Google Optimize shut down on September 30, 2023, so a store now needs a separate A/B testing tool. On Shopify, the cheapest place to start is the built-in Rollouts feature or a low-cost testing app like Elevate or Shoplift. If you need full control, a test on any page, or a tool that also works off Shopify, look at a paid platform like Convert or VWO.
This page explains what happened to Optimize, what each replacement costs as of November 2026, who each one is not for, and how to switch without losing a month of testing. Wherever we can, we point to real test results from the LiftArchive database.
What happened to Google Optimize
Google announced on January 20, 2023 that it would discontinue Optimize and Optimize 360. Both stopped working on September 30, 2023, and any experiments still running that day ended, according to Google's sunset notice.
Google's own explanation was that Optimize "did not have many of the features and services that our customers request and need for experimentation testing." Google did not build a replacement. It said it would work on Google Analytics 4 integrations with third-party testing tools, naming AB Tasty, Optimizely and VWO, and that it had opened its APIs so other vendors could connect too.
So "Google Optimize A/B testing" no longer exists. GA4 can report on experiments, but the test itself (splitting traffic, showing the variant, choosing the winner) now has to run in another tool.
For many stores this was the first time testing cost money. That changes the math. A tool that costs $99 a month has to find at least one winner large enough to pay for itself, so it matters more which tool you pick and what you test with it.
Free and cheap replacements
None of these is a like-for-like copy of Optimize, but between them they cover what most small stores used it for: theme tests, headline and layout tests, and simple split URL tests.
Shopify Rollouts (built in)
Shopify's June 5, 2026 changelog added A/B tests "between two entirely different themes or checkout setups" inside the admin. Shopify copies your live theme for the experiment, so you can keep editing the live store separately.
The requirements page says rollouts are available on the Basic plan and up, but experiments need the Grow plan or higher. It doesn't work with vintage themes, you can't change Liquid templates as part of a rollout, and experiments only apply to the online store, not headless storefronts.
Whole-theme tests can be worth a lot. Kasper Dolk tested a new theme with a BOGO offer against the old theme and saw conversion rate go up 44.54% (2.44% to 3.53%), with a 99.7% probability to beat the old theme.
Not for: stores that want to test one element at a time, prices or shipping rates, or stores on a plan below Grow.
Elevate
Elevate is a Shopify testing app. As of November 2026, its pricing page lists Basic at $49 a month (theme, template and split URL tests, plus a no-code visual editor) and Standard at $99 a month, which adds price, subscription, product image and checkout testing. All plans have a 14-day free trial, and yearly billing is listed as 40% off.
Elevate also publishes its own results. Elevate A/B Testing's pink vs green add-to-cart button test found pink won, 11.49% vs 10.81% add-to-cart. The gallery photo swap moved revenue per visitor from £3.32 to £3.60 in 10 days.
Not for: stores that aren't on Shopify, or teams that want a dedicated account manager without paying for the $499 Premium plan.
Shoplift
Shoplift is another Shopify-only app. Its pricing page lists Core at $99 a month for up to 50,000 visitors as of November 2026, with theme testing, URL split testing and Bayesian analysis. GA4 integration, price testing and country targeting start on Advanced at $299 a month. Every plan has a 14-day free trial. Our Shoplift review goes deeper.
Not for: stores that want GA4 reporting or price tests on the entry plan, and anyone not on Shopify.
GrowthBook and PostHog (free tiers, more technical)
If someone on your team can write code, two developer tools have free plans. GrowthBook's cloud Starter plan is free for up to 3 users, and there is a free self-hosted open-source version. Its AI visual editor is on the Pro plan at $40 per seat a month as of November 2026.
PostHog has a free tier and supports no-code web experiments. Its docs say these work best for small text and formatting edits and may get overwritten on single-page apps.
Not for: store owners who want a point-and-click Shopify app with ecommerce reports built in. These tools are made for product and engineering teams.
Microsoft Clarity (research, not testing)
Microsoft Clarity is free and gives you heatmaps and session recordings. It won't run a test, but it shows you what to test.
@elevate_ab
@kasper_dolk@elevate_abPaid replacements
Paid platforms are worth it when you test on every page type, need stricter stats and targeting, or run sites that aren't on Shopify. Here is how they compare on the facts we could check.
| Tool | Works on | Entry price (as of November 2026) | Free option | Not for |
|---|---|---|---|---|
| Shopify Rollouts | Shopify only | Included; experiments need Grow plan or higher | Yes, on eligible plans | Element, price or shipping tests |
| Elevate | Shopify only | $49/month | 14-day trial | Non-Shopify sites |
| Shoplift | Shopify only | $99/month (up to 50k visitors) | 14-day trial | Price tests on a starter budget |
| Intelligems | Shopify only | From $69/month (build your own) | 14-day trial | Stores with no plans to test price or shipping |
| Convert | Any website, including Shopify | $299/month billed yearly ($399 monthly) | 15-day trial | Very small stores with tight budgets |
| VWO (now Wingify) | Any website | Not published; plans by traffic | 30-day trial | Buyers who want a public price list |
| GrowthBook | Any site or app (needs code) | Free; Pro $40/seat/month | Free plan and open source | Non-technical teams |
Convert
Convert works on Shopify and any other site. As of November 2026, its pricing page lists Growth at $399 a month, or $299 a month billed yearly, for 100,000 tested users a month. Pro is $599 a month, or $420 billed yearly. There is a 15-day free trial with no credit card.
That is a big step up from a $49 app. In return you get a full testing platform that isn't tied to Shopify's theme system, which matters if you run a headless store or more than one platform. Read our Convert review for the detail.
Not for: stores with too little traffic to finish tests at 100,000 tested users a month. If that's you, test bigger changes first, or start with a cheaper app.
VWO (now Wingify)
VWO was one of the three tools Google named for GA4 integrations. It now goes by the parent company name, Wingify, and vwo.com/pricing redirects to the Wingify site. The pricing page lists Growth, Pro and Enterprise tiers but doesn't show prices, so you will need a demo or quote.
VWO once had a free Starter plan pitched as "a better Google Optimize." Wingify says that plan has been sunset, and new users now get a 30-day free trial instead. VWO bundles testing with heatmaps and session recordings. See our VWO review for more.
Not for: anyone who wants to know the price before talking to sales, or who only needs a simple Shopify theme test.
Intelligems
Intelligems is built around price, shipping and offer tests on Shopify. Its pricing page shows a build-your-own plan from $69 a month and an Unlimited plan at $1,279 a month with a 3-month minimum, as of November 2026. Smart Pricing needs a 3-month commitment because the team builds a custom integration.
This is the kind of test Optimize could never run well. Dave Diederen tested shipping price levels and found a $9.49 shipping fee beat the control by 43.81% revenue per visitor. Josh Thompson ran a new page design against the existing page in Intelligems and reported +137.26% conversion rate and +175.64% revenue per visitor.
Not for: stores that only want to test layouts and copy, where a cheaper app does the job.
Optimizely and AB Tasty
Both were named by Google as GA4 integration partners, and both are aimed at large companies with sales-led pricing. If you're weighing them, see our list of Optimizely alternatives. Most Shopify stores under enterprise size will find the tools above a better fit. Our full A/B testing tools roundup covers more options.
Migration tips
Switching tools is a good moment to fix habits that Optimize let slide. These steps keep the move from costing you weeks.
1. Export what you can and list your old winners
Optimize data is gone from the product, so use whatever exports or notes you saved. Write down each past test, the result and the date. That becomes your backlog of tests to re-run.
2. Re-test old winners instead of assuming them
A winner on one store, or in one year, can lose later. Dave Diederen saw +3.16% revenue per visitor from adding a countdown timer to checkout, but a later checkout timer test lost 7.1% revenue per visitor. Kasper Dolk removed a checkout timer for a Dutch audience and saw conversion rate rise 25.61%.
3. Check the new tool's page speed and flicker
Client-side tools load the variant in the browser, and how a page is built can change results on its own. Oliver Kenyon rebuilt a landing page natively in Shopify and tested it against the same page made in a page builder. The native build won by 9.9% conversion rate and 20.9% revenue per visitor. Run an A/A test (the same page in both arms) in your first week to confirm the new tool splits traffic evenly and reports no false winner.
4. Connect GA4, but pick one source of truth
Most tools send experiment data to GA4, though Shoplift only includes the GA4 integration from its Advanced plan up. GA4 and your testing tool will count visitors a little differently. Decide up front which one decides the winner, and choose one main metric, ideally revenue per visitor.
5. Run tests for full weeks
New tools tempt people to call results early. Ruben saw +17% conversion rate after one week on an authority badge test, but after four weeks the lift was only 1.66%, which was most likely noise. Use our sample size calculator before you launch, and plan to run at least two full weeks.
6. Start with tests that already have evidence
Your first tests in the new tool should be ones with a track record, so you learn the tool on likely winners. Look for changes in our test database that won on more than one store, and remember the timer results above: even common winners can lose.
@oliverkenyon
@DaveDiederen@DaveDiederen@kasper_dolk@DaveDiederen
@Rubendgrf