How to Run a CRO Audit on Your Shopify Store
You run a CRO audit on your Shopify store by pulling four funnel rates from Shopify analytics, circling the lowest one, and walking that page on your phone to list every friction point: the audit’s only job is to find the step where you lose the most buyers. It takes an afternoon, free tools, and seven steps.
What You Need
- Shopify admin with access to Analytics and the theme editor
- GA4 connected to the store, for the device split
- Google Search Console, for the Core Web Vitals report
- Microsoft Clarity installed, for scroll maps and session recordings
- One blank spreadsheet with five rows and five columns, which becomes the audit
Step 1: Pull Your Funnel Counts From Shopify
Pulling your funnel counts from Shopify comes first because every later step is a calculation on these four numbers, and Shopify already holds them in the conversion summary under Analytics. Set the date range to the last 30 days, or 90 days if the store gets under a few thousand sessions a month.
The summary shows sessions, sessions that added to cart, sessions that reached checkout, and sessions that converted. Write the four raw counts in your sheet before you touch a calculator, and add a fifth number: sessions that viewed at least one product page, which you get from the sessions by landing page report or the product analytics view.
Just know that a 7-day window is not enough here. One weekend sale or one bad ad day moves the rates enough to send you fixing the wrong page.
While you are in Analytics, also note the sessions by landing page for your top three collection pages. A collection with high sessions and almost no product clicks is a fifth leak the conversion summary does not show, and it belongs in the homepage and collection rows of your sheet.
Step 2: Turn the Counts Into Four Step Rates
Turning the counts into four step rates comes next, and a raw count tells you nothing on its own, while the rate between two steps tells you exactly which page is failing. Divide each step by the step before it and write the result in the sheet.
| Rate | How to calculate it | Where I see stores land |
|---|---|---|
| Product view to add to cart | Add-to-cart sessions divided by product view sessions | 5% to 10% |
| Add to cart to checkout | Reached-checkout sessions divided by add-to-cart sessions | 40% to 60% |
| Checkout to purchase | Converted sessions divided by reached-checkout sessions | 40% to 60% |
| Overall | Orders divided by total sessions | 1.4% to 2.7% |
The overall number is the one on your Shopify dashboard, and Shopify’s own conversion rate roundup puts the global average between 1.4% and 2.66% in 2026, with beauty at 5.32% and home and furniture at 1.29%. So compare your store against your category, not the blended number.
Then circle the weakest of the first three rates against the ranges in the table. The page before the weakest rate is where you spend the rest of the afternoon. For most Shopify stores I audit, that is either product view to add to cart or add to cart to checkout.
Step 3: Split Every Rate by Device in GA4
Splitting every rate by device in GA4 comes before you look at any page. A store that converts at 3% on desktop and 0.8% on mobile has a mobile problem, not a general one, and the blended rate hides it. Most Shopify stores get over 70% of sessions from phones, so the mobile number is the one that matters.
Build the same four steps in a GA4 funnel exploration, with view_item, add_to_cart, begin_checkout, and purchase as the steps, then add device category as the breakdown. Google’s guide to funnel exploration reports walks through it, and the closed funnel option keeps the counts honest by only counting users who entered at step one.
Once the device split is in the sheet, you often find the weakest step moves. A cart step that looked fine overall turns out to be leaking only on mobile, where the checkout button sits under three upsell rows. That same mobile view is where the one-page or multi-step checkout question gets decided, since a long form on a phone leaks in a way the desktop numbers hide.
If GA4 and Shopify disagree on the counts, and they will, use Shopify for the rates and GA4 only for the split. The two tools count sessions differently, so never mix a Shopify numerator with a GA4 denominator.
Step 4: Walk the Weak Page on Your Phone
Walking the weak page on your phone comes now, since the numbers have told you where shoppers leave, and only the page itself can tell you why. Open it on cellular data, logged out, and scroll slowly, writing down every moment you hesitate.
On a product page, the hesitations are usually a hero image that pushes the price below the fold, a size chart that opens a new tab, or a shipping cost that only appears at checkout. Count how many swipes it takes to reach the add to cart button, because on most stores I audit the answer is two, and the fix is a shorter hero image. My list of product page CRO tests that lifted revenue doubles as a checklist for this walk, so keep it open in another tab.
On the cart step, the reasons visitors add to cart but never check out are almost always surprise costs or a hidden checkout button. Baymard’s cart abandonment research puts the average abandonment rate at 70.22% across 50 studies, with extra costs as the top reason at 40%, so a “shipping calculated at checkout” line in the cart is the first thing to write down. Whether to go free from there is its own math, and I walk through it in does free shipping always win.
Write one top complaint per row in your sheet even if you found five, and park the rest in a notes tab. The audit only works when it forces a choice.
Step 5: Check Speed and Watch Five Recordings
Checking speed and watching five recordings comes after the walk. A page can look fine in your hand and still fail for a shopper on a slower phone, and recordings catch the bugs you cannot reproduce yourself. Both checks take under 20 minutes.
For speed, open Search Console’s Core Web Vitals report and look for your weak page in the poor or needs improvement group on mobile. Pages that fail there almost always show up as the worst funnel step too, because a layout shift right as the shopper reaches for the button reads as a broken page.
From there, open Clarity and filter session recordings for that URL with rage clicks or dead clicks. Watch five. A shopper tapping a variant swatch four times and getting nothing is a bug that costs you more than any copy change, and the behavior signals that show where customers drop off all start with this kind of recording.
While Clarity is open, pull the scroll map for the same page too. If the average fold line sits above your add to cart button, you have found a fix before watching a single recording, and it goes to the top of the sheet.
Step 6: Score Every Fix on Impact and Effort
Scoring every fix on impact and effort comes before you change anything: a Shopify store owner can fix six things in an afternoon and the wrong six move nothing. Score each row for impact, meaning how many sessions hit the page and how big the drop is, and effort, meaning theme settings today or a developer next month.
| Fix | Where it lives | Effort |
|---|---|---|
| Show shipping cost and delivery timing on the product page | Theme editor, product section text block | Low |
| Move price, variant picker, and add to cart above the fold on mobile | Theme editor, section order | Low |
| Turn on Shop Pay and express checkout | Settings, Payments | Low |
| Add a sticky add to cart bar on long product pages | Theme settings, most current themes | Low |
| Replace vague trust badges with a one-line return policy near the button | Theme editor, text block | Low |
| Cut checkout form fields to the eight most stores need | Settings, Checkout | Medium |
| Rebuild the cart drawer so the checkout button sits under the line items | Theme editor or developer | Medium to high |
The low-effort rows land in the high impact, low effort box on almost every audit, and most match the CRO tests you can run without tools using only the theme editor. Each row that becomes a test also gets a planned length, since how long an A/B test should run is set before it starts, not when the graph looks good. The checkout row is worth more than it looks, since Baymard’s checkout form research found the average checkout carries 11.3 form fields when most stores need eight.
Also turn on Shop Pay if it is off, because it saves the email, payment method, and shipping details for returning shoppers so they skip the address form entirely. That one setting fixes more checkout-to-purchase drop than any copy change.
The checkout rows deserve their own line in the sheet even when checkout to purchase is not your weakest rate. Baymard’s checkout UX benchmark scores 63% of mobile checkouts as mediocre or worse and puts the average gain from checkout design changes alone at a 35% lift in conversion rate, and Shopify lets you edit most of that flow from Settings without touching code.
Step 7: Ship Two or Three Fixes and Retest in Four Weeks
Shipping two or three fixes and retesting in four weeks closes the audit because that is long enough for the new numbers to mean something and short enough that you still remember what changed. Ship the top rows from your impact and effort scoring, note the ship date in the sheet, and leave the store alone.
Then rerun the same conversion summary in a fresh column and compare the same four rates. If the weak step improved and the drop moved somewhere else, your next audit starts there, which is how a store works its way down the funnel one leak at a time.
Before you compare, check that nothing else changed in the window. A new ad campaign, a sale, or a theme update in the same four weeks muddies the read, so log those in the notes tab with their dates and compare the weeks around them separately if you have to.
Whatever you do, do not ship all seven fixes at once. You will get a lift and have no idea which change caused it, and the next round of decisions is guesswork again.
On a store under 5,000 sessions a month, stretch the retest to six or eight weeks instead. A rate that moves from 5.1% to 5.6% on 300 add to carts is noise, not a win, and calling it early is how owners end up reverting good changes and keeping bad ones.
What to Do If the Numbers Don’t Move
The weak rate stayed the same after the fix
The weak rate stays the same after a fix in about a third of audits, and the complaint you wrote down was real but not the one that stopped the most shoppers. Go back to the recordings, filter for the same URL, and watch five more with fresh eyes, this time looking for what shoppers do right before they exit rather than what annoyed you.
Every rate looks fine and the store still converts under 1%
Every rate looking fine on a store converting under 1% usually means the traffic is wrong, not the pages, because a funnel full of cold ad clicks with no intent shows the same shape as a healthy one, just smaller at every step. Check the landing page report for where the sessions come from before you touch the theme again, and cut the source sending bounces. On the stores where I have seen this, one broad-interest Meta campaign was sending 40% of sessions and under 2% of orders, and pausing it lifted the conversion rate overnight without changing a single page.
The One-Afternoon Audit Checklist
- Build a sheet with five rows, homepage to checkout, and columns for visits, rate, drop-off, complaint, and fix
- Pull the last 30 days from Shopify’s conversion summary, or 90 days if sessions are low
- Calculate the four step rates and circle the weakest one
- Split every rate by device in a GA4 funnel exploration before deciding anything
- Open the page before the weak rate on your phone, on cellular, logged out, and log one top complaint
- Check the page in Search Console’s Core Web Vitals report and watch five rage-click recordings in Clarity
- Score each fix on impact and effort, then ship the top two or three low-effort rows
- Wait four weeks, rerun the same summary, and compare the same four rates
Want Me to Pull Your Funnel Before the Call?
A CRO audit on a Shopify store comes down to four rates, one sheet, and one page walked on a phone. Pull the funnel, find the weakest step, list the friction on that page, and fix the cheapest high-impact item first. An audit like this one kicked off the work behind Ergomotion’s 102% growth in sales from Google.
If you would rather have someone run it with you and turn the sheet into a testing roadmap, sized to whether you should even A/B test with low traffic yet, my Shopify CRO services start with exactly this process. Book a call and I will pull your four funnel rates before we talk, so the first conversation starts at the leak.
Sources:
- Shopify: Ecommerce Conversion Rate Benchmarks
- Google Analytics Help: Funnel Exploration
- Baymard Institute: Cart Abandonment Rate Statistics
- Google Search Console Help: Core Web Vitals Report
- Microsoft Clarity: Session Recordings
- Baymard Institute: Checkout Flow Form Fields
- Baymard Institute: Current State of Checkout UX
