# 30 A/B test ideas for ecommerce (with the hypothesis for each)

> Thirty concrete A/B test ideas for an online store, from homepage to checkout, each with an If-then-because hypothesis and the metric to measure. No invented uplifts.

- Source: https://bixlow.com/en/blog/30-ab-test-ideas-for-ecommerce
- Language: en
- Published: 2026-11-02T13:00:00.000Z
- Updated: 2026-11-03T09:15:25.686Z
- Author: fernando
- Topics: A/B testing, ecommerce CRO, experiment design, product page optimization, checkout optimization, mobile CRO, trust signals
- Publisher: Bixlow

---
Here are 30 A/B test ideas for an online store, grouped by area (homepage, category listing, product page, cart, checkout, trust and mobile), each with its hypothesis in "If…, then…, because…" form and the metric we would measure it with. There are no promised uplift percentages, because what worked on one store need not work on yours. What there is, for every test, is a reason, and that is what separates an experiment from a blind change.

## How to use this list

Do not test them all. Pick two or three that answer something you already suspect about your store: a customer email asking about shipping, lots of abandoned carts, a checkout that feels long on mobile.

Three rules before you start:

1. **One hypothesis per test.** If you change the button copy, the colour and the photo at once, you will not know what worked.
2. **Measure orders whenever you can.** Clicks are easy to move. Orders pay the bills.
3. **Work out how long it will take first.** With few orders a month, a small change can need months. The [sample size calculator](/en/tools/sample-size-calculator) tells you in a minute.

## Homepage

**1. A clear sentence instead of the slider.** If we replace the homepage carousel with one sentence saying what we sell, to whom and with what advantage, then more visitors will move into the catalogue, because they know within two seconds whether they are in the right place. *Metric: clicks to a category or product.*

**2. Free shipping in the top bar.** If we show "Free shipping over €49" in a fixed bar at the top, then average order value will rise, because customers know the threshold before filling the cart. *Metric: average order value and orders.*

**3. Category tiles below the header.** If we show the four to six main categories as image tiles right at the start of the page, then more people will enter the catalogue, because they do not need to open the menu to find their way. *Metric: clicks to a category.*

**4. Best sellers before new arrivals.** If the first product block is best sellers rather than new arrivals, then product page visits will rise, because what others already buy removes doubt for first-time visitors. *Metric: clicks to product and orders.*

## Category listing

**5. Free shipping flag on the card.** If we mark the products that clear the free shipping threshold on their own, then average order value will rise, because customers see which products save them the shipping fee. *Metric: average order value.*

**6. Visible filters on mobile.** If we show the two most used filters (size and price, say) as buttons above the listing instead of inside a panel, then more people will reach a product page, because opening a panel on mobile is a step many never take. *Metric: clicks to product from the listing.*

**7. Sort by best sellers.** If the default sort changes from "newest" to "best selling", then product page visits will rise, because the first thing customers see is what people like most. *Metric: clicks to product and orders.*

**8. A second photo on the card.** If each card lets you see a second image (the product in use or a detail), then clicks to product will rise, because a single studio shot does not show what it looks like in real life. *Metric: clicks to product.*

## Product page

**9. Shipping and delivery time next to the price.** If we put shipping cost and delivery time under the price, then orders will rise, because it is the question most people carry into the cart and the one behind most drop-offs there. *Metric: orders.*

**10. Payment methods under the button.** If we show the icons for the payment methods you accept under the add to cart button, then more people will continue to payment, because knowing they can pay the way they want removes a reason to leave. *Metric: add to cart and orders.*

**11. The essentials as bullets at the top.** If we summarise the description in four bullets next to the photo and move the long text down, then add to cart will rise, because on mobile nobody scrolls through paragraphs to find out what it is made of. *Metric: add to cart.*

**12. "Which size should I pick?" next to the selector.** If we place a size guide link right by the selector rather than at the bottom of the page, then orders will rise, because doubt about size stops the purchase at exactly the moment of choosing. *Metric: orders (and, later, size-related returns).*

**13. Rating under the title.** If we show stars and the number of reviews right under the product name, then add to cart will rise, because the social proof is visible without looking for it. *Metric: add to cart.*

**14. Returns in one line.** If we write "Free returns within 30 days" next to the buy button, then orders will rise, because fear of getting it wrong weighs less when you know you can send it back. *Metric: orders.*

## Cart

**15. What's left for free shipping.** If the cart says "You're €12 away from free shipping" with a progress bar, then average order value will rise, because the target is visible and close. *Metric: average order value.*

**16. Low-priced add-ons to get there.** If we suggest two or three inexpensive products that close the gap to the threshold, then average order value will rise, because we give customers an easy way to avoid paying for shipping. *Metric: average order value and orders.*

**17. Shipping shown in the cart.** If we show the shipping cost in the cart instead of waiting for the last step, then checkout abandonment will fall, because the shipping surprise comes early and not after entering all your details. *Metric: orders (not checkout visits).*

**18. Coupon field collapsed.** If we hide the discount code field behind a "Have a code?" link, then orders will rise, because a large empty field reminds people there is a discount they do not have, and they leave to look for it. *Metric: orders.*

## Checkout

**19. Guest checkout first.** If buying without an account is the first and most visible option, then orders will rise, because creating an account is a chore nobody wants for a first purchase. *Metric: orders.*

**20. Fewer fields.** If we hide optional fields (company, second address line) behind a link, then orders will rise, because a shorter form gets finished sooner, especially on mobile. *Metric: orders.*

**21. The favourite payment method on top.** If we put the method your customers use most on mobile at the top of the list, then mobile orders will rise, because they do not have to hunt for it. *Metric: mobile orders.*

**22. A delivery date instead of a range.** If the shipping step says "Arrives Thursday 12th" rather than "2 to 3 working days", then orders will rise, because a specific date is easier to grasp than a range you have to work out. *Metric: orders.*

**23. Order summary expanded on mobile.** If the mobile checkout shows the summary with products and total open rather than collapsed, then orders will rise, because people want to confirm what they are paying for before handing over their card. *Metric: mobile orders.*

## Trust

**24. Reviews with photos.** If we feature the reviews that include a customer photo on the product page, then add to cart will rise, because a real photo answers what the studio shot does not. *Metric: add to cart.*

**25. A person behind it.** If we show a support phone number or WhatsApp in the checkout, then orders will rise, because knowing someone is on the other side reassures people who do not know the brand. *Metric: orders.*

**26. Who we are in one line.** If we add a sentence such as "Family shop in Valencia since 2015, we ship ourselves" to the product page, then orders from new visitors will rise, because a shop with a face and an address inspires more trust than an anonymous one. *Metric: orders from new visitors.*

**27. Imperfect reviews.** If we also show three-star reviews instead of filtering down to five-star ones, then add to cart will rise, because a product with no downsides at all looks less believable. *Metric: add to cart.*

## Mobile

**28. Sticky buy button.** If the add to cart button on the mobile product page stays fixed at the bottom while scrolling, then add to cart will rise, because the button is within reach when the customer decides, not three screens up. *Metric: mobile add to cart.*

**29. Open search bar in the header.** If search shows as a visible bar on mobile instead of a magnifying glass icon, then more people will search and reach a product page, because a bar invites typing and an icon gets overlooked. *Metric: searches and clicks to product.*

**30. The signup pop-up, later.** If we delay the newsletter pop-up on mobile until the second page viewed, then add to cart will rise, because a full-screen pop-up in the first second covers exactly what the visitor came to see. *Metric: add to cart, with signups as a secondary metric.*

## Where to start

If you have to choose three, we would start with 9 (shipping on the product page), 15 (what's left for free shipping) and 19 (guest checkout). All three answer doubts that show up on almost every store, and they are quick to build.

For the rest, rank them with three questions: how many people see that area, how much you trust the reason behind it, and how much work it takes. Whatever has lots of traffic, a solid reason and little work goes first.

## Taking them to your store with Bixlow

Bixlow installs with one script in the head or through Google Tag Manager, on PrestaShop, WooCommerce, Shopify or any other site. You ask for one of these ideas in your own words and the agent writes the hypothesis and builds the variant on your live store, no developers involved. You review it in preview and nothing goes live until you approve it. You can see how it works on [A/B testing](/en/platform/ab-testing) and what it means for AI to do the heavy lifting in [AI A/B testing](/en/glossary/ai-ab-testing).

You set the goal by URL (the order confirmation page), by click, by form submission or with a dataLayer event. And the verdict does not come from eyeballing the numbers: Bayesian statistics will not call a winner until a variant has at least a 95% probability of being the best, with a minimum sample and duration.

## Frequently asked questions

**How many tests can I run at once?**
As many as your traffic can support without them overlapping. Do not run two tests on the same area of the same page at the same time, because one contaminates the other.

**Why are there no uplift percentages in this list?**
Because they depend on your store, your traffic and your starting point. A number taken from another store only sets expectations that do not get met.

**What if a test produces no winner?**
Nothing bad has happened: you have learned that the change matters less than you thought. Keep the simpler version and move on to the next idea.

**Do these work for a small store?**
Yes, but with few orders, pick the ideas that change something big (shipping, payment, sign-up) and leave visual details for when you have more volume.

---

## Meet our AI CRO agent

What you just read, put to work by an agent: it understands your visitors, decides what to test or personalize, builds it on your site without a developer and tells you when a result is real.

https://bixlow.com/en/ai-cro-agent