Shopify store audit checklist: read your data, then shop as a customer

A Shopify store audit has 2 halves: read your own data (Shopify’s conversion rate breakdown, GA4, session recordings, PageSpeed Insights, a crawler and Search Console) to find the step, device and page where visitors stop, then shop the store as a new customer on a phone to test the sign up, welcome code, basket, checkout, payment methods, discount codes and every email. When sales drop, find the step in the data first, then reproduce it as a customer on the same device.

A Shopify store owner can run a full store audit without a developer or an agency, before a season or on the day sales fall. Every step uses a free tool, a phone or a test order.

A Shopify store audit checks 2 things: what your data says about where visitors stop, and what a customer receives when they use the store. Crawlers and other website audit tools check pages for search, speed and broken links, and Shopify’s reports count visitors at each step. Neither signs up, opens the welcome email, uses the code or places an order, so the second half of the audit is done as a customer.

Faults and surprise costs both lose orders. In Baymard Institute’s survey of US online shoppers, leaving out people who were only browsing, 40% of those who abandoned an order said extra costs such as shipping, tax and fees were too high, and 12% couldn’t see or work out the total up front.

What the audit covers

Run the 2 halves in order. Part 1 finds the step, device or page where visitors stop. Part 2 shows what happens there, and checks what reports can’t count: emails, codes, payment methods and policy answers. Give each finding a severity.

• Critical: customers can’t order, or pay the wrong amount. Fix it the same day.
• High: some orders are lost or a promise breaks, such as a code, a price or an email. Fix it this week.
• Medium: customers are slowed or confused. Fix it this month.

Part 1: read your own data

Compare the last 7 days with the 7 before and with the same week last year. A fall against both is worth tracing.

Shopify’s conversion funnel

In Shopify, open Analytics, then Reports, then Conversion rate breakdown. Shopify’s help centre defines its 4 steps as shares of all sessions: sessions, sessions with cart additions, sessions that reached checkout and sessions that completed checkout.

• Reached checkout counts input. A session counts only once the shopper types or clicks in checkout. If cart additions hold and this step falls, look at the cart and the checkout button.
• Check the open funnel too. The closed funnel keeps only sessions that took every step in order, so shoppers who skip the cart with an express checkout button can drop out of it.
• Filter to human sessions before you compare rates.
• Allow for the September change. Shopify changed how it measures sessions on 21 to 23 September 2026, and says sessions and conversion rate may differ afterwards with no change in traffic or orders. Across those dates, compare orders as well as rates.

Then split the fall with the Sessions by device and Sessions by landing page reports.

GA4 funnels

Build a funnel exploration with Google’s ecommerce events as the steps, from view_item and add_to_cart to begin_checkout, add_payment_info and purchase, broken down by device category. It shows where users drop and the time between steps.

Use Shopify for orders and GA4 for behaviour. Shopify’s help centre notes that Google counts only visitors with JavaScript and cookies on. If GA4 purchases fall while Shopify orders hold, check the tracking first. Stores not on Shopify Plus had to upgrade their Thank you and Order status pages by 26 August 2026, and scripts from the old additional scripts box have to be set up again on the new pages.

Session recordings

Microsoft Clarity is free with no traffic limit, and Hotjar, now sold through Contentsquare, has a free plan. Filter recordings to the device and step that dropped, and watch at least 10 before you change anything. Clarity flags rage clicks, dead clicks and JavaScript errors. A dead click gets no response, which can mean a broken element. Recordings end when the visitor leaves, so the emails that follow are outside them.

PageSpeed Insights

Run PageSpeed Insights on mobile for your home page, a collection page and your best selling product. Read the field data first, which comes from real Chrome users over the previous 28 days. A pass is good on all 3 Core Web Vitals: Largest Contentful Paint up to 2.5 seconds, Interaction to Next Paint up to 200 milliseconds and Cumulative Layout Shift up to 0.1. The Lighthouse score below it is a lab test that helps find the cause, and Google counts 90 or above as good.

A crawler and Search Console

A crawler finds broken links, redirects, missing titles and pages blocked from search. Screaming Frog crawls up to 500 URLs free, and Ahrefs Webmaster Tools audits sites you verify free. In Search Console, also free, the Page indexing report lists the pages Google didn’t index and why. Google expects only canonical pages to be indexed, so check your product and collection pages and leave filtered and duplicate URLs aside.

Part 2: shop as a customer

Use an email address and phone number your store has never seen, in a private window on your phone, logged out of the admin. Your own account won’t work, because your store treats you as staff or a past buyer. The Black Friday checklist sets this up step by step and adds the sale week tests for texts, reminders and ads. These 12 checks are the year round audit.

Discount codes not working

Try every live code on a phone, including one on a sale item. Each Shopify discount’s settings decide what it combines with, and a customer can enter up to 5 product or order codes and 1 shipping code per order. When checkout says a code “isn’t valid for the items in your cart”, check the products and collections it covers, its minimum purchase, its dates and its usage limit. Discounts built on Shopify Scripts stopped running on 30 June 2026, so test each custom discount again.

How to test Shopify checkout

Enter a real address for each country you sell to. Open each payment method without paying: cards, Shop Pay, PayPal, Apple Pay on an iPhone and Google Pay on Android. Wallet buttons depend on the device, and Shopify orders express buttons to show each customer’s fastest method first, so test on both phones. For a whole order, Shopify offers 3 ways to place a test order: its test payment gateway, Shopify Payments test mode, or a real order that you cancel and refund.

Accounts, policies and AI answers

Shopify’s new customer accounts sign customers in with a 6 digit code sent by email, so test that it arrives. Legacy accounts use a password, so test the reset email. Check your chat and help answers against your policy pages with the chatbot returns test, and AI assistants’ answers with what ChatGPT says about your store.

A test by hand covers 1 inbox, 1 device and 1 moment. A theme publish, an app update or a discount edit can break a step the day after you test it, and Shopify Status notes that issues affecting a small share of stores may not appear there. Obsession’s website audit signs up as a fresh test customer after every release or every morning, fills in every form, opens every email and link, and tries your checkout.

Your company/Watches/Mystery shopperSearch or type a task⌘KAMJO

Mystery shopper

3 businesses · 3 AI test customersObsession AI agent

Day 1Day 2Day 3

PPaywrenPayroll · free trialBot won’t give a price

Ask sales Signed up as AI Welcome Setup Asked price Rep emailed Stopped when a rep wrote

hhollinStore · first orderWelcome code fails at checkout

10% × Retest after 48h × Stopped before payment Rejected again

MMolenna DentalBooking enquiry

Ended Day 1 Sent 10:40 Reply 15:27 First reply after 4h 47m

Each owner’s OK · nothing bought

3 journeys walkedPosted in

Open

When sales or conversion rate drop

A broken step often reads as a slow week. A post in a Shopify Community thread from September 2026 described it as orders slowing and add to carts dipping a little.

1. Confirm the drop in Shopify orders against the same weekdays last week and last year, and check Shopify Status for those dates.
2. Split traffic from conversion. If sessions fell, look at ads, email sends and clicks in Search Console. If sessions held and conversion fell, look in the store.
3. Find the step whose rate fell in the conversion rate breakdown.
4. Find who it affects by device, browser, country and landing page, on human sessions.
5. Match the date to a change: theme publishes, app installs and updates, discount edits, prices, shipping rates, markets and payment settings.
6. Watch recordings of the affected sessions at that step.
7. Reproduce it as a customer on the same device and browser.
8. Fix it, test again and watch the rate for 7 days.

Add to cart but no sales

If cart additions hold and reached checkout falls, test the cart drawer, the cart page and the checkout button on a phone, and any basket or upsell app that changed. If reached checkout holds and completed checkout falls, test delivery rates, payment methods and live codes.

Why a Shopify store gets no sales

If no orders arrive at all, check as a customer that the password page is off, products are on the Online Store sales channel and available in each market, payments are set up and a test order completes. If those pass, look at traffic by source and in Search Console.

Revenue at risk from a fault

Put a figure on each fault with your own numbers, so that the most expensive one is fixed first.

Revenue at risk = sessions a day × share of sessions that reach the broken step × conversion rate × average order value × days broken

An invented store, brindle.example, takes 1,500 sessions a day. A theme published on 7 October breaks its cart drawer’s checkout button in Safari on iPhones, which carry 30% of its sessions. Its conversion rate is 2%, the low end of the 2% to 3% global average on Shopify’s blog, and its average order is £50. Over 6 days, 1,500 × 30% × 2% × £50 × 6 = £2,700. Some shoppers come back on a laptop, so treat the result as the most the fault could cost. The ad spend that brought those sessions is spent either way.

Shopify store audit tools compared

The free tools cover Part 1 for most stores, and the paid options save time or run checks on a schedule. We make Obsession, which does some of this, so every line about another tool is taken from that tool’s own pages, checked on 23 October 2026.

For one store that changes rarely and runs little paid traffic, the free tools and a monthly test by hand cover the audit, and no paid tool is needed.

Limits and next steps

• Reports find the step where visitors stop. A recording or a customer test shows why.
• A customer test covers 1 inbox, 1 device and 1 moment. Test again after every change.
• Test mode leaves your live payment provider untested. A real order that you cancel and refund tests it, and may cost a fee.

Run both halves this week, set a monthly date, and test again on the day of any theme, app, discount or shipping change. Before Black Friday, on Friday 27 November 2026, run the Black Friday checklist ahead of your code freeze.

Store audit questions