Ecommerce CRO Checklist
A walk-through audit of every conversion surface in your store — homepage, PLP, PDP, cart, checkout, thank-you — used to surface obvious wins before you spend a sprint designing tests.
Ecommerce CRO Checklist
A surface-by-surface audit of an online store used to find obvious conversion wins before designing A/B tests.
An ecommerce CRO checklist is a structured walk-through of every page that touches a purchase decision — homepage, category (PLP), product (PDP), cart, checkout, and thank-you — scored against a fixed set of heuristics. The goal is diagnostic, not creative: you're hunting for the broken trust signal, the missing shipping line, the buried size guide, the form field nobody asked for.
Most teams use it as a pre-experimentation gate. Anything the checklist flags as a clear UX defect (not a preference call) gets shipped as a fix. Anything ambiguous — a hero swap, a price-anchoring change, a new badge — becomes a hypothesis for the experiment backlog. That separation is what keeps your test queue full of real questions instead of guesses.
The checklist works because most online stores leak conversions in predictable places. Shipping costs revealed too late, a PDP that hides stock status, a checkout that demands an account, a thank-you page that wastes the highest-intent screen real estate on a generic 'thanks!' graphic. These aren't preferences — they're known friction points with a decade of session-recording evidence behind them.
Run it before you touch your experimentation tool. A team that A/B tests button colours on a PDP with no size guide is optimising the wrong layer. Fix the structural defects first, then test the judgement calls. This is the standard sequencing inside any mature ecommerce CRO programme.
How to score each item
For every check, mark one of three states: PASS (the surface meets the heuristic), FIX (clear defect — ship it, don't test it), or TEST (judgement call — write a hypothesis). Aim to close every FIX within two weeks; the TEST items become your next sprint's experiment backlog.
The checklist, surface by surface
Homepage and PLP. On the homepage, the hero should name what you sell and to whom within three seconds — no clever headlines that require a second read. Primary navigation should expose your top 5-7 categories without a hover gymnastic. Trust signals (reviews, press, return policy) need to sit above the fold on mobile, not buried in the footer. On the PLP, every product card should show price, a swatch or variant cue, an average rating with review count, and stock urgency where it's truthful. Filters should be sticky on scroll; sort should default to something better than 'featured' if you can't explain what 'featured' means.
PDP. The product page does more work than any other surface in your store, so it gets the longest section of the checklist. Image gallery: minimum 5 shots including a scale/context shot and one zoomable detail. Variant selectors should never let the user pick an out-of-stock combination silently. Price should sit next to a shipping line ('Free over €60' or 'Ships Tuesday') because the cart is where shipping reveals murder conversion. Size guides, materials, and care instructions belong in accordion panels — visible without leaving the page. Reviews need a count, a distribution histogram, and the ability to filter by photo or by variant. A sticky add-to-cart on mobile typically lifts add-to-cart rate 5-12% on its own.
Cart and mini-cart. The cart's job is to remove the last excuses, not introduce new ones. Show the line-item price, the shipping cost (or a clear threshold and how far the user is from it), any applicable discount applied automatically, and the total — in that order. Free-shipping progress bars consistently lift AOV when the threshold is within 20% of typical cart value. Avoid surprise fees at this stage; if a cost only appears in checkout, the cart has failed. Cross-sells are fine if they're relevant and low-priced relative to the cart; otherwise they read as distraction.
Checkout and thank-you. Force-account-creation is the single most common defect we see on Shopify and WooCommerce checkouts — guest checkout should be the default path, with account creation offered post-purchase. Address autocomplete, express wallets (Shop Pay, Apple Pay, PayPal) above the form, and a visible order summary on every step are baseline. Form fields should be the legal minimum: no 'company name' on a B2C store, no phone unless your carrier requires it. On thank-you, replace generic confirmations with a referral offer, a relevant cross-sell, or a survey question ('where did you hear about us?') — that page has 100% attention and is criminally under-used.
Frequently asked questions
For a single-brand Shopify or WooCommerce store, plan 4-8 hours of focused work across desktop and mobile. Multi-brand or multi-region stores (separate Shopify Markets, localised checkouts) take 1-2 days because you need to repeat the checkout and PDP sections per locale.
Mobile first, every time. Mobile usually accounts for 65-80% of sessions and a lower conversion rate than desktop, so the gap between the two is where most of your lost revenue lives. Audit mobile, then sanity-check the same checks on desktop.
The checklist surfaces defects you should fix without testing — broken trust signals, missing shipping info, force-account-creation. A/B tests are for judgement calls where you genuinely don't know the answer. Testing a defect wastes traffic; fixing a judgement call without evidence wastes credibility.
No, the heuristic walk-through works without data. But pairing it with GA4 funnel data and session recordings is strictly better — analytics tells you where the leak is, the checklist tells you why. If you import historical GA4 data into a CRO platform, you can rank the checklist findings by the revenue they affect on day one.
Mostly the same, with two additions for subscriptions: the PDP needs an explicit 'how cancellation works' link near the subscribe button, and the checkout needs the billing cadence and next charge date stated, not buried in T&Cs. Subscription stores that hide cancellation terms see 15-25% higher chargeback rates.
On Shopify and Woo stores between €1M-€15M, the two highest-impact items are consistently (1) revealing total shipping cost before the checkout step and (2) enabling guest checkout. Either one alone typically moves checkout completion 5-15%.
Quarterly for a stable store, and after any major change — new theme, new checkout, new payment provider, a big SKU expansion, a re-platform. Drift is real: a checkout that passed in January often fails by July after three app installs added friction nobody audited.
Yes, if you can. The order confirmation, shipping notification, and review-request emails are the natural extension of the thank-you page and carry the same intent. Check that they include a clear next-purchase hook, accurate delivery ETA, and a reply-to that goes somewhere a human reads.
The heuristics apply identically — every store has a PDP, a cart, and a checkout regardless of stack. The difference is that fixing what you find takes a dev sprint on a headless build, whereas on Shopify or Woo a non-technical operator can ship most fixes themselves through theme settings and apps.
Treat it as the diagnostic phase of every CRO cycle. The checklist generates two outputs: a fix list (shipped this sprint, no test) and a hypothesis list (next sprint's A/B tests). Without it, experimentation teams end up testing surface defects instead of real strategic questions.
Get an AI expert review of your site
Paste your URL — Metricuno's AI runs the same heuristic checks a senior CRO consultant would, scoring your page and prioritising the fixes that'll move conversion fastest.