E-commerce Checklist: 30 Website Checks That Protect Sales

An ecommerce website checklist should help you find the problems most likely to cost sales, not produce a backlog of cosmetic changes. Start with broken purchase journeys and unreliable measurement. Then fix high-friction steps, validate the result and only test changes where the commercial upside justifies the work.

This checklist is ordered by commercial risk. It covers mobile performance, product discovery, product pages, cart and checkout, GA4 ecommerce tracking, and a practical testing process. Use it on a live store before a redesign, campaign scale-up or peak trading period.

Key takeaways

• Fix blockers before optimisations: Payment failures, incorrect prices and broken purchase journeys come before cosmetic CRO ideas.
• Verify measurement before making budget decisions: Test the complete GA4 ecommerce event sequence, parameters, duplicates and revenue reconciliation.
• Prioritise by commercial exposure: Combine reach, severity, evidence and implementation effort instead of treating every observation equally.
• Test uncertainty, not defects: Repair obvious failures directly. Reserve controlled experiments for changes where the better decision is genuinely unclear.

How to use this ecommerce website checklist

Do not score every item equally. A failed payment, incorrect price or missing purchase event deserves immediate attention. A minor visual inconsistency can wait. Work through the checklist with a real product, a real mobile device and a test order.

• Critical: The issue can block purchases, corrupt revenue data or create a legal or trust risk. Fix it before increasing traffic.
• High: The issue affects a major buying decision or a high-volume funnel step. Investigate and prioritise it this sprint.
• Test: The change may help, but evidence should come from user research or a controlled experiment.

Record the affected page, device, traffic share, funnel step and estimated revenue exposure. That turns a generic checklist into a decision tool.

Priority rule: Fix Critical issues before increasing traffic. Schedule High issues by revenue exposure. Put Test ideas into research or experimentation only when the evidence and potential upside justify the work.

Mobile experience and site performance

Test the complete journey on a mid-range phone using mobile data. Lab tools help diagnose technical causes, but field data shows what real visitors experience. Google recommends assessing Core Web Vitals at the 75th percentile for mobile and desktop, with good thresholds of LCP at 2.5 seconds or less, INP at 200 milliseconds or less and CLS at 0.1 or less in its official Web Vitals guidance.

• ☐ Complete a mobile test purchase: Open a landing page, find a product, select a variant, add it to cart and pay. Mark any point that requires zooming, repeated taps or switching orientation as high priority.
• ☐ Check field performance: Review Core Web Vitals in Search Console or CrUX, segmented by page type. Do not treat one Lighthouse run as proof of the experience across users.
• ☐ Find the largest loading cost: Check hero images, product galleries, fonts, video, third-party scripts and tag-manager bloat. Fix the dominant constraint before applying small optimisations.
• ☐ Test interaction delays: Open menus, filters, variant selectors, accordions and add-to-cart controls on a slower device. Visible content is not enough if controls respond late.
• ☐ Check layout stability: Watch for banners, reviews, recommendations and images that move the purchase controls after the page begins loading.

Navigation and product discovery

Navigation should reduce the work required to reach a suitable product. Evaluate it against real shopping tasks, not the neatness of the information architecture.

• ☐ Use customer language in categories: Category names should match how buyers describe the products. Avoid internal merchandising terms that require interpretation.
• ☐ Test site search: Search for top products, common misspellings, attributes and problem-led queries. Review zero-result searches and what users do next.
• ☐ Make filters useful on mobile: Keep the applied state visible, show the result count and preserve selections when users return from a product page.
• ☐ Show decision information in listings: Include the attributes that determine a click, such as price, variant availability, delivery cue or rating, when relevant to the purchase.
• ☐ Handle empty states: Out-of-stock categories and zero-result searches should offer clear alternatives, not a dead end.

Product pages that answer buying questions

A product page must answer the questions that determine purchase, including suitability, total cost, availability, delivery and returns. More copy does not automatically reduce uncertainty. Put the most decision-relevant information close to the product title, price and primary action.

• ☐ Make the offer unambiguous: Show the exact product, price, included items and any subscription or minimum-order condition before add to cart.
• ☐ Explain variants clearly: Use understandable labels, disable unavailable combinations and confirm the selected option near the purchase button.
• ☐ Show delivery information early: State realistic dispatch or delivery expectations before checkout where the platform and fulfilment data allow it.
• ☐ Put returns within reach: Summarise the policy in plain language and link to the full terms. Do not make shoppers leave the buying context to find basic conditions.
• ☐ Use product-specific evidence: Place genuine reviews, useful specifications, sizing, materials or compatibility information where they resolve objections. Avoid decorative trust badges that prove nothing.
• ☐ Check images for decision value: Confirm that images show scale, important details, variants and usage context without slowing the page unnecessarily.
• ☐ Test the primary action: The add-to-cart control should remain obvious after variant selection, validation errors and stock changes. Sticky controls should not cover important content.

Cart and checkout checks

Checkout friction deserves priority because the shopper has already shown purchase intent. Baymard's checkout research separates avoidable usability problems from natural abandonment by people who were only browsing. That distinction prevents teams from treating every abandoned cart as recoverable revenue.

• ☐ Reveal the full payable amount early: Show shipping, taxes and unavoidable fees as soon as they can be calculated. A late cost surprise is a commercial problem, not a copy problem.
• ☐ Allow quantity and removal changes: Make cart edits obvious and update totals immediately without losing selected variants.
• ☐ Offer guest checkout where appropriate: Do not require an account unless the business model genuinely depends on identity before purchase.
• ☐ Remove unnecessary fields: Ask only for information needed to fulfil, pay for and support the order. Explain any field that a shopper may consider unusual.
• ☐ Validate errors in context: Preserve valid entries, identify the exact field and explain how to fix it. Test rejected cards, invalid postcodes and expired sessions.
• ☐ Test every payment method: Verify success, failure and return paths on mobile and desktop. Check that double-clicks or page refreshes do not create duplicate orders.
• ☐ Check the confirmation journey: The order confirmation page and email should agree on products, totals, address and fulfilment expectations.

GA4 ecommerce tracking verification

Analytics is part of the customer journey audit because bad tracking can send budget and CRO decisions in the wrong direction. Google's ecommerce implementation documentation defines recommended events and item data. Its validation guide recommends using DebugView to inspect event-level and item-level parameters in real time.

• ☐ Verify the event sequence: Trigger view_item, add_to_cart, begin_checkout, add_shipping_info, add_payment_info and purchase where those steps exist.
• ☐ Inspect required parameters: Check transaction_id, value, currency and the items array. An event appearing in GA4 does not prove that ecommerce reporting will populate correctly.
• ☐ Check for duplicates: Complete one test action and confirm that one event fires. Pay special attention to duplicate tag implementations, repeated data-layer pushes and purchase events firing on reload.
• ☐ Reconcile orders and revenue: Compare GA4 purchases and revenue with the ecommerce platform for a defined period. Document expected differences such as consent, refunds, time zones and excluded test orders.
• ☐ Test consent behaviour: Repeat the journey after accepting and declining consent. Confirm that the implementation matches the site's policy and regional requirements.
• ☐ Build a usable funnel: Segment product views, add-to-cart, checkout and purchase by device, landing page and channel. Use the funnel to locate investigation areas, not to claim causation.

How to prioritise the findings

Score each issue using reach, severity, confidence and effort. Reach estimates how many relevant sessions encounter it. Severity reflects whether it blocks a purchase, creates doubt or merely adds inconvenience. Confidence should reflect evidence from analytics, recordings, support contacts or usability tests.

Start with defects and measurement failures. Then address high-reach friction supported by evidence. Use session replay to investigate behaviour that analytics cannot explain, while protecting personal and payment data. A recording can reveal a pattern, but it does not quantify the revenue impact on its own.

Practical tests before you approve a change

• Task-based usability test: Ask representative users to find a product, assess its suitability and complete a purchase. Observe where the interface makes them stop or ask questions.
• Browser and device test: Cover the highest-volume combinations from analytics, then include at least one slower mobile device and an unstable connection.
• Operational test: Follow a test order through payment, stock reduction, confirmation, fulfilment, cancellation and refund where practical.
• Controlled experiment: Test a change only when there is meaningful uncertainty, enough traffic and a decision tied to the result. Review whether you are ready to run an A/B test before choosing the tool.

Define the primary outcome and guardrails before launch. Revenue per visitor or contribution per visitor may be more useful than conversion rate when a change affects order value, discounts or returns. Decide in advance what result would justify rollout, another test or stopping.

What this checklist cannot prove

A checklist can find broken journeys, missing information and weak measurement. It cannot prove that a design change caused more incremental revenue. Before-and-after comparisons can be distorted by traffic mix, promotions, seasonality, stock and competitor activity.

ConversionNest's view: the best ecommerce audit ends with a shorter, commercially ranked backlog. If every observation becomes a priority, the checklist has failed to guide a decision.

Your next action

Run one mobile purchase and one GA4 DebugView session today. Fix any blocker, payment failure or tracking error first. Then rank the remaining issues by revenue exposure and evidence, assign owners and review the highest-value change with product, engineering and trading teams.

Frequently asked questions

What should an ecommerce website checklist include?

An ecommerce website checklist should cover mobile performance, navigation, search, product information, trust, cart and checkout, payment handling, GA4 ecommerce tracking and a method for prioritising findings by commercial impact.

How often should you audit an ecommerce website?

Run focused checks after material releases, tracking changes, payment updates and before peak trading periods. Use ongoing monitoring for payment failures, event anomalies, site performance and funnel changes rather than waiting for a calendar-based audit.

Which ecommerce issues should be fixed first?

Fix purchase blockers, incorrect pricing, payment failures, stock errors and unreliable revenue tracking first. Next, prioritise high-volume friction with supporting evidence. Cosmetic changes should not displace issues that affect buying or decision quality.

How do you verify GA4 ecommerce tracking?

Complete a test journey in DebugView, inspect event and item parameters, check for duplicates and reconcile purchases and revenue against the ecommerce platform for a defined period.