An introduction to Meridian GeoX
Meridian GeoX is an open-source solution designed to establish a Google standard for geo-experimentation. It provides a framework for advertisers and agencies to rigorously measure the incrementality of their marketing efforts through geographic-based testing.
Core benefits of Meridian GeoX
The revamped methodology in Meridian GeoX offers reliable incrementality testing that is both cost-effective and time-saving. Benefits of using Meridian GeoX include:
Streamlined all-in-one solution: Functions as a single library for all current and future methodologies, covering both study design and incrementality analysis. It also allows for comparing study designs across different methodologies.
Tailored experiments: Provides support for many types of experiment designs to support your needs, including holdback, go-dark, and heavy-up.
Multi-cell capability: Enables multi-cell execution, which saves both cost and time by comparing multiple treatment arms with a common control group.
Flexible design: Offers a flexible API to accommodate operational constraints or geo and statistical constraints in study design, such as forcing certain geos to be excluded from testing to avoid large media disruptions.
Meridian MMM integration: Seamlessly integrates with Meridian Marketing Mix Model (MMM), by suggesting new GeoX experiments for your Meridian model based on MMM results. These experiments help calibrate your model by using GeoX's open-source library to test and compare real-world data across different methodologies to provide incrementality-driven priors.
Overview of the journey
Meridian GeoX empowers you to meet specific marketing objectives through a seamless and mediated process. The user journey for an advertiser typically follows these main steps:
(Optional) MMM suggested experiment: If you're a Meridian MMM user, the MMM model output might suggest running a experiment to better calibrate your Meridian Model. Consider using this information to set up your GeoX experiment.
Data collection and preparation: Before designing a geo-experiment, you'll need to prepare a pre-test dataset. All experiments require gathering historical geo-level KPI daily time series data. For designs that modify existing campaign budgets (such as heavy-up or go-dark), you must also collect historical daily geo-level spend data. Incorporating this data allows Meridian GeoX to estimate the budget required for the experiment, helping to streamline budget planning and alignment.
Study design: This stage involves selecting a experiment type and using the Google design algorithms to generate a list of ranked, viable designs based on your specific marketing objectives. Finalize one optimal design based on your marketing needs and proceed with experiment implementation.
Implementation: Here, your experiment is tested in-platform through Google or other media channels. This typically entails setting up geo-level targeting in your designated campaigns, by enabling a marketing intervention and modifying your campaign budget accordingly.
Results data collection: After the testing period, Geo-level KPI daily time series data is gathered for the duration of the experiment. All conversions and spend for tested campaigns are required for each geo. An optional cooldown period can occur after the experiment ends.
Analysis and inference: Analyze results using counterfactual modeling (such as time-based regression) and robust inference methods to evaluate the statistical significance of incremental effect.
(Optional) MMM calibration: If data from the experiment passes quality checks, Meridian MMM users can use the incrementality findings to calibrate their MMM model.