- Click-through rateClick-through rate is the proportion of people who saw something and then clicked it, expressed as clicks divided by impressions. It is used to judge search listings, ads, emails and internal links. A higher rate usually means the message matched what the audience was looking for.
- Consent ModeConsent Mode is a Google framework that lets tags adjust their behaviour based on the consent choices a visitor has made. Instead of tags being blocked outright, they receive signals about whether analytics and advertising storage are allowed, and act accordingly. Version 2 added parameters covering the use of personal data for ads and for ad personalisation.
- Conversion trackingConversion tracking is the practice of recording the actions you care about, such as purchases, form submissions or calls, and connecting them back to the channel, campaign or session that led to them. It gives advertising platforms and analytics tools the outcome data they need to report performance and optimise bidding. Accuracy depends on correct tag implementation, consent handling and clear conversion definitions.
Primary research · August 2026
How ChatGPT Shortlists Software Brands
An audit across 10 categories and 60 buying questions. I recorded what ChatGPT reads, throws away and links to when a buyer asks it which software to buy, and what that decides.
- 60
- Questions asked
- 10
- Software markets
- 2,680
- Results read
- 367
- Links shown
Free download
Get the full report
35 pages · PDF · 536 KB. Enter your details and it downloads straight away.