Sellforte Geo Tests

Measure the incremental sales impact of your media

Increase, reduce, or pause media in selected regions. Use Sellforte to measure the effect on sales, understand iROAS and uncertainty, and make your next budget decision with evidence.

Part of Sellforte Incrementality Testing. Available without MMM.

Prioritize, measure, and interpret your geo tests

1. Prioritize your next geo test

Test Prioritizer ranks testing opportunities using Impact, Confidence, and Ease, drawing on your media investment and past testing activity.

Filter by country, ad platform, and suggested test method to focus on geo tests. Use the ranking and gaps in recent testing coverage to decide where new evidence would be most valuable.

Explore the Test Prioritizer

Ranked testing opportunities
ImpactConfidenceEase
#Media & marketWhy it ranksPriority / 10
1Meta SalesMeta Ads · Germany9.6
2Meta SalesMeta Ads · United States9.2
3Meta ProspectingMeta Ads · Germany9.2
4Meta Advantage+Meta Ads · United States9.1
5Meta ProspectingMeta Ads · United States9.1

2. Measure incremental sales

Sellforte uses Bayesian synthetic control to build a comparison from untreated regions. The model learns how to match the test markets before the intervention, then estimates their no-change sales trajectory.

The gap between actual sales and this counterfactual is the estimated treatment effect.

See how the methodology works

Illustrative geo test: actual sales fall below the counterfactual while media is paused, then recover.
Example data. In a go-dark test, lower sales during the pause can show the value of the media removed.

3. Interpret return and uncertainty

Review incremental sales and, when media data is available, test iROAS. The results also show uncertainty and how well the synthetic control matched the pre-treatment period.

Use the charts to check both sides of the test: whether spend changed as intended and how sales responded.

Learn to read the results dashboard

Return and uncertainty Illustrative result
Test iROAS3.0
90% credible interval2.4–3.6
2.43.03.690% credible interval
The interval describes uncertainty under the model. Read it alongside the test design and pre-treatment fit.

Build your test on consistent regional data

Align your test geography with Sellforte Market Areas

Sellforte Market Areas harmonize regional definitions across media platforms and business systems. Give media spend and sales a common geographic foundation for testing.

Geo Tests can use finer regional data than your MMM when that granularity is available in Sellforte.

  • Consistent media and KPI data across regions
  • Treatment kept separate from control markets
  • Enough history to assess the pre-treatment match

Learn more about Sellforte Market Areas

Sellforte Market Areas: A shared regional foundation
US example
US map divided into simplified SMA regions, with selected test regions filled teal and control regions blue. Other SMAs are light gray.
Test SMAsControl SMAsOther SMAs
Illustrative test/control selection. SMA boundaries simplified.

Turn the result into a budget decision

Understand what the media was contributing

A positive iROAS in a go-dark test can reflect sales lost when spend was removed. The signs of the sales and spend changes matter.

Compare the estimate and its uncertainty with your business economics. A revenue return alone does not establish profitability, or show that the same return will hold at a different spend level.

Reading a go-dark result Illustrative example
Sales change vs. counterfactual−$150k
Spend change vs. counterfactual−$50k

Test iROAS: (−$150k) ÷ (−$50k) = 3.0

In this example, each $1 of media removed was associated with $3 of lost incremental sales.

Use matching regions and measurement windows for sales and spend. These numbers illustrate the calculation; they are not a customer result.

Support a budget decision

Use credible evidence to evaluate the investment tested in those markets, for that channel and period.

Investigate a disagreement

When the result differs from current MMM, compare the scope, timing, and definitions before changing your measurement.

Learn from uncertainty

An inconclusive test leaves a question open. Review the design, signal strength, and data before deciding on a follow-up.

Keep the learning after the test ends

Build on your geo tests and compare results across test types in your test library

Keep completed and planned tests together. Filter by test type, country, channel, status, or dates, and open the detailed result when a new decision arises.

Use the AI summary and experiment comments to make the findings easier to share.

Explore Sellforte Incrementality Testing

Test library
TestSpendiROASConfidence
Meta Cart RetargetingConversion Lift · United States$61.11k5.0196%
Meta + Google go-darkGeo test · United States$1.08M5.06100%
Email winback pushA/B Test · United States$7.63k9.50100%
Meta full-funnelConversion Lift · United States$350.41k3.1498%
Meta ProspectingConversion Lift · United States$360.61k2.6595%

Bring credible test evidence into your model

Where MMM is part of your setup, relevant geo experiments can inform model priors. Compare the channel, market, intervention, and period so the evidence answers the same question.

Geo testing is also available as a standalone product.

Compare plans and capabilities

MMM calibration3 studies linked
Meta Prospecting · RevenueMeta Ads · United States · Short uplift
Weighted Experiment iROAS
1.64
Approx. 90% interval
1.39–1.88
Contributing experimentsiROAS90% intervalWeight
Meta US upper-funnel1.341.21–1.4772%
Meta US prospecting2.401.58–3.2228%
Meta US prospecting · follow-upBelow 50% significance2.10−0.31–4.510%

Bring geo testing into your own AI workflows

Geo test results
You

Did pausing media affect sales?

AI

Let’s compare actual sales with the counterfactual.

Test uncertainty
You

How strong is this geo test?

AI

Let’s review iROAS and its credible interval.

Geo test comparison
You

How do our geo tests compare?

AI

Let’s compare channels, markets, and test periods.

Next geo test
You

What should we test next?

AI

Let’s find channels and markets with gaps in evidence.

Geo test vs. MMM
You

Does this agree with our MMM?

AI

Let’s match the channel, market, and period.

Media change
You

Did spend change as planned?

AI

Let’s check test-region spend during treatment.

Geo test report
You

Summarize this geo test.

AI

Here’s a draft covering sales impact, iROAS, and uncertainty.

Regional context
You

Could a local promotion explain it?

AI

Let’s review the test alongside your campaign brief.

Create reports from your geo tests

Use your own AI tools to turn geo test results into presentations, documents, and team readouts. Bring incremental sales, test iROAS, and uncertainty together, with the evidence connected through Sellforte MCP.
Create reports from your geo tests

Build geo test analysis agents

Give your AI agents access to your geo test library. Have them compare completed tests, summarize new results, and flag uncertain findings for review. Use the evidence to inform follow-up questions and decide which channel or market to test next.
Build geo test analysis agents

Connect geo tests to your measurement

Bring geo test results together with MMM, attribution, and other data available to your AI. Compare findings for the same channel, market, and period to understand where the evidence agrees or needs further investigation.
Connect geo tests to your measurement

Add regional business context

Combine geo test results with campaign briefs, local promotions, store openings, and budget changes. Give your AI context to explain what changed in the test markets, what the findings mean for your business, and which questions remain open.
Add regional business context

Geo Tests FAQs

What is a geo test?

A geo test measures the impact of a marketing intervention in selected regions. Sales or another KPI in those regions are compared with an estimate of what would have happened without the intervention.

What data do I need to run a geo test?

Reliable regional KPI and media data, including sufficient pre-treatment history, must be configured in Sellforte. The available geographic detail needs to support both sides of the analysis. Read the setup requirements.

How long should a geo test run?

The right duration depends on the expected effect, sales variability, channel, and purchase cycle. Plan the measurement window before launch and consider a cool-down for delayed effects. Read the geo experiment design guide.

Can I run geo tests without MMM?

Yes. Geo Tests are included in the standalone Incrementality Testing product. MMM and its calibration capabilities depend on your plan and setup. See the current plans.

What if the result is inconclusive?

An uncertain result does not establish that the channel has no effect. Check the size of the media change, historical fit, treatment separation, measurement window, and any unusual events before deciding whether to repeat the test.

How are Geo Tests different from other incrementality tests?

Geo Tests use regional data configured in Sellforte. Uploaded A/B tests use data you supply for test and control groups. Meta Conversion Lift Studies use an underlying experiment conducted in Meta. Compare the test types.

Plan your next geo test with Sellforte

Bring a channel, market, and budget question. Explore the test design and results with our team.
Subscribe to the Sellforte Impact Newsletter