Stripe-only checkout vs three payment options: three options won
Commentary byLiftArchiveOur analysis, not theirsCutting checkout down to Stripe card payments only did slightly worse than offering PayPal, Google Checkout and Stripe together.
Patrick McKenzie ran this test on Bingo Card Creator and published it on the Kalzumeus blog in 2012. The win for three options was weak, at about 90% significance, and no numbers were shared.
- Offering all three payment options beat the Stripe-only checkout. This counts as a loss for the variant.
- The win was weak, at about 90% significance.
- No conversion rates or sample sizes were shared.
The change
The variant removes PayPal and Google Checkout, leaving card only.
Simplified mockup drawn by LiftArchive from the shared screenshots.
Control
Checkout offered three payment methods: PayPal, Google Checkout and an on-site Stripe card form.
Variant
Checkout offered only the Stripe card form, with PayPal and Google Checkout removed.
Why it likely lost
Our reading of the result, not the practitioner's claim.
- Most buyers chose cards, but a small group may only buy if they can use a wallet like PayPal.
- Removing a familiar option can make some buyers leave rather than switch methods.
- A simpler checkout does not help if it removes the method a buyer trusts.
Caveats: The author first reported this as a null result and later corrected it. The win was only about 90% significant, no numbers were shared, and the data is from 2012.
Who ran it
Patrick McKenzie ran this test and published the results (Kalzumeus (patio11) blog). The numbers are theirs; the summary and commentary on this page are ours.
Read the original case study on Patrick McKenzie →
How to run this test on your site
- Hypothesis
- If we remove PayPal and other wallets and offer only card payment, then purchases will hold or rise because the checkout is simpler.
- Primary metric
- Completed purchases per checkout visitor
- Guardrails
- Checkout abandonment rate
- Payment failure rate
- Revenue per visitor
- Who should test it
- SaaS sellers thinking about cutting payment methods to reduce fees or complexity. If a meaningful share of buyers uses a wallet, expect to lose some of them.
Questions
Should I remove PayPal from checkout?
Test it first. In this case, removing PayPal and Google Checkout did slightly worse than keeping them.
Do fewer payment options increase conversions?
Not here. Even though most buyers paid by card, offering more options won weakly.
How many payment options should a SaaS checkout have?
Enough to cover the methods your buyers actually use.
Check your payment mix and test before removing any method with real usage.