Introducing Sellforte Experiments Hub: Run Experiments, Get Clearer Results, and Make Your MMM More Accurate Over Time

Every serious marketing team runs experiments.

Conversion lift tests on Meta. A/B tests across campaigns. Geo lift studies to understand regional impact. These are the gold standard of marketing measurement as these controlled tests tell you, with confidence, whether your media is actually driving incremental sales.

And yet, for most teams, the results sit in a spreadsheet.

A dense, platform-generated export full of statistical tables that take hours to interpret, can't be compared across platforms, and have no connection to the rest of your marketing measurement. The experiment happened. The insight is in there somewhere. But the value rarely makes it back into how you plan and optimise your media.

Today, we're changing that.

Sellforte Experiments Hub brings all your marketing experiments into one place with AI-powered summaries that make results easy to understand and a calibration flywheel that makes your MMM more accurate every time you run a test.

Experiments That Feed Back Into Your Model

When you ingest an experiment to Sellforte it does not just become a nicer-looking report, it becomes data.

Every experiment you add is stored in your Sellforte environment and used to calibrate your Marketing Mix Model. That means the ROI figures and channel attributions you see in your Sellforte dashboard get more accurate the more experiments you run. Each test generates a data point with some certainty attached to it that the model can anchor to, reducing risk and giving you tighter confidence intervals across your entire measurement stack.

This is the calibration flywheel.

The more you use Experiments Hub, the more trustworthy your MMM becomes. And the more trustworthy your MMM becomes, the better the decisions you make with it.

What's Inside Experiments Hub

Experiments Hub currently supports three experiment types, all presented in a unified interface with a consistent structure regardless of platform or methodology.

Conversion Lift

If you have run a conversion lift study through Meta Ads, you can export the results as an Excel file and upload it directly to Experiments Hub. Sellforte processes it and returns a clean, structured dashboard with an AI-generated summary, key highlights, and ROI confidence intervals, replacing an unintelligible platform report with something you can actually present to a stakeholder.

Geo Lift

Sellforte's geo lift methodology uses our proprietary Bayesian synthetic control algorithm to measure the incremental impact of media spend across geographies. It is not an off-the-shelf package. It is built and maintained internally, and it is now part of the same unified Experiments Hub experience.

A/B Tests

The most flexible experiment type we offer. A/B tests in Experiments Hub work with any two-group comparison where you changed something for one group and not the other. You do not need geographies. You do not even need marketing data: any KPI time series will work. If you ran a test, we can analyse it.

From Raw Output to Clear Insight

Here is what the workflow looks like in practice.

You ran a conversion lift test on Facebook last quarter. Facebook gave you a spreadsheet. You come to Experiments Hub, select Conversion Lift, give the test a name, choose Facebook Ads as the ad platform, upload the Excel export, and set the start and end dates. Then hit Analyze.

What comes back is a structured result with everything you need to make a decision. In this example, the FB Ads Test ran for 21 days from February 16 to March 8, 2026. The results are immediate and clear: €125k in spend drove €542.5k in incremental sales, an iROAS of 4.34 with a 90% confidence interval of 3.74 to 4.94, and a 14% sales lift. Confidence sits at 96%, indicating high reliability in the positive ROI estimate.

Below the headline metrics, the AI Summary translates the statistical output into plain language. You can see exactly what the test measured, what the result means, and how much confidence to place in it without needing to interpret a single table from a platform export.

Previously, getting to this point required an analyst, a data export, and several hours. Now it takes minutes.

And once the experiment is processed, it is stored. The next time your model runs a calibration, this test is part of the input.

Built for Marketers, Not Just Analysts

Experiments Hub is designed to be used by the same person who uses the Sellforte MMM dashboard, not just by the analyst who set the experiment up.

The AI summary exists precisely for this reason. Advanced users get full statistical detail: confidence intervals, control group comparisons, experiment metadata. But the summary at the top of every result translates the data into plain language that a performance marketer or marketing manager can read, understand, and act on without a statistics degree.

The goal is to make experiment results part of the normal marketing workflow, reviewed in the same place, with the same familiarity as any other Sellforte dashboard.

A Unified View of Everything You Have Measured

One of the less obvious benefits of Experiments Hub is what it looks like after you have uploaded several experiments.

You have a geo lift from Q3. A conversion lift from Q4. An A/B test from earlier this year. All three are now in one place, with the same layout, the same metrics, and the same presentation structure, regardless of which platform they came from or what type of test they were.

Compare that to the alternative: three different platform reports, three different formats, no easy way to see what they collectively tell you about your media.

Experiments Hub is the single source of truth for what your experiments have shown, and with the calibration flywheel running in the background, that body of evidence is actively making your measurement better.

What's Coming Next

Experiments Hub starts rolling out today with manual upload via Excel export. 

This is the starting point and the roadmap from here is significant.

Kacper Solarski, Experiments Hub Product Lead, said, "We are really excited to bring Experiments Hub to our customers. What we are launching today is already a step change in how marketing teams can use their experiment results but the vision goes much further. We want Experiments Hub to become the most powerful experiment analysis tool available to marketing teams, built on methodology that is genuinely state of the art."

On the methodology side, we are already working on upgrading our core algorithm from Bayesian Synthetic Control to Bayesian Synthetic Difference-in-Differences, a more recent and more precise approach that will produce better ROI estimates and tighter confidence intervals. Tighter intervals mean more reliable experiment results, and more reliable experiment results mean better MMM calibration. In short, the flywheel spins faster.

On the platform side, direct API connections to Meta and other ad platforms are on the roadmap, removing the need to download and upload files entirely. As those integrations land, the friction of getting experiments into Sellforte drops to zero.

Further ahead, as part of our Agentic MMM roadmap, we are building the Sellforte Experiments Agent. It will use your MMM data to recommend which channels you should be testing, which geographies to pause investment in, and how long to run the test for to get a result you can trust. It will include an experiment dashboard with automatic test detection, so you can see the data underlying your geo lift studies and identify experiments that may already be running without you realising it. And through a conversational interface, you will be able to ask questions across your full library of experiments and get answers in plain language. Which test should I trust most? What has performed best in North America? How have my lift results trended over the past year? The answer will be one question away.

Get Started With Experiments Hub

Experiments Hub is available as part of the Sellforte Growth Plan, included within Sellforte Experiments alongside Geo Lift, Conversion Lift, and A/B Tests.

If you have conversion lift studies, geo lift results, or A/B tests sitting in a spreadsheet somewhere, now is the time to do something with them.

You can browse example experiment results with dummy data in the Sellforte open demo to get a feel for how the product looks in practice. Please note that data upload is not enabled in the open demo for data privacy and security reasons. To explore Experiments Hub with your own data, log in to your Sellforte account or book a demo with our team.

👉 Book a demo to see Experiments Hub in action

👉 Explore the Sellforte open MMM demo to browse example results

Author

Edward Ford is VP of Marketing at Sellforte. He has over 15 years of marketing experience in B2B SaaS and Tech with specialization in marketing measurement and intelligence. Before joining Sellforte, Edward spent over 6 years at Supermetrics where he joined as an early-stage employee.