Why is your Shopify conversion rate low? Find the leak first

What does "conversion rate" actually measure on Shopify?

On Shopify, the online store conversion rate is the share of sessions that ended in an order. Shopify's analytics shows it as a funnel: sessions, sessions with add to cart, sessions that reached checkout, and sessions that completed checkout. The overall rate is simply the product of the three steps between them.

That matters because a single number hides which step is weak. Two stores can both convert at the same overall rate. One loses most visitors on the product page. The other loses them at payment. The first needs a better product page, the second needs a better checkout, and the headline number cannot tell them apart.

A few things also move the rate without the store changing at all. A new ad campaign that sends cold visitors lowers it. An email send to existing customers raises it. A sale lifts it for a week and flatters the month. So before anything else, write down what changed in your traffic over the period you are looking at.

Is your conversion rate actually low?

There is no number that every store should reach. Published benchmarks vary widely by category, price point, device and traffic source, and the averages they report blend very different businesses. A store selling a low-cost consumable and a store selling a mattress should not be compared, and neither should paid social traffic and returning email subscribers.

Use three comparisons instead:

1. Your own history. Compare the last four full weeks with the four before, and with the same weeks a year ago if you have them.
2. Device against device. Open Analytics and split by mobile and desktop. Most stores get the larger share of sessions on phones, and many see a clear gap in conversion between the two. A big gap points at the mobile experience, not at demand.
3. Source against source. Compare direct, email, organic search and paid. If paid traffic converts far below the rest, the problem may sit in the ad, the audience or the landing page the ad points to, not in the store as a whole.

If none of these shows a gap, the rate may not be low at all. It may simply be what your traffic produces, and the better lever is the offer or the audience.

Where in the path are visitors stopping?

Read the funnel from the top and find the step with the biggest drop relative to what you would expect.

Be careful with the middle row. Many stores treat the cart as a formality, but it is where shipping and total cost first appear, and an unexpected total is a well-documented reason people leave. We covered this in showing shipping cost before checkout and, for the last step, in the checkout optimisation guide.

Record at least two weeks, including a weekend, before you decide which step is weakest. A few days of data is noise.

Can slow pages or a broken mobile path explain a low rate?

Often. Speed and mobile usability are the two causes you can verify without running a single test, which is why they come first.

Google's Core Web Vitals give thresholds you can read against: for Largest Contentful Paint, 2.5 seconds or less is rated good, per web.dev. Check your product page in the field data for mobile, not just a lab score. If the main image or the buy button appears late, the visitor is already scrolling away. In one build we rebuilt the theme so that LCP went from 4.1s to 1.2s and the JavaScript shipped fell from 2.4MB to 310KB. We do not claim that speed alone produced any particular lift. It removes a cause of loss so that your tests are measuring the page and not the waiting. Our page weight guide shows how to find the largest files, and the third-party scripts guide covers the apps that add to them.

Then walk the buying path yourself on a 390 pixel wide phone. Open a product page from an ad link, choose a variant, add to cart, open the cart, start checkout. Note every place where you hesitate. Buttons below the fold, a size chart that opens a new page, a cart drawer that does not scroll: each of these is a leak you can see in two minutes.

What should you test, and in what order?

Once the measurable faults are fixed, choose a change for the weakest step and test it alone. Write it as a hypothesis: what you will change, for which visitors, and which number should move.

Evidence from real tests is the only guide worth trusting here, and it is often not what the team expected. In one test we ran, one variant showed the shipping cost on the product page and the control showed it at checkout. Each group had 24,105 sessions. The control produced 265 orders and the variant 578, which is a +118% difference in orders per session. A separate test of a sticky buy box ended at +31%. Those results come from specific stores and specific products, and yours will differ. What carries over is the method: pick the step that leaks, remove a surprise or a point of friction, and measure it against a control with the same number of visitors.

Two cautions. First, how much traffic you need depends on your baseline rate and the smallest change you want to detect, so check that before you build, as set out in how to A/B test a Shopify store. Second, do not run two tests on the same page at once, or you will not know which one did the work. The CRO roadmap shows how to queue ideas so that each test gets a clean slot.

When is a low conversion rate not a store problem?

Sometimes the store is fine and the traffic is the issue. Signs of this: sessions are high and add to cart rate is very low across every product; most of the visitors arrive from one cold paid audience; bounce is high on the landing page the ad points to; or the products people look at are not the ones the ad promised. In that case, changing buttons will not help. The fix is on the ad side or in the landing page match, and you should say so before spending on store changes.

The opposite also happens. Traffic is warm, product pages are viewed at length, and carts are created, but checkout completion is weak. That is a checkout, payment or shipping problem and deserves attention before anything about the product page.

What should you fix first, and how long does it take?

Start with a one-week job: split the funnel by device and by source, and write down the weakest step. That costs nothing and settles most arguments about where to work.

Next, spend a few days on what you can fix without testing: the speed of the product page on mobile, any obvious break in the phone buying path, and the surprise costs that appear late. Then let the numbers settle for two weeks before you start your first test.

After that, test one change on the weakest step with a stopping date set before launch. Allow for a first cycle of several weeks from measurement to a readable result; how many depends on your traffic.

If you would rather have the funnel measured and the leaks ranked by likely revenue impact on your own store, that is what our Store Teardown does. The services page lists the build and optimisation work that follows, and the Shopify CRO agency page covers the ongoing testing work.