As D2C Product Ops
I wasn’t convinced the newer checkout was better.
- Test
- Four weeks, in steps
- Traffic
- 250 to 300 orders a day
- Decision
- Mine
The situation
Our checkout runs on a third-party provider. We were on its newer version. Most other brands using it were still on the older one. The easy assumption was that newer converts better.
What I did
For four weeks I moved shoppers between the two versions in steps: 10% to the older version, then 25%, then 50%, then everyone.
- Step 1
- Step 2
- Step 3
- Step 4
older versionnewer version
Result
The older version won. Checkout conversion rose 12 to 18%. That is a relative rise, not percentage points.
My part
The decision was mine, made from where shoppers dropped out of the funnel.