Google just dropped the minimum spend for an incrementality experiment from $100,000 to $5,000 (Google Ads Help, 2025). That means your WooCommerce store can now access the same measurement methodology that enterprise brands use to prove which ads actually drive new sales. Three approaches are within reach: platform-native lift studies, DIY geo-holdout tests, and simple on/off experiments. But here’s what every guide skips—if a third of your visitors are invisible to your tracking, the experiment is measuring fiction.
What WooCommerce Stores Running Ads Actually Need to Know
Attribution dashboards lie. Facebook says you got 85 sales. Google says 60. WooCommerce says 50. The numbers never match, and they never will—because each platform takes credit for the same conversions using different rules.
Incrementality testing sidesteps this problem entirely. Instead of asking “who gets credit?”, it asks a better question: “Would this sale have happened without this ad?”
Until 2025, answering that question required a $100,000 minimum commitment to Google’s experimentation tools—or a $5,000-$50,000 monthly subscription to platforms like Triple Whale, Haus, or Northbeam. That locked out every WooCommerce store owner spending $1,000-$10,000/month on ads.
73% of marketing leaders now view incrementality testing as essential, up from 41% in 2023 (Gartner, 2025). The methodology went mainstream. The price just caught up.
Three Incrementality Tests Your WooCommerce Store Can Run
Not every test requires the same resources. Here are three tiers matched to different budgets and comfort levels—each one practical for a WooCommerce store owner without a data science team.
Tier 1: Platform-Native Conversion Lift Studies (Free)
Both Google and Meta now offer built-in incrementality experiments. A Conversion Lift study is a platform-managed test where the ad network randomly assigns users into exposed and holdout groups, then measures the difference in conversions.
Google’s updated methodology delivers conclusive results up to 50% more frequently than previous versions (Google Ads Help, 2025). The new $5,000 minimum spend threshold means a WooCommerce store running even modest Google Ads campaigns can qualify.
Meta offers similar Conversion Lift studies for Facebook and Instagram advertisers. The platform splits your audience automatically—you don’t need to configure audiences or set up complex experiments.
The limitation: these tests only measure one platform at a time. Google’s lift study tells you if Google Ads drove incremental sales. It says nothing about whether your Facebook campaigns are pulling their weight.
Tier 2: DIY Geo-Holdout Tests ($0 Extra)
A geo-holdout test compares geographic regions where your ads run against regions where they’re paused. It’s platform-agnostic, meaning you can measure the combined impact of all your advertising in one test.
The practical version for WooCommerce stores: pick 2-3 states or metro areas where you pause all paid campaigns for 3-4 weeks. Compare conversion rates in those regions to regions where campaigns stayed active.
Brands spending under $5M/year on paid media may not need intensive geo or platform lift testing—observational experiments like this often suffice (Haus/Supermetrics, 2024). That’s your permission slip. You don’t need enterprise tools for enterprise insights.
The recommended test duration is 3-4 weeks with a 2-week post-treatment observation window (Haus, 2025). That post-treatment window matters—some customers who saw your ads before the pause will still convert during it.
Tier 3: Simple On/Off Tests (Run One Tomorrow)
An on/off test is exactly what it sounds like: pause a campaign, watch what happens to sales, then turn it back on. It’s the simplest form of incrementality testing and any WooCommerce store can run one immediately.
Pick a single campaign—preferably one you suspect isn’t pulling its weight. Pause it for two to three weeks. If sales barely move, that campaign was taking credit for sales that would’ve happened anyway. If sales drop, the campaign was driving real incremental revenue.
This method has real limitations. Seasonality, competitor activity, and external events can all skew your results. But it gives directional evidence with zero setup—and directional evidence beats guessing.
The Prerequisite Nobody Mentions: Your Tracking Data Must Be Complete
Every incrementality testing guide assumes your tracking data is accurate. None of them mention this: if 31.5% of your visitors are invisible to browser-based tracking (Statista, 2024), your test and control groups are equally corrupted.
Think about what that means for a Conversion Lift study. Google splits your audience into two groups—exposed and holdout. But browser-based tracking can’t see visitors using Safari with ITP restrictions, Firefox with Enhanced Tracking Protection, or any browser with an ad blocker installed. Safari, Firefox, and Brave collectively block 20-25% of your tracking data before ad blockers even enter the picture.
Incrementality tests need at least 1,000 people per group to detect a 10% lift confidently (Amplitude, 2025). If a third of those people are invisible to your measurement, your sample size effectively shrinks—and your confidence intervals blow out.
An incrementality test built on incomplete data doesn’t tell you which ads work. It tells you which ads work for the subset of visitors your tracking can see.
The fix isn’t complicated, but it is non-negotiable: your tracking data needs to be complete before you run experiments on it. Browser-based tracking pixels are becoming less reliable with every update—and that trend isn’t reversing.
Server-Side Tracking: The Foundation for Trustworthy Experiments
Server-side tracking captures conversion data on your server before it reaches the browser—where ad blockers and privacy restrictions can’t touch it. For WooCommerce stores planning incrementality tests, this isn’t an optimization. It’s a prerequisite.
Transmute Engine™ is a first-party Node.js server that runs on your subdomain (e.g., data.yourstore.com). The inPIPE WordPress plugin captures WooCommerce events and sends them via API to your Transmute Engine server, which routes them simultaneously to GA4, Facebook CAPI, Google Ads, and BigQuery—all from your own domain, bypassing ad blockers entirely.
Clean data in. Trustworthy experiments out. That’s the sequence.
Key Takeaways
- Google reduced incrementality experiment minimums from $100,000 to $5,000—WooCommerce stores with modest ad budgets now qualify for the same measurement enterprise brands use.
- Three practical approaches exist: platform-native Conversion Lift studies (free), DIY geo-holdout tests ($0 extra), and simple on/off tests you can start tomorrow.
- 73% of marketing leaders now consider incrementality testing essential (Gartner, 2025)—this isn’t optional methodology anymore.
- Every test is only as good as your tracking data. If 31.5% of visitors are invisible to your pixels, your experimental results have the same blind spot.
- Server-side tracking is the prerequisite, not the afterthought. Fix data completeness first, then run experiments on data you can trust.
Platform-native Conversion Lift studies from Google and Meta are free to run—you just need to meet minimum spend thresholds ($5,000 for Google). DIY geo-holdout and on/off tests cost nothing extra beyond your existing ad spend. The real cost is the 3-4 weeks of test duration where you may need to pause some campaigns.
Yes. Meta’s Conversion Lift studies are available to advertisers meeting their minimum budget requirements. Meta randomly splits your audience into exposed and holdout groups, then measures the true incremental lift. The key requirement isn’t budget size—it’s tracking completeness. If your pixel misses visitors due to ad blockers and browser restrictions, your lift measurement reflects partial data.
The simplest method is an on/off test: pause a single campaign for 2-3 weeks, then compare sales during the pause to sales when the campaign was running. If sales barely change, that campaign wasn’t driving incremental revenue. This method has limitations—external factors can skew results—but it gives a directional answer with zero setup.
No. Platform-native tests from Google and Meta handle the statistical analysis automatically—you get a clear result showing incremental lift. On/off tests are even simpler: compare revenue during the pause to revenue when the campaign ran. Geo-holdout tests require slightly more effort in setup but the comparison is still straightforward. The critical factor isn’t statistical expertise—it’s ensuring your tracking data is complete.
Before running any incrementality test, make sure your tracking captures complete data. Server-side tracking closes the gap that browser-based pixels leave in your experimental results. See how Seresa makes it work for WordPress.
