CRO Audit Template Checklist
A diagnostic checklist that walks every surface of your store — homepage, PDP, PLP, cart, checkout, thank-you — so you know what to fix and what to test, in that order.
CRO Audit Template
A surface-by-surface diagnostic checklist that finds conversion leaks across an online store before any test is designed.
A CRO audit template is a structured walkthrough of every page type a shopper touches — homepage, category (PLP), product (PDP), cart, checkout, and thank-you — scoring each surface against a fixed set of usability, clarity, trust, and performance checks. The output is a prioritised list of fixes (quick wins) and hypotheses (worth testing) ranked by traffic exposure and revenue at risk.
The point is sequencing. Most teams jump to A/B testing the hero banner; an audit forces you to first answer where the money is actually leaking, so test velocity gets spent on the surfaces with the most upside. It is the document that turns 'we should run more experiments' into a six-month roadmap.
Run an audit before you run a test. The reason is sequencing: A/B tests are expensive in traffic and time, and you only have so many concurrent slots. If 60% of your checkout drop-off is a broken Apple Pay button on mobile Safari, no amount of hero-banner testing will move the needle.
A good audit gives you two outputs side by side. A 'fix it now' list — broken links, untracked events, slow LCP, illegible form errors — that ships this sprint. And a 'test it' list — value-prop wording, PDP layout, shipping-threshold nudges — that becomes your experiment backlog ranked by ICE or PIE.
Don't audit without funnel data open in another tab
A heuristic walkthrough tells you what looks wrong; your analytics tells you what actually costs money. Pull a 90-day funnel report (session → PDP view → add-to-cart → checkout start → purchase) before you start clicking. A 2% drop at PDP→ATC on a page with 200k monthly views outranks a 15% drop on a page with 800 views, every time.
Homepage & PLP: clarity and discoverability
Start with the five-second test on the homepage. Can a first-time visitor name what you sell, who it's for, and one reason to buy from you (not Amazon) within five seconds of the hero rendering? If your value prop is 'Discover the Collection', you have a clarity problem before you have a CRO problem.
On the PLP, audit filters first. For an apparel store with 400 SKUs, missing size and colour filters above the fold can cost 20-30% of category-page conversion. Check filter discoverability on mobile (most stores hide them behind a button most users never tap), sort defaults (is 'featured' actually surfacing your best-converting SKUs?), and image consistency across product cards.
Then check load behaviour. LCP above 2.5s on mobile, layout shift on hero images, and lazy-loaded product grids that block scroll are the three highest-frequency offenders. A 1-second improvement in mobile LCP typically lifts homepage→PLP click-through by 5-10% on stores in the €1M-€15M revenue band.
PDP: where most stores leak revenue
The PDP is the highest-leverage surface in almost every audit. Score yours against eight checks: image quality and zoom, variant selection clarity, price + shipping transparency above the fold, social proof placement, stock urgency (truthful), returns policy visibility, delivery date estimate, and a sticky add-to-cart on mobile. Missing any one of these is typically worth 1-3% of PDP conversion.
Pay specific attention to variant UX. On a beauty SKU with five shades, a dropdown beats a swatch grid for conversion roughly never — yet half of Shopify themes still ship dropdowns by default. On apparel, a size selector without a size-guide link inline (not three clicks away) drives 8-12% of all returns according to most post-purchase surveys.
Cart & checkout: the last 90 seconds
Cart drawer or cart page — pick one and commit. The drawer keeps users in flow but hides shipping math; the page makes shipping costs explicit but introduces a navigation step. Whichever you use, audit for: free-shipping progress bar, gift-with-purchase logic, discount-code field placement (visible fields invite users to leave and Google a code), and a single clear primary CTA.
In checkout, run the audit on a real mobile device, not Chrome DevTools. Time how long Apple Pay and Shop Pay take to surface, check that address autocomplete works for your top three markets, and confirm that error states are inline (not at the top of the page where mobile users never see them). On Shopify, also verify your Shopify Markets configuration isn't quietly showing the wrong currency to ~5% of EU visitors.
Thank-you & post-purchase: the surface everyone skips
The order-confirmation page is the highest-attention surface in your entire store — and the most under-optimised. Audit it for: post-purchase upsell logic, account-creation prompt (post-purchase converts 3-5x better than pre-purchase), review-request setup, referral program entry point, and that your conversion event is actually firing exactly once with the right revenue value into GA4 and your ads platforms.
Cross-check tracking last, but cross-check it. Audit your purchase event in GA4 DebugView on a real transaction, verify deduplication between GTM and the Shopify pixel, and confirm Klaviyo's placed-order webhook is firing within 30 seconds. Tracking holes silently inflate CAC for months — the audit is when you catch them.
CRO audit template — FAQ
For a store with 50-500 SKUs, plan two to three focused days: one day on heuristic walkthrough across all surfaces, one day pulling funnel and session-recording data to validate hypotheses, and a half-day writing up the prioritised backlog. Anything longer usually means you're trying to solve problems inside the audit instead of cataloguing them.
A full audit every six months, plus a focused mini-audit any time you change theme, replatform, or push a major checkout update. Quarterly is overkill for most stores in the €1M-€15M range — you won't have shipped enough tests to invalidate the previous backlog.
At minimum: GA4 for funnel data, a session-recording tool to see what users actually do on PDP and checkout, PageSpeed Insights for Core Web Vitals, and a checkout test order on a real mobile device. A spreadsheet to score each surface and rank fixes is the only deliverable that matters.
Yes. Paid-traffic landers behave differently from organic homepage visits — visitors are colder, intent is narrower, and the value prop has to land in two seconds. Audit your top 3-5 ad landing pages as a distinct surface, scored against ad-creative-to-page message match in particular.
Quick wins ship if they're (a) clearly broken or (b) backed by an established UX principle with no real downside — fixing them doesn't need a test. Anything involving copy, layout, pricing presentation, or social-proof positioning is a hypothesis and belongs in the test backlog, ranked by traffic exposure × expected lift × confidence.
The framework is platform-agnostic — every store has the same six surfaces. Platform-specific checks differ: Shopify audits focus on theme limitations and Shop Pay, WooCommerce on plugin bloat and checkout-page conflicts, Magento on category-page performance and the M2 default checkout. The scoring rubric is identical.
A UX audit scores usability against heuristics regardless of business impact. A CRO audit weights every finding by traffic exposure and revenue at risk, so a confusing nav on a 200-view page is a footnote and a confusing variant selector on a top-10 SKU is the top priority. Same techniques, different ranking function.
The audit itself, yes — it's diagnostic, not implementation. Roughly 60-70% of the quick-wins on a typical Shopify store can be shipped via theme settings, app config, or no-code CRO tools. The remaining items (page-speed work, custom checkout extensions, advanced tracking fixes) need a developer or a plugin-based experimentation layer.
Run two sanity checks first: does your GA4 purchase count match Shopify orders within 5%, and does your GA4 revenue match Shopify revenue within 10%? If either is off by more than that, fix tracking before you trust funnel numbers. Cookie consent, ad blockers, and duplicate purchase events are the three usual suspects.
Yes — that's the whole point. The output of the audit is two artefacts: a fixes ticket-list for the next sprint, and a prioritised hypothesis backlog with at least 8-12 testable ideas ranked by ICE score. Without both, you've written a document instead of changing a number.
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.