# Fast home-built cart replacing hosted cart: +94% conversions

- Outcome: Won
- Page type: Checkout (SaaS)
- Elements: Checkout steps, Friction
- Date shared: 2009-02
- Test run by: [Patrick McKenzie](https://liftarchive.com/practitioners/patrick-mckenzie/)
- Commentary by: LiftArchive (our analysis, not theirs)
- Original source: https://www.kalzumeus.com/2009/02/17/ninety-four-percent/
- Page: https://liftarchive.com/test/patio11-bcc-new-shopping-cart/

Replacing a slow hosted shopping cart with a fast, home-built lightbox cart lifted purchase conversions by 94%. Patrick McKenzie ran this split test for Bingo Card Creator and reported it on the Kalzumeus (patio11) blog. Several cart changes shipped together, so the lift can't be credited to any one of them.

## Result

**+94% conversions**

- Purchase conversions: +94% for the new cart.
- In a later post the author said they had reset the test and still saw roughly a 100% improvement.
- Sample size, duration and confidence level were not shared.

## Control

The hosted e-junkie shopping cart. It was slow to open and included Continue Shopping and Update Cart steps.

## Variant

A fast lightbox-style cart built in-house, with prices calculated in the page, no Continue Shopping or Update Cart buttons, and product branding instead of shopping-cart wording. Payment was still processed by e-junkie.

## Why it likely won (our reading)

- A cart that opens instantly gives buyers less time to hesitate or give up.
- Removing Continue Shopping and Update Cart cut steps that made no sense for a single software product.
- Branded wording may have felt more like buying the product and less like a generic store checkout.

## Caveats

The results post shares only the headline number, with no sample size or confidence. Several cart changes were bundled, and Bingo Card Creator was downloadable desktop software in 2009, not SaaS.

## How to run this test

- Hypothesis: If we replace our slow, generic checkout with a fast, single-product checkout without extra cart steps, then more buyers will complete payment because there is less delay and fewer decisions.
- Primary metric: Completed purchases per checkout start
- Guardrail: Refund rate
- Guardrail: Payment errors
- Guardrail: Average order value
- Who should test it: SaaS and software sellers with one main product who use a generic hosted cart. It matters less if you already use a fast, streamlined checkout built for subscriptions.

---

Credit: Patrick McKenzie ran this test and owns the result. Summary and commentary by LiftArchive (https://liftarchive.com/test/patio11-bcc-new-shopping-cart/).
Usage: Free to use, quote and build on, with credit and a link: name the practitioner or organization that ran the test and link to its LiftArchive page. The results belong to the practitioners who ran and shared them; the summaries and commentary are LiftArchive's.