The fastest way to increase ecommerce conversion rate is to find the step where most shoppers drop off, then test small changes there: clearer product titles, visible reviews, honest shipping costs, and reassurance near the buy button. In the 285 real A/B tests in our database, those kinds of changes won far more often than redesigns built on opinion. But some popular tactics lost money, so test before you roll anything out.
Below are 25 changes that won in real tests, grouped by where they happened, plus the ones that sounded good but lost. Every number comes from a practitioner who ran the test.
Start with the leak: find where your funnel drops
Before you change anything, look at your funnel in order: sessions, product page views, add to carts, checkouts started, and orders. The step with the biggest drop compared to your own history is where a test is most likely to pay off.
Most stores lose the most shoppers between cart and order. Baymard Institute puts the average documented cart abandonment rate at 70.22%, based on 50 studies. In the same research, 40% of shoppers who abandoned checkout (excluding people who were only browsing) said extra costs like shipping, taxes and fees were the reason, 20% said delivery was too slow, and 18% said the site wanted them to create an account.
That gives you a useful filter. If your product page add-to-cart rate is weak, start with the product page. If carts are full but orders are not, start with costs, delivery promises and trust in the cart and checkout.
- Low add-to-cart rate: work on the title, images, reviews and the buy box.
- High cart abandonment: work on shipping messaging, guarantees and payment reassurance.
- Low order value: work on bundles, thresholds and offers.
Track revenue per visitor (RPV) next to conversion rate. A change that adds orders but shrinks order size can look like a win on conversion rate alone. You can browse every result by page type in our test database.
Product page changes that won
The product page is where most of our tests ran, and the wins cluster around three things: saying the benefit clearly, putting proof near the decision, and removing doubt from the buy box.
Lead with the benefit
- Benefit-led product title. Dave Diederen saw +15.61% conversion rate after changing the product title to an "Achieve [outcome] with [product]" format.
- Outcome headline in the hero. Kasper Dolk reported $823,115.54 extra revenue per year from changing one product hero headline to the shopper's desired outcome.
- Two benefit words in the title. Adding words like "Sweat-Proof" to a product title produced +$267,576 profit in a test by Dave Diederen.
Move proof closer to the buy button
- Review stars above the title. Moving the review stars above the product title was worth +$855,804/year in one of Dave Diederen's tests.
- Reviews section higher on the page. Moving the existing review section directly below the above-the-fold area gave +11.76% conversion rate.
- Customer video reviews. Adding customer video reviews to the product page lifted conversion rate by +5.80%.
Clean up the buy box
- Sticky add-to-cart bar. A sticky bar that scrolls back to the main buy box gave +11.67% conversion rate.
- Savings in the button. Changing the button to "Add to Cart - Save 50%" gave +14.41% RPV.
- Removing a contradiction. On a page that promised free shipping, removing "Shipping calculated at checkout" below the button gave +14.49% conversion rate.
- Thumbnails instead of dots on mobile. Replacing gallery dots with a scrollable thumbnail strip gave +15.10% RPV.
Note that some of these remove things. Dave Diederen also saw +9.77% conversion rate after hiding payment icons below the add-to-cart button. More elements near the buy box is not always better.
@DaveDiederen@DaveDiederen@DaveDiederen@DaveDiederen@DaveDiederen@DaveDiederenCart and checkout changes that won
Cart and checkout wins line up closely with the abandonment reasons Baymard lists: surprise costs, doubts about delivery, and doubts about the store. Most of the winning changes were small blocks of reassurance placed right where the shopper commits.
- Reviews on the checkout page. Dave Diederen saw +17.96% conversion rate by showing customer reviews on the checkout page.
- Payment icons under the cart button. Adding payment method icons directly below the checkout button in the cart gave +7.37% conversion rate.
- A clear free shipping message. Changing cart copy from "Free shipping will be applied at checkout" to a "Congrats! You unlocked" message gave +11.08% conversion rate.
- Hiding a pre-ticked add-on. Hiding the pre-ticked shipping protection checkbox in the cart gave +6.39% conversion rate, which fits Baymard's finding about unexpected extra costs.
- Guarantee under the checkout button. Adding a "30-day money-back guarantee" under the cart checkout button was worth +£68,640/month.
- Three USPs below Pay Now. Adding guarantee, shipping and delivery-speed points below the Pay Now button gave +9.90% conversion rate.
- Delivery and returns info in the cart drawer. Oliver Kenyon saw +23.8% revenue per visitor after adding a dispatch, returns and payment block under the checkout button in the cart drawer.
- A checkout countdown, in some markets. Dave Diederen saw +3.16% RPV (+$139,029/month est.) from a countdown timer at checkout. Read the losing section below before you copy this one.
For more on what to put near the checkout button, see our collection of trust badge tests.
@oliverkenyon@DaveDiederen@DaveDiederen@DaveDiederen@DaveDiederen@DaveDiederenOffer and pricing changes that won
Offer tests often move revenue more than design tests, because they change what the shopper gets, not just how it looks. They also carry more risk to margin, so measure profit per visitor where you can.
- A flat shipping fee instead of a threshold. Dave Diederen saw +7.90% RPV after removing the free shipping threshold and charging a flat fee on every order.
- A higher free shipping threshold. Raising the threshold to $165 gave +6.27% profit per visitor in another of Dave Diederen's tests. Our free shipping threshold tests show the right number varies by store.
- A clear guarantee under the button. Oliver Kenyon saw +€63,000/month after adding 60-day guarantee and easy returns blocks under the add-to-cart button.
- A named sale. Naming the sale in the discount badge ("Summer Sale - Save 50%") gave +11.12% conversion rate.
- A one-click extra discount. Ruben at Tailored Commerce saw +28% RPV from a one-click 10% extra discount banner above the add-to-cart button.
- A simpler price. Oliver Kenyon found that lowering the price and removing the free gift ($82 vs $85 plus gift) was worth +$43k/month.
- Subscribe and save on repeat products. A subscribe-and-save checkbox (20% off) on top repurchase products brought in $346K MRR in 30 days for Dave Diederen.
Two of those wins involve charging shoppers more for shipping, which goes against most advice. That is the point of testing: the tactic everyone recommends is not always the one that makes money.
Changes that sound good but lost
Losing tests are just as useful as wins, because they stop you from rolling out a change that costs money. Our analysis of losing A/B tests covers more of them.
- Skipping the cart. Sending shoppers straight to checkout sounds like less friction, but Dave Diederen saw -10.23% conversion rate.
- A "better" bundle offer. Changing a 2-pack to "Buy 2 Get 1 Free" at the same price lost -52.38% RPV (-$104,679/month) in a test by Milos.
- Countdown timers in the wrong place. One checkout countdown lost -7.1% RPV, and one for a European market lost -35.61% RPV. Kasper Dolk saw conversion rate rise +25.61% after removing a checkout countdown. See all countdown timer tests before you add one.
- Removing the email popup. Popups feel annoying, but removing one lost -20.51% RPV.
- The price inside the button. Adding the price to the add-to-cart button lost -1.98% conversion rate.
- Swapping written reviews for videos. Replacing written reviews with only product videos lost -1.60% RPV, even though adding videos alongside reviews won.
@kasper_dolk@DaveDiederen@DaveDiederen@DaveDiederen@DaveDiederen@DaveDiederenThe pattern: the same element can win on one store and lose on another. A countdown timer tested across several brands gave about +10% conversion rate on a low-priced brand, was flat on a high-priced brand, and lost -9% on another.
How to test before you roll out
Every change above was tested against a control. Here is a simple process to do the same.
- Write a hypothesis. Tie it to the leak you found: "Showing delivery dates in the cart will cut cart abandonment because shoppers worry about slow delivery."
- Check you have enough traffic. Use our A/B test sample size calculator to see how many visitors each variant needs before you start.
- Pick RPV as the main metric. Watch conversion rate and order value as secondary metrics.
- Run full weeks and don't stop early. Dave Diederen saw an authority badge show +17% after week 1, which turned out to be noise. After four weeks it was +1.66% and flat.
- Roll out, then retest. A win on one product page is a hypothesis for the others, not a guarantee.
You need a testing tool to split traffic. As of November 2026, Convert's Growth plan is listed at $399/month, or $299/month billed annually, for 100,000 tested users per month, with a 15-day free trial. It works on Shopify and any other site. It is not a fit for small stores that can't reach a reliable sample size, since you pay a fixed fee either way.
VWO combines A/B testing with heatmaps and session recordings. Its pricing page lists Growth, Pro and Enterprise tiers, and as of November 2026 it does not show prices for them publicly, so expect to request a quote. It suits teams that want research and testing in one place; it is a poor fit if you only need a simple split test and want to see the price upfront.
If you want ideas with the biggest upside, start with the biggest A/B test wins in our database, then pick the ones that match your leak.