Shopify A/B testing · Shopify split testing · Done for you

Shopify A/B testing: prove what makes people buy before you change your store

Shopify A/B testing shows half your visitors the current page and half a new version, then lets real orders decide which one makes more money. I run the whole programme for you: research, test design, AI-built variants, launch, analysis and rollout of the winners. One five-star review under Add to Cart, tested on almost a million visitors, added $67,365.

+$67,365from one review under Add to Cart
+$87,665from a main product video
60+A/B tests in 12 months for one brand
90%minimum confidence before I call a winner
In partnership with Variant
$128Mrevenue measured in A/B experiments
1,200+active A/B tests
2.8Mconversions measured
PandoraKarma and LuckDanielle GerberChatelier CosmeticsMokoBokoTimiTomiLifestyleCamper

Why A/B test instead of just changing the page

Every Shopify store gets advice: add trust badges, add a countdown, shorten the page. Some of it works on your store, some of it does nothing, and some of it costs you orders. On one store, a review block that “should” have helped cut orders by more than half in its first week. Without a test, nobody would have known.

An A/B test removes the guessing. Same ads, same prices, same week, two versions of the page. The one that makes more money stays.

Real Shopify A/B tests and their numbers

TestResult
One five-star review under Add to Cart+8% revenue per visitor, +$67,365
Video instead of a photo as the main product image+16% revenue per visitor, +$87,665
Price raised by 7.4%Conversion slightly down, +$37,219 revenue
Honest low-stock counter across the journey+5.84% cart-to-purchase, 98% confidence, ~$68,000
Customer videos on the mobile homepage+33% revenue per visitor, +$58,094
A “Selling fast” tag based on real sales+17.1% revenue per visitor, +$18,534
Trust box under Add to Cart (high-ticket store)+6%, two orders: not proven, still running

The last line matters as much as the first. Most tests don’t win, and a programme that hides its losers isn’t testing honestly. (Six tests, winners and losers.)

What I test on Shopify stores

  • Product pages (product page optimization): main image or video, reviews near the button, description placement, size guides, delivery promises, bundles. Usually the biggest wins. (Description above the button: ~$760K a year.)
  • Collection pages: sort order, filters, product card information, badges.
  • Homepage: hero message, social proof, brand and customer videos, navigation to bestsellers.
  • Cart: free-shipping progress, upsells, payment logos, express checkout buttons.
  • Checkout (Shopify Plus, checkout optimization): trust, delivery dates, upsells. A rebuilt checkout took one brand from $400K to $530K a month. (Checkout customization.)
  • Offers and prices: price points, bundles, free-shipping thresholds, discount types. (The price test.)

How the programme works

  1. Research. AI reads your analytics, thousands of session recordings and every review and support message, and groups the reasons people almost didn’t buy.
  2. Plan by money and traffic. Each idea is ranked by the revenue it could move, and each test gets a sample size and a planned length before it starts.
  3. Build. AI builds the variant in your store’s style in hours; I review every one on mobile and desktop.
  4. Run 50/50. No peeking and no early calls. A test that is clearly losing is stopped early to protect your sales.
  5. Decide on revenue. Revenue per visitor on most stores; conversion rate on high-ticket stores, where one big order can distort revenue.
  6. Roll out and stack. Winners go live for everyone and the next test builds on them.

Shopify Rollouts vs a testing app vs done for you

Shopify Rollouts (built in)A/B testing appDone for you (me)
CostIncluded in some plansFrom a small monthly fee, rising with trafficFrom $4,000 a month (app cost separate)
What you can testTheme versionsPages, prices, offers, often checkoutEverything an app can test
Statistical confidenceNot shown at the time of writingUsually shownPlanned before the test, reported after
Who builds and decidesYouYouAI builds, I decide and check
Best forSimple theme tests on stores with a teamStores with an in-house CRO personBrands doing $100K+ a month without a CRO team

How much traffic you need

Visitors needed per side, at 95% confidence and 80% power:

Your conversion rateTo prove +20%To prove +10%
1%~43,000~163,000
2%~21,000~81,000
3%~14,000~53,000

Less traffic? Test bigger changes on your busiest pages, run one test per page, and plan longer tests. More in how AI A/B testing works.

Who this is for

A good fit if you: make at least $100,000 a month, get 10,000+ visitors a month, and can commit to at least six months of testing.

Not a fit yet if you: have little traffic, want results in a week, or don’t want to change anything on your site. Start with the Shopify CRO audit checklist and the free calculators instead.

Done-for-you Shopify A/B testing starts at $4,000 a month: CRO pricing. How AI makes it faster: AI CRO agency.

Alexander Skibinskiy, Shopify CRO consultant
Who you'll work with

Alexander Skibinskiy

17 years in digital marketing and e-commerce. Head of growth at Karma and Luck, a nine-figure brand, where conversion rate grew 50% in one year of A/B testing. Founder of a seven-figure e-commerce brand. 100+ brands helped since 2015, including Danielle Gerber (0 to a multi-million brand) and Chatelier Cosmetics (0 to 7 figures, then acquired). Record with AI CRO for one brand: $7M extra revenue, +91% conversion rate, 60+ tests in 12 months.

Watch the free training →

Frequently asked questions

How do you A/B test on Shopify?

Pick one page and one change that answers a real customer doubt, build the new version, and split traffic 50/50 with a testing app or Shopify's built-in Rollouts. Let it run for a planned length until each side has enough visitors, then judge it on revenue per visitor. Keep the winner and test the next idea.

Does Shopify have built-in A/B testing?

Yes. Shopify Rollouts lets you test a duplicated theme against your live theme. At the time of writing it can't test prices, discounts or the checkout, and its reports don't show statistical confidence, so most serious programmes still use a dedicated testing app.

How much traffic do I need to A/B test on Shopify?

It depends on your conversion rate and the lift you want to prove. At a 2% conversion rate, proving a 20% lift takes about 21,000 visitors per side; a 10% lift takes about 81,000. Stores with less traffic should test bigger changes on their busiest pages.

How long should a Shopify A/B test run?

At least two full weeks so every weekday is covered twice, and until the planned sample size is reached. Product page tests on high-ticket stores can need 6–12 weeks. Stopping a test early because it looks good is the most common way to ship a false winner.

What should I A/B test first on my Shopify store?

The page with the most visitors and the biggest drop-off, usually the product page. Start with what customers doubt most (fit, quality, trust, delivery), taken from your reviews and support messages, not with button colours.

Can you A/B test prices on Shopify?

Yes, with a testing app that supports price tests. In one test, a 7.4% price increase lowered conversion slightly but added $37,219 in revenue, because the higher price more than made up for it. Price tests need careful setup so every customer sees one price.

How much does Shopify A/B testing cost?

The testing app usually costs from a small monthly fee to enterprise plans, depending on traffic. Done-for-you Shopify A/B testing with me starts at $4,000 a month, including research, design, building, analysis and rollout.

Real tests and guides
More ways to work together