The Shopify CRO Checklist: 60 Actionable Items

Most CRO checklists are a list of opinions. This one is built so that every item is either verifiably done or verifiably not done on your store, and so that working top to bottom is a sensible order of operations.

Two rules before you start.

Measure first, change second. Section 1 exists because every item after it is worthless if you cannot tell whether it worked.

Rank by traffic times drop-off. Do not work this list linearly. Find the stage where the most people leave, then jump to that section. A checklist is a menu, not an itinerary.

1. Measurement foundations (1–6)#

1. Confirm you have one agreed source of truth for conversion rate — Shopify Analytics or GA4, not both. They count sessions differently and will never reconcile.
2. Exclude internal traffic: your own IPs, staff devices, and theme previews.
3. Verify the purchase event fires exactly once per order, including for wallet and express checkouts.
4. Record a trailing twelve-month baseline of session-based conversion rate, weekly, so you know your normal range rather than a single number. See what a good conversion rate actually means.
5. Build a funnel view with five stages: sessions, product views, add-to-carts, checkouts started, orders. You cannot prioritise without it.
6. Start a dated change log. Every theme edit, app install, price change, and campaign launch, with the date. This single habit resolves most future arguments.

2. Traffic quality (7–11)#

7. Segment conversion rate by channel. Branded search, non-branded search, paid social, email, direct, referral.
8. Segment by device within each channel — never blended, because device and channel correlate.
9. For each paid campaign, check that the landing page matches the promise in the creative. Mismatched message is the most common cause of a "traffic that never buys" problem.
10. Check bounce or single-page-session rate by source. A source with very short sessions is an ad problem, not a store problem.
11. Review your top ten landing pages by entry volume. These are the pages worth optimising; everything else is a rounding error.

3. Homepage (12–17)#

12. State what you sell and who it is for in the first viewport, in plain words. A brand slogan is not a value proposition.
13. Give the hero exactly one primary action. Two competing buttons is the same as none.
14. Put your top three collections within one tap of the fold — do not make the hamburger the only route to product.
15. Check scroll depth on the homepage. Anything below the point where most visitors stop is decoration; move what matters above it.
16. Move at least one trust signal — reviews, guarantee, shipping promise — above the fold.
17. Verify the homepage carousel is not the only path to your best sellers. Carousels are seen by far fewer people than the merchant assumes.

4. Collection pages (18–23)#

18. Show price, and any variant availability, on the card without requiring a click.
19. Add filters that match how customers actually shop — size, colour, price band, use case — and make them usable on mobile.
20. Ensure sort defaults to something commercially sensible rather than alphabetical or arbitrary.
21. Keep product card images consistent in crop and background so the grid scans quickly.
22. Show enough products above the fold that the page reads as a selection, not a dead end.
23. Add a short intro paragraph for SEO but keep it collapsed or below the grid — it must not push product below the fold.

5. Product pages (24–33)#

24. Put price, variant picker, and add-to-cart in the first viewport on mobile without scrolling.
25. Show a delivery estimate or shipping cost near the price, not only at checkout.
26. Put the returns policy in one sentence near the button, linked to detail.
27. Include at least five images covering scale, detail, in-use, and what is in the box.
28. Add a size or fit guide adjacent to the variant picker if fit is a purchase risk.
29. Make swatch selection change the gallery image. If it does not, shoppers cannot see what they are choosing.
30. Disable or clearly mark out-of-stock variant combinations rather than letting them fail at add-to-cart.
31. Surface reviews near the buy area, not only at the page bottom — most visitors never reach the bottom.
32. Answer your five most common pre-purchase support questions on the page. Read your inbox for the list.
33. Check for dead clicks on product images, swatches, and specification text. Non-interactive elements that look tappable cost you intent.

Full method: the product page optimization playbook.

6. Cart (34–39)#

34. Make checkout the single dominant action. "Continue shopping" should be visibly secondary.
35. Show shipping cost or free-shipping progress in the cart, before checkout.
36. Show express wallet buttons at the top of the cart, not below the fold.
37. Allow quantity edits and removal without a full page reload.
38. Restate the returns policy and any guarantee in the cart. This is where hesitation peaks.
39. Measure cart-to-checkout rate as its own number. It is frequently the weakest step and the least examined — see cart overview.

7. Checkout (40–46)#

40. Enable guest checkout. Forced account creation is a documented, avoidable loss — Baymard Institute's cart abandonment research puts it at 18% of US shoppers citing forced account creation as a reason they abandoned an order.
41. Enable Shop Pay and the wallets your top markets expect.
42. Remove every optional field. Company name, address line two, and phone are usually not required.
43. Make the discount code field understated so you are not sending shoppers off-site to hunt for a coupon.
44. Show a progress indicator so the shopper knows how many steps remain.
45. Check that error messages appear next to the field that failed and explain how to fix it.
46. Complete a real purchase on a phone, on cellular data, in each of your top markets. Then do it again with a wallet.

More: 10 Shopify checkout fixes that actually recover revenue.

8. Mobile (47–51)#

47. Verify every tap target is comfortably thumb-sized, especially close buttons and variant swatches.
48. Check that sticky headers, chat widgets, and cookie banners do not consume the viewport together.
49. Confirm the keyboard does not cover the input it was opened for on checkout forms.
50. Test with one hand on a small phone. If you need two hands to buy, so does your customer.
51. Compare mobile and desktop conversion within branded search specifically. A gap there is a UX problem, not a demand problem.

9. Speed and stability (52–55)#

52. Test a real product page and a real collection page on a throttled mobile connection — not just the homepage.
53. Eliminate layout shift above the fold. Content that jumps causes mis-taps and erodes trust.
54. Compress and correctly size hero and product images; lazy-load below-fold media.
55. Audit your installed apps. Uninstall anything not actively earning its script weight — see storefront speed.

10. Trust, popups, and post-purchase (56–60)#

56. Replace stock photography with real product and lifestyle images. Nothing signals "dropshipper" faster.
57. Publish a real contact route, a physical address if you have one, and plain-language policy pages.
58. Trigger popups on behaviour — exit intent, scroll depth, second page view — never on a timer, and never over the cart or checkout. Watch for rage clicks on close buttons.
59. Set the order confirmation and shipping notification emails to answer the questions customers actually ask next. Post-purchase anxiety drives support load and hurts repeat rate.
60. Ask for the review at the right moment — after the product has arrived and been used, not on dispatch. Reviews are the input to item 31.

How to actually work this list#

Pick your stage. From your funnel view, find the single largest drop-off weighted by the traffic reaching it. Go to that section only.

Ship one change at a time. If you change five things and the number moves, you have learned nothing about which one did it.

Measure against the same period last year, not last week. Week-over-week comparisons are dominated by traffic mix and seasonality.

Test where you can. If you have the volume, run a real experiment rather than a before-and-after. How to A/B test on Shopify covers what volume you need and how to avoid calling a winner too early. If you do not have the volume, change one thing at a time, log it, and compare year over year — that is an honest method too.

Watch ten sessions before you redesign anything. Every merchant who does this finds at least one problem that was not on any checklist, including this one. Session replay and heatmaps exist because the specific way your store breaks is specific to your store.

Sixty items is a lot. You do not need all sixty. You need the three that are costing you the most, done properly, and measured honestly. Start with section 1 so you can tell.