Test it before you ship it.
You shipped the new product page. You still don't know if it sold more.
Experimentation is 4 A/B tests a month on your store. On Shopify I build and launch all 4 myself. On other platforms I design and spec all 4, check your build before it goes live and analyze every test you launch. Each test starts from your existing data and a written hypothesis. I measure what happened before a change becomes permanent. €6,799 a month, the same price as the Design Subscription.
Call: keep variant B
Your conversion rate moved last month. Nobody can tell you why.
A new hero image went live on a Monday. A promo started on Thursday. By the end of the month the number had changed, and every change gets the credit or the blame depending on who you ask. So the next decision gets made on taste again.
? Which change moved it
A fix isn't finished when it ships. It's finished when the data agrees.
An A/B test shows half your visitors the current page and half the new one, over the same days, with the same ads and the same promo running. The change is the only difference, so the result belongs to the change.
How it works
-
I start from your data
Your analytics and any research you already have show me where shoppers drop off and which pages get enough traffic to test.
Funnel drop-offExample dataProduct page → cart−62%Cart → checkout−31%Checkout → paid−18%Testable traffic / month24,300 -
I write the hypothesis
Every test starts with one written hypothesis: what I change, for which shoppers, what should move and why I expect it to. No hypothesis, no test.
Hypothesis H-07Example dataChangeDelivery date above the foldForMobile shoppers on product pagesShould moveAdd-to-cart rateBecauseSession recordings show scrolling for shipping info -
The test goes live
On Shopify I develop the variant and launch it myself. On other platforms the test gets built the way you decide, in the tool you choose.
Live on ShopifyExample dataDay 9 of the testA · 50%6,112 visitorsB · 50%6,087 visitors -
I measure the result
I analyze the result and you get a clear call on whether to keep the change.
ResultExample data+12.6%Keep the change
One plan. 4 experiments a month.
Same price as the Design Subscription. Different work.
- 4 A/B tests per month. On Shopify they are launched; elsewhere they are designed and specced. A test counts once it goes live, win or lose.
- Every test starts from your existing data
- A written hypothesis behind every test
- On Shopify: development, launch and analysis by me
- On other platforms: design, a build-ready spec, a QA check before launch and analysis of every test you launch.
- Every result measured and analyzed
€6,799
Book a callShopify or another platform
Shopify: I run all of it
I develop the test, launch it and analyze the result. Your developers don't have to build anything for the test itself.
WooCommerce, Magento and other platforms
You decide which testing tool to use, and your developers build each test. I write the hypotheses, design the variants and send a build-ready spec. I check your build on staging before it goes live and analyze every test you launch.
Is this for you?
A/B tests need traffic. About 10,000 monthly visitors on the page you're testing is enough for click and add-to-cart tests. Judging a test on purchase rate takes roughly 25,000 to 35,000 sessions per tested page per month. Four tests a month works when they run in parallel on different page types. Below that, a structured audit gets you to an answer faster. Start with a QuickScan (€995, ready in 48 hours) or a Research audit.
Good fit if you...
- Have about 10,000 monthly visitors or more on the pages you want to test
- Already collect data in your analytics
- Want a measured result before a change becomes permanent
- Run Shopify, or have a team that can build tests on your platform
Not a fit if you...
- Have fewer than about 10,000 monthly visitors per page
- Mostly need a steady flow of new designs (that's the Design Subscription)
- Expect every test to win
Frequently asked questions
Not sure your traffic is enough?
Thirty minutes, free. Tell me which platform you're on and roughly how many visitors your key pages get. I'll tell you honestly whether 4 experiments a month is worth it for you, or whether an audit gets you there faster.
Book a call