Varify: A/B testing, personalisation and now the Shopify checkout too

Varify is a German tool for A/B testing and personalisation on your own website. You build variants in a visual editor on top of your existing pages, set goals and read out what works better — without a developer needed for every test. We are a Varify partner and use the tool with clients who want to test seriously without an enterprise budget.

Want to look at the tool yourself? You can via our partner link: varify.io. We also compare the tool with the larger platforms in A/B testing tools compared.

What Varify does

• Visual testing. Adjust copy, buttons, blocks and images in an editor over your own site, and pit that variant against the original.
• Split URL tests. Two separate pages against each other, for example an old and a new landing page.
• Personalisation. Show a variant only to part of your visitors: new versus returning, device, source, country.
• Custom goals. Set clicks, forms, purchases, scroll depth or a custom event as the goal.
• Links to your analytics. Send results to GA4, so your reporting stays in one place.
• Cookieless measurement. The tool works without tracking cookies, which makes setting it up under the GDPR a lot simpler.

Now also testing the Shopify checkout

New, and the most important thing for webshops: since May 2026, Varify has its own Shopify app that lets you adjust and test the checkout. You place components in the checkout flow — think of a USP row, a free-shipping progress bar or extra explanation at a payment step — and launch them as an A/B test or directly for all visitors.

That's a bigger deal than it sounds. The Shopify checkout was practically closed off for years: customising it required Shopify Plus, and testing in practice meant waiting, hoping, or not testing at all. Now the last step of the funnel has become reachable without custom development. The front-end tests (homepage, category, product, cart) work on all Shopify plans, so Shopify Plus is not a requirement there.

The checkout app is a separate app with its own subscription (around $119 per month at the time of writing), alongside your Varify account for the rest of your site. Do check the current price yourself; it changes regularly.

What kind of results you can expect in the checkout is a story of its own: they're structurally smaller than on a product page, while in euros they often deliver more. We explain that in A/B testing in the checkout.

Why we're a partner

We don't work with everything that exists, but with a limited number of tools we use ourselves and whose people we know. With Varify there are three reasons:

1. The price point. For most SMEs and webshops this is the difference between testing and not testing. The enterprise platforms start where many businesses' test budgets end.
2. It fits marketers. The editor is built for people who want to build a variant themselves. That matches how we work: we set it up, you can take it over. See also DIY.
3. The checkout step. With the Shopify app, the tool now also covers the part of the funnel where the money actually comes in.

Being a partner, for us, means: we know the tool inside out, we see updates earlier, and we can move faster on questions. Our links to Varify are partner links. That changes nothing about the advice — if another tool fits better, we say so.

When Varify is not the right choice

Honest about the limits:

• Too few visitors. Below roughly a few hundred conversions a month, every test takes so long that you're better off working on your offer, copy and speed first. Check this with the significance calculator.
• Server-side testing and feature flags. If your test sits in application logic rather than the display layer, look at VWO or Optimizely instead.
• Heavy governance. Large organisations needing roles, rights and audit trails per team end up at the enterprise suites.
• No one owns it. A testing tool with no one thinking up tests and reviewing results monthly is a subscription without a return.

How we deploy it

With clients, we usually approach it like this:

1. Measurement first. Goals, events and conversions correct in GA4 and in the tool, before a single test goes live. See measurement and data.
2. A test list based on figures. Where does the funnel leak, what do we see in session recordings and heatmaps, where do customer service questions cluster. Not: "let's make the button green".
3. Big tests first. Message, offer and page structure give results you can actually see. Detail tests we save for pages with high volume.
4. Set out in advance what we measure and for how long, so a test isn't cut short mid-way just because it happens to be doing well.
5. Document outcomes, including the tests that did nothing. That's the knowledge you keep.

More on the wider approach is in increasing conversion and conversion optimisation. Want to know more about webshops and the technology behind them, take a look at webshops and websites.