Should a supplement brand test first orders or all orders?

Usually first orders, if growth is the question. All orders mixes in reorders from people who would buy anyway, which shrinks the visible gap. Test all orders only when you want to know whether ads to existing customers add anything.

Do subscription renewals count in a conversion lift test?

Only if they send a conversion event during the test. If they do, they land in both groups, because a lift test counts all conversions in each. They add noise, not lift, so measure new subscriptions instead.

When should a sports nutrition brand avoid running a lift test?

During a launch, a big sale, a stockout or a creator push. Each one moves sales on its own and can swamp the gap you want to see. If January or early summer is your rush, avoid those weeks too.

What is a conversion lift test for a sports nutrition brand?

For a sports nutrition brand, a conversion lift test holds a random group back from your ads and compares purchases. If you sell protein or supplements, test first orders or new subscriptions, because reorders land in both groups and hide the gap.

By Joris van Huët, Founder & CEOUpdated 2026-10-015 min read

Run the numbers for your store: the free holdout test planner.

For a sports nutrition brand, a conversion lift test usually means holding a random group back from your ads. You then compare its purchases with the group that could see them. If you sell protein or supplements, decide first whether you are testing first orders or all orders, because reorders can flood both groups.

If you sell sports nutrition

If you sell protein, creatine or pre-workout, your best customers come back when the tub runs out. Many of them never needed an ad to remember you. Some buy on subscription, so their order arrives whether or not anyone saw a campaign that month.

Your calendar is lumpy too. New-year resolutions, summer training, flavour launches and creator codes all move sales on their own. If a creator posts a discount code, buyers may arrive by typing your address, and the ad report never sees where they came from.

A conversion lift test cuts through that, because Meta's lift tests count all conversions in the test and holdout groups. The catch is the same feature. Reorders land in both groups too, so the gap you want sits on a big pile of buying the ads did not cause.

One store's export, read for repeat buying

Below is one store's anonymised GA4 export, 1 January 2024 to 21 August 2026. It holds shares of revenue only: no ad spend, no order counts. Nothing in it says the store sells sports nutrition, so it is not a benchmark for that trade.

What the export shows	Share of revenue	Source cell
Journeys with 1 touch (0.5 days to buy)	79.5%	Journeys sheet, 1 touch row
Journeys with two or three touches (12.5 days to buy)	12.2%	Journeys sheet, 2-3 touches row
Direct, in last click, first click and touched views	57.7%	Channels sheet, Direct row

In this store, journeys with 1 touch carry 79.5% of revenue on the Journeys sheet, at 0.5 days to buy. One touch and half a day looks like someone who already knew where to go. If you sell supplements and your own export looks like this, many of those quick journeys may be reorders.

Direct holds 57.7% of revenue in every view on this store's Channels sheet. If your regulars reorder by typing your address, a lift test on ads shown to them mostly measures loyalty. That is a fair question for a retargeting budget. It is the wrong one for prospecting.

Journeys with two or three touches carry 12.2% of revenue on the Journeys sheet, at 12.5 days to buy. If your new customers compare brands before a first order, a lane like that is where they sit, and a test has to outlast it.

The export cannot show which journeys were first orders, or what any ad caused. Shopify knows the first. Only a test knows the second.

What changes for a sports nutrition brand?

Choose the conversion before the campaigns. Meta's Experiments measure conversions such as purchases or subscriptions. If growth is the goal, measure new subscriptions or first orders, not every purchase. Google adds that conversions that happen more often raise the certainty of a result, so check that the event you pick is not rare.
Test prospecting and retargeting apart. One study per question, one at a time. Mixed together, a weak prospecting campaign can hide behind the loyal buyers your retargeting reaches.
Mind the second tub. A lift test counts conversions between its own start and end dates. Meta does not recommend using its post-test conversion window, and Google projects late conversions only for Demand Gen studies. So the reorder a new customer makes next month falls outside the test. Value a first order with that in mind.
Keep the menu still. Google warns that changing creatives and audiences during a study might affect it. A flavour launch, a stockout or a creator push mid-test does the same, in one group or both. Book the test in a calm month.

What to do this week

Measure your repeat share. In Shopify, go to Analytics > Reports, click the Category filter, choose Sales and open Subscription vs one-time sales. Pass: you know how much of your sales is subscription, so you can pick the conversion to test. Fail: the report is empty, which Shopify says happens unless you sell through the Shopify Subscriptions app, so count repeat orders another way first.
Check what your renewals send. In Meta Events Manager, open Data Sources and select your dataset. Compare a week of Purchase events in the Overview table with Shopify's order count for that week. Pass: the events match your checkout orders only. Fail: they track all orders, renewals included, so test first orders or new subscriptions instead.
Find a calm month. In Shopify, open Total sales over time from the same Sales reports and group it by month. Pass: you can name four weeks with no launch, sale or creator push. Fail: every month has one, so test in the quietest weeks and say so when you read the result.

Check the homework. Your GA4 Attribution paths export already holds the evidence. Causality Engine reads that one file and shows what each channel caused next to what last-click gave it, in 1 to 2 minutes, for €99 once (excluding VAT), refundable within 30 days. Check the homework

Sources, 1 October 2026: Differences between Conversion Lift test results and other reporting tools (Meta Business Help Center); About Experiments (Meta Business Help Center); Set up Conversion Lift based on users (Google Ads Help); About Conversion Lift (Meta Business Help Center); Sales reports (Shopify Help Center); About Meta Events Manager (Meta Business Help Center).

Frequently asked questions

Should a supplement brand test first orders or all orders?
Usually first orders, if growth is the question. All orders mixes in reorders from people who would buy anyway, which shrinks the visible gap. Test all orders only when you want to know whether ads to existing customers add anything.
Do subscription renewals count in a conversion lift test?
Only if they send a conversion event during the test. If they do, they land in both groups, because a lift test counts all conversions in each. They add noise, not lift, so measure new subscriptions instead.
When should a sports nutrition brand avoid running a lift test?
During a launch, a big sale, a stockout or a creator push. Each one moves sales on its own and can swamp the gap you want to see. If January or early summer is your rush, avoid those weeks too.

Go deeper: Causal attribution, explained.

Sixty-second versions of these ideas: Causality Engine on YouTube Shorts.

Keep reading

Terms in this article

Browse the full glossary