Apex vs Varify: 2026 Comparison for E-Commerce A/B Testing

01How Do Apex and Varify Compare at a Glance?

Most tool comparisons treat A/B testing software as a feature checklist: editor, targeting, segments, reports. That framing hides the expensive problem. In a normal e-commerce testing program, about 1 in 5 tests produces a real winner. Every other test still costs design time, developer time, QA, and two to four weeks of traffic. The tool decides how fast you ship. Your idea selection decides how much of that shipping was worth doing.

Apex and Varify sit on opposite sides of that split. Varify.io removes the shipping bottleneck: a marketing team can build and launch a test in the browser without touching code. Apex attacks the selection bottleneck: it scores a test idea against a test memory of 4.3 million A/B tests before anyone builds it. The table below is the honest short version, including the rows where Apex has nothing public to show.

Two rows deserve a note. Where the Apex column says “not publicly documented,” we have chosen not to fill the gap with marketing language. And the pricing row is not evasion: Apex is sold with managed execution, so the scope is set in a call rather than on a pricing page. Varify publishes its prices, which is a genuine advantage if you want to buy a tool today without talking to anyone.

02What Does Apex Do That Varify Does Not?

Apex has one argument, and we would rather state it plainly than bury it in a feature list. Apex predicts which tests win before they go live, from a test memory of 4.3 million A/B tests from 151,000 shops, collected over eight years. That memory is one of the largest A/B test databases in e-commerce, and it is the only reason Apex exists as a separate product.

Concretely, that changes the order of operations in a testing program. Instead of prioritizing a backlog by gut feel or by a scoring framework where the numbers are opinions, each idea is scored against what happened when comparable changes ran on comparable shops. The output is not a promise. It is a prior: this idea class has a track record, that one does not, and this third one has a track record of quietly losing.

The three layers, and what each one is for

• Test memory: 4.3 million A/B tests from 151,000 shops, collected over eight years. It scores every test idea before launch, so the backlog is ranked by evidence instead of enthusiasm.
• Testing tool: build, launch, and evaluate A/B tests directly in the shop. The prediction and the execution live in the same place, so the prediction is checked against the result every time.
• Managed execution: tests are built, QA’d, launched, and analyzed with the DRIP team. This is the part no self-service tool sells, and it is why Apex pricing is set in a call.

Varify does not compete on this axis and should not be criticized for it. A no-code editor and a test memory are answers to different questions. If your team already knows which tests are worth running, the Apex argument is worth much less to you, and a cheaper self-service tool is the rational purchase.

03Where Is Varify the Better Choice?

Varify.io is a genuinely good product, and we recommended it before Apex existed. Its design decisions are coherent, and two of them are better than what most enterprise platforms offer.

No developer in the loop

The browser-based visual editor lets a marketer, CRO specialist, or product manager change text, images, layout, and colors on the live page and publish that as a variant. After the one-time snippet install, the marketing team is autonomous. If your developers are fully committed to product work, that autonomy is worth more than any statistical feature, because a test that never gets implemented has a win rate of zero.

No own tracking, which is a privacy feature

Varify deliberately does not build its own analytics engine. Results are read from Google Analytics 4, Matomo, or Piwik Pro. Three consequences follow, and all three are good ones: no additional cookie consent category on your banner, no conflicting numbers between the testing tool and the analytics tool, and one less client-side script on the page. For European shops with a strict privacy posture, that architecture removes a recurring argument with legal.

Published, flat pricing

The Growth plan is €149 per month, the Pro plan €249 per month, both with unlimited traffic and unlimited experiments, and there is a 30-day free trial. No traffic tiers, no per-user fees, no procurement cycle. Among professional testing tools that is close to the low end of the market, and it removes the budget excuse for not testing at all.

04What Actually Predicts Whether a Test Wins?

This is the part of a tool comparison that only volume can produce, so here is what our own data says. DRIP has run more than 4,000 experiments for more than 250 e-commerce brands. In 2024 our win rate was 27%, not far above the industry pattern where about 1 in 5 tests produces a real winner. In the most recent quarter it was 55%. We did not double the number of tests, and we did not change testing tools to get there.

What changed was the rejection rate. The lift came almost entirely from ideas we killed before they consumed design, development, QA, and traffic. The reliable signal was never the element itself. Trust badges, urgency timers, and image galleries all have both winners and losers in the record. The signal was context: had this change class already won on shops with a similar traffic mix, price point, and page type?

• Change class beats element: “Reduce decision cost on the product page” has a track record. “Make the button green” does not, and never will, because the same element wins on one shop and loses on the next.
• Context decides the sign: the same change frequently flips direction between a high-consideration, high-price catalog and an impulse catalog. Prior outcomes on comparable shops carry that information. A hypothesis document does not.
• The best output is a rejection: the most valuable thing a test memory returns is the list of ideas you do not run. Nobody celebrates it, and it is where the win rate actually comes from.

Note what this argument does not claim. A prediction is a prior, not a guarantee, and our win rate is our own program’s record rather than a promise for any single shop. It also does not make Varify’s editor less useful. It says the editor was never the constraint on results.

05Who Runs the Tests, Your Team or an Expert Team?

The operating model matters more than the feature list, and it is where these two products separate most cleanly.

With Varify, your team owns everything after the snippet is installed. That is the appeal, and it also means your team owns the failure modes: a variant that breaks on one mobile browser, a goal wired to the wrong event, a test called after nine days because the numbers looked good on day nine. None of these are Varify problems. They are the standard failure modes of any self-service testing program, and they are the reason a tool with a great editor can still produce a flat year.

Apex includes managed execution: tests are built, QA’d, launched, and analyzed with the DRIP team. In practice, that means the analysis judgment travels with the tool instead of being an internal hiring problem. It is also the honest reason we do not publish an Apex price. Managed scope is not a per-seat number, so the fit is worked out in a call.

06Our Verdict: Should You Choose Apex or Varify?

We sell Apex, so read the recommendation with that in mind. The reasoning is simple enough to check: if about four of five tests fail industry-wide, the highest-leverage improvement available to a shop is picking better tests, and picking better tests requires outcome data at a scale no single shop can generate. That is the case for Apex, and it is the only case we make for it.

Choose Apex if…

• Your win rate sits near the industry base rate and more tests have not fixed it
• You have limited traffic, so every inconclusive test is an expensive month
• You want each idea scored against comparable shops before anyone builds it
• You want tests built, QA’d, launched, and analyzed with an expert team
• Your backlog is long and your confidence in its order is low

Choose Varify if…

• Your marketing team wants to launch tests independently, starting this week
• You need published, predictable pricing and no procurement conversation
• You want zero additional cookies and results read from GA4, Matomo, or Piwik Pro
• You already have in-house testing and analysis skill, so selection is not your bottleneck
• You are running your first structured testing program and want a low-risk start

One last honest note. Varify publishes its price and lets you try it for 30 days. We ask you to book a call. That is a real friction difference, and if you want to buy software without talking to a human today, Varify respects that preference and Apex does not.