@joshtCRO
@RubendgrfWhy variant selectors matter on the product page
The variant selector sits at the moment of highest intent: the shopper wants the product and now has to pick a size, color, flavor or pack.
Dave Diederen's argument is that a dropdown hides important information behind an extra tap, right when the path to add-to-cart should be shortest.
Size doubt is the other problem. Several tests here are less about the control itself and more about helping shoppers answer "which one is right for me?"
Measured results
Six measured tests, six reported wins, no reported losses.
Dave Diederen ran five and Oliver Kenyon one.
Selector-only changes
Oliver Kenyon's test is the cleanest in the collection.
Heatmaps showed shoppers clicking their gender first and then looking for a size, so the team added Men's and Women's tabs and showed only the relevant sizes.
Nothing else on the page changed, and the result was +10.4% conversion rate, +12.9% revenue per visitor and +12.8% profit per visitor.
Dave Diederen's dropdown-to-blocks test produced a smaller but clear lift: +3.95% conversion rate, +3.79% revenue per visitor and +3.95% profit per visitor, AOV unchanged.
Dave Diederen also notes that blocks leave room for labels such as Best seller, Sold out or Low stock.
Changes around the selector
- Model size overlay on fashion images ("Elly is 5'8" and wears a size S"): +2.63% conversion rate, +4.61% revenue per visitor, +1.93% AOV (Dave Diederen).
- Sticky add-to-cart bar that scrolls back to the selector when a variant or bundle must be picked: +$32,565/month (Dave Diederen). No conversion rate figure was shared.
- New sections under the add-to-cart button: +14.19% conversion rate, +11.32% revenue per visitor (+A$52,743/month), +11.22% profit per visitor (Dave Diederen).
- Swapping the model in the first product image: 9.55% conversion rate for the blonde model against 9.32% for the control, about +$18,828/month (Dave Diederen).
@oliverkenyon
@DaveDiederen@DaveDiederen@DaveDiederen@DaveDiederen@DaveDiederenDave Diederen also reports that placing a countdown timer above the variant or bundle selector "won repeatedly", but shared no numbers, so it is not counted here.
Losing or flat tests
No test lost overall, but parts of them did.
In the model swap test the brunette variant fell to 9.01% conversion rate, below the 9.32% control.
The sections-under-the-button test won on conversion rate while AOV dropped 2.52%, and an earlier version of it reported a smaller +3.18% conversion rate and +6.57% revenue per visitor.
A record of only wins is also a sign of what gets shared.
Treat it as a list of things worth testing, not a guarantee.
Untested ideas
Most of the 46 entries are ideas without data:
- Swatches and blocks. Kasper Dolk suggests big, spaced image swatches with an obvious selected state, and a label with the variant name and color code that updates on switch, after recordings showed confusion over similar colors.
- Sold-out handling. Kasper Dolk suggests disabling add-to-cart and marking sold-out options inside the selector.
- Sizing help. Ruben suggests moving the size quiz higher, since 30% of buyers almost didn't buy over size worries but only 20% of visitors scrolled far enough to see the quiz. Kasper Dolk suggests a UK/EU/US size switch and fit scales above add-to-cart.
- Content next to the selector. Dave Diederen suggests three USPs above the selector, an almost-out-of-stock note below it, and a free-gift highlight inside it.
Josh Thompson describes sizing chart and survey-driven fixes with claimed lifts, but no test data was shown, so treat those as ideas.
Placement and design patterns that won
- Show options as visible blocks rather than a dropdown when the choice matters.
- Group sizes the way shoppers already think, such as men's and women's.
- Give a fit reference, such as the model's height and size.
- If the product has variants, make the sticky add-to-cart send people back to the selector instead of throwing an error.
To test this yourself, change only the selector, track conversion rate and revenue per visitor, and watch returns for apparel.
A lift near +4% needs a lot of traffic, so check the sample size calculator first.
Related tests elsewhere on the store
See all variant selector tests across the store, including cart tests, plus product page sticky add-to-cart tests, product image tests and bundle tests.