Overview
Where we came in
A direct-to-consumer home fragrance brand with a strong repeat rate and a leaky first purchase. Traffic quality was fine and product pages were converting to cart at a healthy rate. Everything went wrong between the cart and the payment screen, and it went wrong at the same moment for almost everyone.
The baseline problem
- Delivery cost first appeared on step three of four, after the customer had entered a full shipping address.
- Exit timing analysis showed a spike within four seconds of the shipping line rendering.
- The free delivery threshold existed but was never mentioned until the total was already calculated.
- Cart abandonment sat at 51.8%, with mobile ten points worse than desktop.
Hypothesis
What we expected, and why
Written before the test was built. Because we observed X, we expect Y, measured by Z.
Because the total cost is concealed until the customer has already invested effort in the checkout, we expect that showing delivery cost and the free delivery threshold from the cart onwards will increase checkout completion and raise average order value toward the threshold.
Experiment plan
How the test was set up
Sample size and stop date were both fixed before launch, so no result here was called early.
Two-arm split test across the full checkout. Control kept cost reveal at step three. Variant showed an estimated delivery cost in the cart drawer, a persistent order summary on every checkout step, and a progress line showing the remaining spend to free delivery.
- Primary metric
- Checkout completion rate, cart to order confirmation
- Traffic
- ~24,100 sessions per arm
- Duration
- 18 days, covering two full weekly cycles
- Guardrails
- Average order value, to make sure completion gains are not just cheaper baskets
- Return rate over the following 60 days
- Add-to-cart rate, in case the cart drawer change suppressed adds
Results
What actually moved
| Metric | Control | Variant | Change |
|---|---|---|---|
| Checkout completion rate | 48.1% | 59.0% | +22.7% |
| Mobile checkout completion | 41.3% | 53.7% | +30.0% |
| Average order value | £42.10 | £46.80 | +11.2% |
| Orders above free delivery threshold | 28.4% | 41.9% | +13.5pp |
| 60-day return rate | 4.2% | 4.1% | -0.1pp |
Order value went up as well as completion, which does not always happen. The threshold progress line gave people a reason to add one more item, and the returns data over the following two months showed those extra items were not regretted purchases.
What changed
Shipped to production
- Estimated delivery cost now shows in the cart drawer, before checkout starts.
- A persistent order summary sits on every checkout step, on mobile as a sticky collapsed bar.
- A progress line states the exact remaining spend to reach free delivery.
- Checkout reduced from four steps to three by merging contact details into the address step.
Next experiments
Already in the queue
- 01Test a delivery date estimate alongside the cost, since speed is the second most common pre-purchase question.
- 02Test express payment buttons at the top of the cart drawer for returning customers.
- 03Test threshold-aware product recommendations in the cart.