Mobile browsing
5 tests · 2 winners
+0.4% pooled effect, -0.3% to +1.2%
Medium confidence. Enough tests, but the pooled effect is still compatible with no effect.
A category page shapes the choice before the product page.
Sorting, filters, and featured products: what decides which products get looked at. The examples come from the demo store.
- Sort by popularityProduct clickNot conclusive yet…day 3 of 14+1.5%, range -2.8% to +6.1%
Maison Lune, a fictional demo store · data computed by the engine
Test a category page on what it shows first.
Visitors browse a list. Tests here target the order, sorting, filters, and what comes with each product.
- Default sortNewest, most popular, price: the order the list appears in.
- FiltersWhich ones to offer, and where to put them.
- Featured productsCurated picks, flash sales, or new arrivals at the top of the list.
- Price displayPrice alone, crossed-out price, price including shipping.
- ThumbnailsPhoto size, products per row, information under each product.
- BreadcrumbsFinding your way through the departments and going back.
Maison Lune, a fictional demo store
The popularity sort test has just started.
In a test’s first days, the range is wide. That’s expected, and it’s why the plan sets no checkpoint until enough visitors have come through.
Sort by popularity: not conclusive yet.
Day 3 of 14, measured on product click. Today’s lift +1.5%, range -2.8% to +6.1%.
First checkpoint on day 5. Until then, no reading counts, whatever color the day shows.
If category pages are sorted by popularity, visitors will see the best-liked products first, because sorting by newest puts slow-selling lamps up front.
An extra click isn’t an extra sale.
On a product list, the fastest metric is the click through to a product. “Simplified filters on mobile” raised it by +4.0%: fewer filters, more products viewed.
Before going to everyone, the gradual rollout tracks revenue per visitor as a guardrail, so a gained click doesn’t cost an order. See this rollout.
Simplified filters on mobile
Conclusive. Proven gain+4.0%, range +1.4% to +6.6%Product click · Shipped
Category ideas are ranked before they’re tested.
Expected effect drawn from past tests on the same theme; duration based on category page traffic. A promotion idea inherits what promotions have delivered so far.
Category page trafficdemo store
3,900visitors a day
5.2%smallest effect measurable in 30 days on “Commande”
1.9%on “Product click”, which happens more often, in the same time
An idea with a smaller expected effect won’t reach a conclusion in a month, and Probant tells you before launch.
- Rank 6Delivered price shown on category pages
- Expected effect
- +1.7%prior2 pricing tests, too few
- Time to conclude
- 38 daysmeasuring product clicksinstead of 286 days on orders
- Estimated gain
- +€22,400per month
- Rank 16Sticky filters at the top of category pages on mobile
- Expected effect
- +0.4%based on 5 mobile choices tests
- Time to conclude
- Undetectableover a year on ordersmeasurable in 30 days from +1.9%
- Estimated gain
- +€5,700per month
- Rank 20Flash sales at the top of category pages
- Expected effect
- -0.9%based on 3 promotions tests
- Time to conclude
- —expected loss
- Estimated gain
- -€12,100per month
3 pending ideas, in Probant’s ranking order. Expected effect drawn from past tests on the same theme; duration based on the area’s traffic.
The library keeps the tests on product lists.
In the demo store, these tests are filed under the Navigation area: they covered sorting, filters, and the product list grid.
Lessons by theme
Promotions
3 tests · 1 loser
-0.9% pooled effect, -2.1% to +0.3%
Medium confidence. Enough tests, but the pooled effect is still compatible with no effect.
Pricing
2 tests · 1 loser
-2.2% pooled effect, -4.4% to -0.1%
Low confidence. Fewer than three tests: a lead, not yet a rule.
4 completed tests
- Simplified filters on mobileJul 23 → Aug 14 · Product click · ShippedConclusive. Proven gain+4.0%, range +1.4% to +6.6%
Fewer filters, more products viewed
- Breadcrumbs on category pagesMay 18 → Jun 8 · Product click · Original keptNo detectable effect.+0.3%, range -1.5% to +2.1%
Useful to a few visitors, with no overall effect
- Sorting cut to three options on mobileMar 30 → Apr 20 · Product click · Original keptNo detectable effect.-0.6%, range -1.8% to +0.6%
Fewer sort options neither helped nor hurt
- One product per row on mobileDec 8 → Dec 29 · Product click · Original keptNo detectable effect.-0.2%, range -1.6% to +1.2%
Bigger photos but fewer products seen, so no net change
A click on a product hints at a sale without guaranteeing one.
A category page is quick to measure on clicks through to a product. What matters is decided further on.
- Revenue per visitorTracked as a guardrail, it checks that a gained click doesn’t cost an order.
- PromotionsHeavy featuring draws clicks, sometimes toward smaller carts.
- SpeedBigger or more numerous thumbnails weigh down the page.
Demo store guardrails
Monitored during every test and every rollout step. A proven drop stops the test or sends visitors back to the original.
- Page speed (LCP)
- 75th percentile of page views · increase of +4% at most
The Probant features that help in this area.
Other use cases
The same engine, for other site areas and other teams. The demo opens on the screen this page describes.
- Open demo, no sign-up
- Free to start, no overage charges
- Database and compute in Amsterdam, with Scaleway