Liftlab - Logo
Request a Demo

How a Retailer Uncovered a $6.6M Revenue Opportunity with a Meta Prospecting Incrementality Test

+92% modeled iROAS (1.63 → 3.13)
$6.6M annual revenue opportunity
24.4% lift in incremental conversions
42-day geo test
How a Retailer Uncovered a $6.6M Revenue Opportunity with a Meta Prospecting Incrementality Test
Summarise with

Here’s how a luxury footwear and accessories retailer used a geo experiment to find that its model had under-credited Meta Prospecting, and mapped a funded path to scale it.

A 42-day geo incrementality test showed that a luxury footwear and accessories retailer’s marketing mix model was crediting Meta Prospecting with about half its true incremental return. Comparing dark markets with matched live ones revealed a 24.4% lift in conversions. Recalibration raised iROAS from 1.63 to 3.13 and showed room to more than double daily spend, worth an estimated $6.6M a year.

Business Challenge

Meta Prospecting was one of the brand’s biggest media investments, accounting for a significant share of total spend. The team’s strategy on the platform had also changed substantially from the year before. When a channel’s strategy shifts that much, a model fit on past data can lag behind how the channel actually performs today. With that much budget at stake, the team didn’t want to scale, hold or cut Meta Prospecting on the strength of an estimate alone. They needed a causal answer to a simple question: how much of this spend is incremental, and how much of the revenue would have happened anyway?

Objectives

  • Measure the incremental conversions and revenue Meta Prospecting drives, using a live experiment instead of a modeled estimate.

  • Compare that result with the model’s pre-test estimate, and recalibrate the model where the two disagree.

  • Find the spend level where Meta Prospecting’s next dollar stops outperforming the rest of the mix, and identify where any extra budget should come from.

How LiftLab Tested Meta Prospecting and Recalibrated the Model

LiftLab designed a single geo experiment that would both answer the incrementality question and feed a better model.

  • Go Dark with Pacing geo test: For 42 days, Meta Prospecting was paused completely in a 20% control cell of markets, while a matched 20% treatment cell kept running as normal.

  • Lift by customer type: The test read incremental conversions for new and existing customers separately, so the team could see who the channel was actually reaching.

  • Calibration through the Trust Engine: The experiment results were fed into the brand’s Agile MMM, updating Meta Prospecting’s iROAS and response curve.

  • Marginal-return analysis: Using the recalibrated curve, the team located the spend level where Meta Prospecting’s mROAS falls to the brand’s blended media mROAS, and checked which channels were returning less on their next dollar.

[Find this interesting? We have more. See how a 42-Day Meta awareness experiment revealed $1.1M in incremental revenue opportunity for another LiftLab customer]

Results & Business Impact

  • After calibration, Meta Prospecting’s modeled iROAS rose from 1.63 to 3.13, a 92% increase. The pre-test model had been crediting the channel with roughly half of its measured incremental return.

  • Meta Prospecting drove about 695 incremental conversions during the test, a 24.4% lift.

  • New customers responded far more strongly than existing ones, with a 42.5% lift compared with 9.9%.

The recalibrated response curve then showed how far the channel could scale.

  • Daily spend could rise from $9,730 to $22,621 before Meta Prospecting’s mROAS falls to 1.09, the brand’s blended media mROAS.

  • At that level, the additional spend is projected to bring in about $18,000 in revenue a day, or roughly $6.6M a year.

The brand now has a corrected iROAS benchmark for its largest prospecting channel and a funded plan to scale it, built on its own experiment rather than on assumptions.

Key Takeaways

  • Re-test when strategy changes. A channel whose strategy shifts year over year can outrun a model built on its history. A live experiment resets the benchmark.

  • Models can miss in either direction. Here the model under-credited Meta Prospecting. In a related Meta Awareness test, the same kind of experiment showed a model over-crediting one outcome and under-crediting another.

  • Read lift by customer type. A 42.5% lift among new customers against 9.9% among existing ones says the channel is doing its job: acquiring customers, not just reaching people who were already going to buy.

  • Scale in steps, and keep testing. The $6.6M figure comes from a response curve projected well beyond the tested spend level. Increase spend in stages and re-test along the way to confirm the curve holds.

Methodology note: iROAS (incremental return on ad spend) is the incremental return observed during the test window, comparing markets where Meta Prospecting went dark (control) with markets where it ran (treatment). mROAS (marginal return on ad spend) is the estimated return on the next dollar at a given spend level, read from the recalibrated response curve. Results reflect one anonymized customer experiment run in Q3–Q4 2026. Individual results vary by category, baseline demand and media mix.

Our customers’ names are kept anonymous to protect their confidentiality.

Want to know how much of your Meta spend is really incremental?

FAQs about the Meta Prospecting Case Study

What is a Meta Prospecting incrementality test?

It measures how many conversions Meta Prospecting campaigns cause that wouldn’t have happened otherwise. In a geo test, the campaigns are paused in a set of control markets and kept running in matched treatment markets. The difference in conversions between the two groups is the incremental lift.

What is the Go Dark with Pacing methodology?

Go Dark with Pacing is LiftLab’s experimental method for measuring incremental media impact. It pauses delivery in selected control markets while maintaining normal delivery elsewhere, allowing measurement of the difference in business outcomes.

What’s the difference between iROAS and mROAS?

iROAS is the average incremental return of a channel delivered during the test. mROAS is the return on the next dollar at a given spend level, read from a response curve. Because returns diminish as spend rises, mROAS is the better guide to where to add or cut budget.

Why would a marketing mix model under-credit a channel?

A model learns from historical data. If a channel’s strategy, audiences or creative change significantly, its recent performance can differ from what its history suggests. Calibrating the model with a live experiment brings its estimate back in line with how the channel performs now.

How does LiftLab keep its MMM calibrated with experiments?

The <a href=”https://liftlab.com/platform/incrementality-testing-engine/”>Incrementality Testing Suite</a> runs geo experiments like this one, and the <a href=”https://liftlab.com/platform/real-time-mmm-intelligence/”>Trust Engine</a> feeds each result into Agile MMM to update the affected channel’s response curve. The model stays anchored to measured outcomes, and every budget recommendation it produces goes to your team to act on.

Download Success Story

See LiftLab in Action

Get a guided demo tailored to your priorities—scenario planning, experiments, and marginal ROI optimization—so your team can move faster without sacrificing long-term growth.
Footnote Background
About Cta Bg Mobile - Footnote