Should we personalize the home page for returning customers?
Our answer
Keep the home page built for new visitors. Add a row of products the visitor viewed before, shown only to visitors with history, and test a returning-visitor layout on revenue per visitor, read by segment, before keeping it.
- Confidence
- Low
- Strongest proof
- 5 controlled tests
- 1 official rule, 1 dataset, 1 opinion or claim
- Sources
- 5 sources
- 8 statements
- CCBC
O
New here? Take the 1-minute tour
Why this is the answer
3 reasons and our judgmentMost home page visitors are new. They are often 60 to 80% of home page traffic (Intelligems), and new visitors made up 77.6% of sessions in one store's test (Blend Commerce).
A row of products a visitor viewed before can lift revenue without touching new visitors. Blend Commerce measured a 13% lift in one store's revenue per visitor from such a row, hidden from visitors without history.
A full returning-visitor layout can pay more, though results vary. Clean Commit measured a 44.3% lift in revenue per visitor from one, while its viewed-products section on product pages moved conversion only 1.0%.
Our judgment
We would personalize only for visitors with history and leave the default page for new ones, since most visitors are new and the row that worked stayed hidden from them. We start with the viewed-products row because it is the one change with a clear home page result. Clean Commit's larger layout tests include one that never reached significance, and its product page row was inconclusive. Read every personalization test on revenue or profit per visitor, split by new and returning visitors.
Our judgment, built from Intelligems, Blend Commerce and Clean Commit: our reading of these sources, not a finding from any one of them.
What the sources say
8 statements from 5 sourcesIn one Blend Commerce client test, a strip of value points (guarantee, returns, fast shipping, free shipping threshold) under the homepage banner lifted conversion 5.82% and revenue per visitor 7.76%.
“The variant increased conversion rate by 5.82%, revenue per visitor by 7.76% and average order value by 1.83%.”
Blend Commerce says a recently viewed products carousel on one coffee brand's homepage raised revenue per visitor and conversion rate 13% each. No sample size, length or significance given.
“Variant 1 increased revenue per visitor by 13% and eCommerce conversion rate by 13%, while subscription orders per visitor rose 2%.”
Intelligems, which sells testing tools, estimates 60 to 80% of homepage visitors are new and suggests A/B testing a new-visitor homepage, judged on profit per visitor and split by visitor type.
“Blended numbers across both segments can hide the real signal.”
Shopify's Rollouts, under Markets > Rollouts in the admin, can schedule, gradually publish and A/B test themes, checkout and customer account setups, including two entirely different themes.
“between two entirely different themes or checkout setups to find the winner”
Clean Commit, which sells conversion work, reports one store's personalized homepage for returning visitors lifted conversion 29.9% and revenue per visitor 44.3%. No sample size, length or significance given.
“Highlighted product offers for returning visitors.”
Clean Commit's test of a recently viewed products section on one store's product pages was inconclusive: conversion up 1.0% and revenue per visitor up 1.7%, no meaningful uplift.
“Simplifying the shopping journey didn't lead to any meaningful uplift.”
Optimizely reports three unnamed retailer homepage tests: visitor-specific recommendations lifted clicks to product pages 136%, but a broad outfit carousel cut add-to-cart 6.07%, with no sample sizes given.
“Serving full head-to-toe outfits on the homepage was overwhelming for visitors, decreasing engagement.”
Clean Commit, which sells conversion work, showed new visitors an educational homepage and returning visitors a sales layout on one store. It reports large lifts, but the test did not reach significance.
“Other tests and holiday promotions unfortunately got in the way of this test reaching significance.”
In the full answer
For readers who sign in- What to do this week3 steps
- The numbers behind it3 figures from cited sources
- Where each source standsA chart placing each source on the question
- The strongest disagreementThe best case against the answer, from a cited source
- How sure we areWhat its confidence rests on, and what would change the answer