A back-in-stock notification is an opt-in alert, usually email or SMS, that tells a shopper when a sold-out product or size is available again.
It saves some sales, but on its own it is a weak fix: in Baymard's usability testing, 30% of shoppers who hit an out-of-stock product were likely to leave and buy elsewhere, and many ignored the "email me" button.
In the A/B tests in our database, the bigger wins came from giving shoppers something to do right now, such as similar products, with the alert kept as a backup.
Wishlists show the same pattern. Two tests in the database found small positive effects on conversion or transactions, but neither showed a clear revenue gain.
Below is what each tool is good for, what the tests show, and how to test them on your own store.
Wishlists in ecommerce: what they are for
A wishlist, often a heart icon on product cards and product pages, lets shoppers save items they are not ready to buy.
It serves people who are comparing options, waiting for payday, or building a gift list.
The catch is the login wall. Baymard's product page benchmark found that 75% of sites require users to create an account to use "save" features, and its checkout abandonment survey found 18% of US shoppers who abandoned a checkout did so because the site wanted them to create an account.
A wishlist that asks for a password first will mostly go unused.
Wishlists also do quiet work behind the scenes.
Saved items tell you which products people want but are not buying yet, which helps with restock decisions, price tests and email targeting.
That value often matters more than any direct conversion lift.
- Guest wishlists: save to the browser first, and ask for an email only when the shopper wants the list on another device.
- Visible entry point: put a wishlist link in the header so saved items are easy to find again.
- Follow-up: a saved item is a signal of intent. Price-drop and low-stock emails tied to the wishlist give it a job after the visit ends.
Wishlist conversion: what the A/B tests show
The two cleanest wishlist tests in our database point the same way: a modest lift in conversion or transactions, with no proof of a revenue gain.
| Test | Who ran it | Change | Result | Caveats |
|---|---|---|---|---|
| Adding a wishlist | Clean Commit | Heart icon on product cards and product pages, login required, plus a header wishlist link | +4.2% conversion rate, +4.3% add-to-cart, revenue per visitor flat | 113,055 visitors over 48 days, 83.9% probability; not significant; AOV -3.9% |
| Wishlist on listing and product pages | Clarins UK | Save-for-later option on listing and product pages | +1.84% transactions; basket page views up | Sample size and significance not published |
| Order total and lock icon on checkout button | Convertibles | A/B/n test of cart drawer changes, one of which was a save-for-later option | The winner was the total-plus-lock button (+$389,565/month); save for later did not beat it | Five variations; the save-for-later result is reported only as "did not win" |
The Clean Commit result is the most useful because it reports everything.
Conversion rose 4.2%, but average order value fell 3.9%, so revenue per visitor stayed flat.
With an 83.9% probability to beat control, it was counted as inconclusive.
A plausible reading is that the wishlist helped some hesitant shoppers buy one item, but did not grow baskets.
The Clarins UK test shows a smaller +1.84% lift in transactions.
It came from a VWO case study without sample sizes, so treat it as a direction, not a forecast.
The honest takeaway: a wishlist is unlikely to be your biggest win.
It is worth testing on stores with long consideration cycles, gift buyers or large catalogs, and less so on single-product stores where shoppers either buy now or leave.
Judge it on revenue per visitor, not on wishlist clicks.
Back-in-stock notifications: how they work
The usual setup is a "Notify me" button or form that replaces Add to Cart on a sold-out product or size.
The shopper leaves an email or phone number, and an automated message goes out when inventory returns.
In Klaviyo, for example, a sign-up records a "Subscribed to Back in Stock" event that triggers a flow, and subscribers wait in a back-in-stock delay until the item is restocked.
Their help center notes that alerts can go out in batches, and that after the first message, Klaviyo cannot check whether the item has sold out again before a follow-up.
Batch sends matter when you restock 50 units for 2,000 subscribers.
The weakness is timing. Baymard's testing found shoppers often read the button as a signal to shop elsewhere.
In the words of the study: "Why wait to receive an email from one site, alerting them that a product is in stock, when they can likely purchase the product immediately at a competing site?"
Notifications work best for products people cannot easily find elsewhere: your own brand, a limited drop, a specific size in a popular style.
What beats a lone "Notify me" button
Three tests in our database added an alternative to sold-out pages that had only a notify-me option. All three reported higher conversion.
DRIP Agency
Visually
Blend Commerce
Clean Commit- Oceansapart, via DRIP Agency: a "Similar Products" section under the Notify Me button on sold-out variants. Conversion +11.6%, revenue per user +21.7%. Significance and sample size were not stated.
- The Oodie, via Visually: a similar-products block on out-of-stock product pages. Conversion +9.5%. This is vendor marketing without sample size.
- A plant brand, via Blend Commerce: on that store, the back-in-stock and similar-products modules sat far down the page. A message in the image gallery linking down to them lifted conversion 4.59%, revenue per visitor 6.97%, and back-in-stock clicks 8.48%.
The plant brand test is worth noting: making the options easier to see raised both sales and notification sign-ups.
An alert and recommendations are not competing for the same shopper. They serve different people.
Not every notify-me placement wins, though. In a five-way collection page test by Convertibles, the winning version used filter pills and quick-add buttons (+$42,308/month).
A version with Notify Me on sold-out items did not beat it.
On listing pages, sold-out items with a sign-up form may just add clutter.
Golden Planet's test for Lotte Duty Free added both Add to Cart and Restock Notification buttons to mobile bestseller cards and saw a 45.7% higher add-to-cart rate across 17,440 users.
Since both buttons changed at once and revenue was not reported, it does not isolate the restock button's effect.
Baymard's main advice goes further: when a product is only temporarily out of stock, let people buy it now with a longer delivery estimate.
Its benchmark found 68% of sites don't allow orders for temporarily out-of-stock products.
On Shopify, this is the "Continue selling when out of stock" setting, which requires inventory tracking to be on.
Only use it when you can restock reliably and show the real ship date.
Messaging when the item comes back
Restocks invite scarcity copy, and the results here are mixed.
In a Convertibles test on a golf product page, a "Back in stock – only N% left" box with a progress bar lifted revenue per visitor 26.4% (368 vs 297 orders, 98% probability).
The meter ran on a timer, not live inventory, which is a trust and consumer-protection risk we would not copy.
Dave Diederen saw the reverse on another Shopify store: showing "In Stock" instead of "Almost Out of Stock" lifted revenue per visitor 16.78%, conversion 5.72% and AOV 10.46%.
Scarcity effects depend on the product and the shopper, so test your restock copy instead of assuming urgency helps.
Our scarcity and urgency guide and low-stock test collection cover more of these results.
How to test wishlists and back-in-stock alerts
- Size the problem first. Check how many product page sessions land on a sold-out product or size. If it is 1% of traffic, a test will take months to reach a result. Use a sample size calculator on the affected traffic, not the whole site.
- Start with the dead end. On sold-out pages, test similar products or backorders before polishing the notify-me form. That is where our database shows the clearest wins.
- Pick revenue as the main metric. Wishlist clicks and alert sign-ups are easy to inflate. The Clean Commit test shows conversion can rise while AOV falls.
- Count delayed sales. Back-in-stock revenue arrives days or weeks after the visit. Keep the test running, or track orders from alert emails by variant, so the control is not credited with sales the alert produced.
- Remove the login wall. If your wishlist requires an account, test a guest version before deciding wishlists do not work.
For more on sold-out states and product pages, see our product page design guide and the full CRO guide.