Skip to main content
inconclusive

Checkout: Optional or Confident Recommendation

Hypothesis

If we implement 'Optional or Confident Recommendation' on checkout pages (In this experiment, copy around an upsell was changed from using "optional" to "our recommendation for you"), then key conversion metrics will improve.

PricingCheckoutCross-Industrycopypersonalizationdesktop

Test Results

58,597
Sample size

Key Learning

Context: Friction during the checkout process causes users to abandon right when they're closest to converting.

What was tested: REAL-WORLD TEST: 'Optional or Confident Recommendation' was tested on a live checkout page. The test involved 58,597 real visitors. Full statistical results require paid access. Test methodology: In this experiment, copy around an upsell was changed from using "optional" to "our recommendation for you". The idea was to recommend two upsells wit...

Result: No statistically significant difference was detected. Pricing tests that are inconclusive may indicate the price itself isn't the issue — the perceived value or the framing might matter more.

How to Apply This to Your Site

This experiment tested checkout: optional or confident recommendation but produced no statistically significant change. The test was run on a checkout page in the cross-industry industry. Inconclusive results suggest this particular change may not be a priority — focus testing effort on higher-impact areas.

Before you test: Consider that pricing tests typically require large sample sizes to detect small effects. Run your test for at least 2 full business cycles to account for weekly traffic patterns.

What Was Tested

In this experiment, copy around an upsell was changed from using "optional" to "our recommendation for you". The idea was to recommend two upsells with more confidence. Impact on the two upsells (secure, and express) as well as overall transactions was measured.

Methodology

Confidence Level
70%

Build On These Learnings

Save your own experiments, spot winning patterns across your test history, and stop repeating what's already been tried.

Related Experiments

Explore More Experiments