Auto-scroll to recommendations after add to basket: +1.2% AOV
Commentary byLiftArchiveOur analysis, not theirsMaking the add-to-basket button scroll shoppers down to the recommended products raised average order value by 1.2% and clicks on recommendations by 23%.
Good Growth ran this A/B test on the Pets at Home product page, and it was published as a VWO case study.
It reached statistical significance after 3 weeks on desktop and mobile.
- Average order value: +1.2%. A small lift, but it applies to every order, so it adds up for a large retailer.
- Clicks on recommended products: +23%. More shoppers saw and used the section once they were taken to it.
- The result was statistically significant after 3 weeks, across desktop and mobile traffic.
The change
Adding to basket jumps the shopper down to the recommendations at the bottom.
Simplified mockup drawn by LiftArchive from the shared screenshots.
Control
The product page had a recommended products section at the very bottom. Most shoppers added an item to their basket and never scrolled far enough to see it.
Variant
Clicking add to basket triggered an anchor link that moved the page down to the recommended products. Nothing else on the page changed.
Why it likely won
Our reading of the result, not the practitioner's claim.
- The recommendations were already there. The test only fixed the fact that few people saw them.
- Right after adding an item is a natural moment to think about what else is needed, like food plus a toy or bedding.
- Pet supplies have many related items, so suggestions are more likely to be useful than in a single-item category.
Caveats: The source does not share sample size or the exact confidence level. The AOV lift is small, so it needs high traffic to detect.
Who ran it
Good Growth ran this test for Pets at Home and published the results (VWO case study). The numbers are theirs; the summary and commentary on this page are ours.
Read the original case study on Good Growth →
How to run this test on your store
- Hypothesis
- If we scroll shoppers to the recommended products right after they add an item to the basket, then average order value will rise because more people will see related items they would otherwise miss.
- Primary metric
- Average order value
- Guardrails
- Conversion rate
- Checkout starts from the product page
- Bounce or exit rate after add to basket
- Who should test it
- Stores with a large catalogue of related or complementary items and recommendations placed low on the product page. It probably won't help if shoppers usually buy one item, or if the add-to-basket action already opens a cart drawer they rely on.
Questions
Does auto-scrolling a page annoy shoppers?
It can, if it moves people away from something they still need.
Here it happened only after the item was added, so the main task was already done.
Watch checkout starts to make sure people can still find the basket.
Where should product recommendations go on a product page?
Wherever shoppers will actually see them. If your data shows few people reach the bottom of the page, move the section up or bring people to it after a key action, as this test did.
How much traffic do I need to test average order value?
Usually a lot. AOV varies a lot from order to order, so small lifts like 1.2% take many orders to confirm.
Similar tests
@DaveDiederen@DaveDiederen@DaveDiederen
Blend CommerceBlend CommerceBlend Commerce