Ecommerce Website Audit: The Complete 50-Point Checklist

An ecommerce website audit is only as useful as its order. Fifty observations in a random sequence produce a document nobody acts on. Fifty checks that follow the customer from landing to confirmation produce something your team can work through in an afternoon and argue about with evidence.

This is the checklist I work from, split the way the customer actually moves.

How to use this checklist

Work in order. The sequence matters because problems compound downstream: a category page that hides your best products makes every product page metric look worse than it is, and you will misdiagnose the product page if you audit it first.

Mark each check red, amber or green as you go. Do not fix anything yet. The temptation to stop and repair the first broken thing is what turns a two-hour ecommerce website audit into a three-week project that never finishes the list.

Homepage (5 checks)

• What, who, why. A stranger can state what you sell, who it is for and why you rather than a competitor, within five seconds.
• A primary path exists. There is one obvious next step, not six competing banners of equal weight.
• Delivery and returns are visible without scrolling, or at least promised above the fold.
• Entry points match your catalogue. If 60% of revenue is one category, that category is not buried three clicks down.
• It loads before people leave. Largest Contentful Paint under 2.5 seconds on mobile, measured on a throttled connection rather than office wifi.

Navigation and search (5 checks)

• Menu labels use customer words, not internal category names or brand-invented terminology.
• Every product is reachable within three clicks from the homepage.
• Internal search tolerates mistakes. Partial words, plurals and common misspellings return results.
• Zero-result searches are handled. Suggestions or alternatives appear rather than a dead end.
• Search terms are logged and reviewed. What people type is the cheapest product research you will ever get, and most stores never read it.

Category and listing pages (7 checks)

Which products appear first here is a commercial decision, not a design one. The deeper treatment is in ecommerce merchandising.

• Filters match how buyers decide, which is usually size, price, compatibility or use case rather than internal attributes.
• Filters survive a back button. Selections persist when someone views a product and returns.
• Default sort is deliberate and defensible, not whatever the platform shipped with.
• Listing tiles carry enough to choose: price, a key variant signal, availability, and a review count if you have one.
• Out-of-stock items are handled deliberately, either demoted or clearly marked, never silently mixed in.
• Pagination or lazy loading works without losing position or breaking the back button.
• Category pages have real copy that helps someone choose, placed where it does not push the products below the fold.

Product pages (10 checks)

The densest section, because this is where the decision is made. If your suspicion already points here, the 30-point product page audit goes considerably deeper than the ten below.

• Above the fold answers the basics: what it is, what it costs, whether it is available, how to buy it.
• Images cover real use, including scale, detail and the product in context, not only studio shots on white.
• Variant selection is unambiguous. Unavailable combinations are visible before selection, not after.
• Delivery timing appears on the page, not only at checkout. This single item moves more revenue than most redesigns.
• Returns terms are stated plainly, in words rather than a link to a legal page.
• Specifications are complete enough to compare against a competitor without leaving.
• Reviews are present and credible, including the mediocre ones. Uniformly perfect ratings reduce trust.
• The add-to-cart control stays reachable on long pages, particularly on mobile.
• Cross-sells are relevant, meaning genuinely compatible or complementary rather than algorithmically adjacent.
• Stock and lead time are honest. Nothing costs future revenue like a delivery promise the warehouse cannot keep.

Cart (6 checks)

• The full cost is visible in the cart, including shipping and tax, before checkout begins.
• Quantities and variants are editable without removing and re-adding the item.
• The cart persists across sessions and devices for logged-in customers.
• Free-shipping thresholds show progress rather than only appearing once reached.
• The coupon field is not a distraction. A prominent empty box sends people off to hunt for codes they never find.
• Continuing to shop is possible without losing the cart or the place in the catalogue.

Checkout (10 checks)

The most expensive place in the store to have a problem, because everyone here has already decided to buy. The 25-point checkout audit covers this surface properly, including a ten-minute self-test.

• Guest checkout exists and is not visually subordinate to account creation.
• Field count is minimal. Every field you cannot justify commercially is costing orders.
• Shipping cost appeared earlier. If checkout is the first time it is seen, treat this as the priority finding.
• Errors are specific and inline, telling people what to fix rather than that something is wrong.
• Entered data survives an error. Nothing is cleared when validation fails.
• Payment methods match the market, including the local options people in that country expect.
• Address entry is assisted through autocomplete or postcode lookup.
• Progress is visible so people know how much remains.
• Trust cues sit near the payment step: contact details, returns, security indicators.
• The confirmation page and email are useful, setting delivery expectations rather than only thanking the customer.

Mobile (7 checks)

Run these on a real handset on mobile data. Mobile carries most ecommerce traffic and converts below desktop almost everywhere, which makes it the highest-value section in any ecommerce website audit.

• Tap targets are large enough and not crowded together.
• The keyboard does not hide the button you need to press next.
• Input types are correct, so numeric fields raise a numeric keypad.
• Wallet payments are offered where available, since they remove most of the form.
• Images do not push the price below the fold on a small screen.
• Interstitials do not block the path: cookie banners, newsletter popups and app prompts stacking together.
• The whole purchase completes on mobile data, end to end, without switching to a laptop.

How to score severity

A flat list of fifty items is not a plan. Severity is what turns it into one, and it has two dimensions rather than one.

Then apply the second dimension: implementation effort. A red item that takes a developer a day outranks a red item that requires replatforming. This is the whole reason a professional audit prices each finding, because impact without effort is only half of a decision.

Where this checklist stops

Fifty checks will find your visible problems. They will not tell you what each one costs, because that needs your conversion data split by step, device and market, applied against order value and margin.

They also stop at the storefront. Products sold below contribution after fulfilment, channels whose real cost is masked by last-click attribution, a repeat purchase rate that forces you to buy every customer twice: none of that appears in a page-by-page review, and any of it can outweigh everything on this list.

If you want the shorter version first, the ten-check DIY audit takes thirty minutes. The pillar guide covers what a full audit examines and delivers, and there is a straight answer on what one costs.