Test analysis
You
What did our Meta test measure?
AI
I’ll summarize sales lift, iROAS and test scope.
Sellforte Incrementality Testing Agent
Your ever-curious data scientist who automatically designs, detects, and analyzes experiments to strengthen every decision you make.
Loading the Incrementality Testing Agent…Open the full demo ↗
Analyze incrementality tests with Sellforte AI
Understand what each test measured
Ask the Incrementality Testing Agent to explain a geo test, Conversion Lift Study, or randomized A/B test. Review the incremental sales and test iROAS, with the test period and audience in view.
Follow up on the uncertainty around the estimate, the channels included, or the outcome the study measured. Use the answer to understand which decisions the result can support.
Prepare a clear readout for your team
Turn the findings into a summary for a campaign review or budget meeting. Ask for the main result, the limits of the study, and the questions that remain open.
Keep the uncertainty range alongside the central estimate, so your team can judge how much confidence to place in the result. Explore the supported methods in Sellforte Incrementality Testing.
Sellforte AIChat
What did our latest Meta Retargeting test tell us?
Sellforte AI
The test estimated $407,000 in incremental sales on $57,240 of spend. Test iROAS was 7.11, with a 90% interval of 5.83 to 8.39.
Test iROAS90% confidence interval
Lower5.83
Mean7.11
Upper8.39
Scroll down to see more ↓
Ask a follow-up…
Sellforte AI
Find the evidence in your test library
Find studies for a channel or market
Start with a question such as “Which Meta tests have we run?” or “What have we learned in this market?” Sellforte AI helps you find the relevant studies in your Test Library, which brings geo tests, Conversion Lift Studies, and A/B tests together.
Narrow the request by test type or period, then ask for a summary of the findings.
Build on what your team has already learned
Compare previous studies before planning another experiment. Ask which findings are consistent, which remain uncertain, and whether the tests cover the campaigns you are reviewing.
A high return in one study is a starting point for a closer look. Keep differences in audience, spend and timing visible when interpreting the comparison.
Sellforte AIChat
Find our recent Meta tests and summarize what we learned.
Sellforte AI
Here are the recent Meta tests and their reported iROAS.
Recent Meta tests3 tests
| Test | iROAS |
|---|---|
| Meta RetargetingOff-peak study | 7.11 |
| Meta ProspectingRepeat study | 3.08 |
| Meta Full-funnelFull-funnel study | 3.14 |
Scroll down to see more ↓
Ask a follow-up…
Sellforte AI
Prioritize your next incrementality tests
Identify the evidence gaps behind a budget decision
Tell the agent which decision is coming up, such as increasing a channel budget or expanding into a market. Ask where past testing leaves that decision uncertain.
Use Test Prioritizer to review opportunities across your media mix, then ask Sellforte AI to explain the evidence behind a proposed test.
Turn a priority into a testing brief
Ask for a draft brief that states the business question, the existing evidence, and what the next test should resolve. Compare candidates by potential business impact and practical feasibility.
Review the available control markets, test cost and detectable effect with your team before choosing a design. A useful priority should lead to a test your team can run and a decision it can inform.
Sellforte AIChat
Which test would help us make our next budget decision?
Sellforte AI
These are the proposed tests, ranked by priority:
Ranked testing opportunities
ImpactConfidenceEase
| # | Media & market | Why it ranks | Priority / 10 |
|---|---|---|---|
| 1 | Meta SalesMeta Ads · Germany | 9.6 | |
| 2 | Meta SalesMeta Ads · United States | 9.2 | |
| 3 | Meta ProspectingMeta Ads · Germany | 9.2 | |
| 4 | Meta Advantage+Meta Ads · United States | 9.1 | |
| 5 | Meta ProspectingMeta Ads · United States | 9.1 |
Scroll down to see more ↓
Ask a follow-up…
Sellforte AI
Compare experiment results with your MMM
Match the scope before comparing returns
Ask how a test compares with your Marketing Mix Model for the same platform and period. Check the market, campaigns and sales outcomes included in each estimate.
A Retargeting test and an all-Meta model result answer different questions. Comparing a more closely matched set of campaigns makes the difference easier to interpret.
Investigate differences and decide what to test next
Review both estimates with their uncertainty. Ask whether the difference points to a scope mismatch, a measurement question, or a gap that another experiment could resolve.
Use the comparison to prepare a discussion with your measurement team about calibration or follow-up testing. Keep the assumptions visible alongside the recommendation.
Sellforte AIChat
Do our latest test results agree with MMM?
Sellforte AI
The Retargeting test reports iROAS 7.11, with a 90% interval of 5.83 to 8.39. MMM estimates Retargeting iROAS at 8.64 for the same period.
The model estimate is just above the test interval. Would you like to include this new test as a calibration input for your MMM?
Ask a follow-up…
Sellforte AI
Bring the Incrementality Testing Agent into your own AI workflows
Test library
You
Find our previous Meta tests.
AI •••
I’ll find past studies and compare the evidence.
Test priorities
You
What should we test next?
AI •••
A Google-only geo test could fill a gap left by studies that combine Meta and Google.
Test planning
You
Draft a brief for our next test.
AI •••
I’ll draft the question, design and checks.
MMM comparison
You
Does this result agree with MMM?
AI •••
Let’s align the period, campaigns and outcomes before comparing the estimates.
Uncertainty
You
How confident is this result?
AI •••
I’ll review the test’s interval and explain which decisions the evidence can support.
Team readout
You
Summarize this test for our team.
AI •••
I’ll summarize the result, limits and next steps.
Follow-up decisions
You
What evidence do we need to scale?
AI •••
Check whether the test covers the intended audience and spend, then identify gaps.
Ask about your tests in Claude or ChatGPT
Connect the AI tools your team already uses to Sellforte through the Model Context Protocol (MCP). Ask about incrementality test results, find past studies, and explore which questions to test next.
Bring the Incrementality Testing Agent into the workspace where you review evidence and prepare your next testing plan.
Turn testing evidence into plans and reports
Use test results alongside your campaign notes and planning documents. Draft a testing brief, compare a study with MMM, or prepare a readout for your team.
Explore the connection and available capabilities on the Sellforte MCP page.
Meet the agents powering Sellforte AI
Incrementality
Testing Agent
Understand what worked and decide what to test next.
Test result
Meta Retargeting
7.11 test iROAS
90% interval5.83–8.39
Media Planner
Agent
Find the media mix that makes your budget work harder.
Recommended mix
€100kMonthly budget
Google Search€45k
Meta€35k
YouTube€20k
Media Buyer
Agent
Put your budget and bidding recommendations into action.
Budget updates
ProspectingMeta+15%
ShoppingGoogle+10%
RetargetingMeta−5%
✓ 3 budget updates applied
