.svg)
.svg)
%20(1).png)
Our Apps Have Driven
5X AOV | 2X Conversions | $30M+ Additional Revenue
5X AOV | 2X Conversions | $30M+ Additional Revenue
.svg)
5.0 (3000+ Reviews)
Shopify Apps
.svg)
CRO gets reduced to a few familiar numbers: conversion rate, bounce rate, cart abandonment, maybe AOV if we're feeling ambitious. But those numbers don't mean much without context. A 3% conversion rate can be healthy for one store and a warning sign for another, just as a high bounce rate can be perfectly normal on one page and a serious problem on another.
What matters is knowing which metric to look at, where in the customer journey to look at it, and what a change actually tells you.
This guide breaks down the CRO metrics that matter for ecommerce, how to benchmark them, how to spot conversion bottlenecks, and which numbers are useful for decision-making versus those that are better treated as supporting data.
CRO analytics connects behavioural data such as clicks, scroll depth, recordings, and funnel drop-offs with conversion data such as purchases, sign-ups, and completed checkouts. Looking at both gives you a better idea of what visitors are doing and where they are struggling.
The goal isn't simply to find where people leave. You also want to understand why they're leaving and what fixing that drop-off could mean for revenue. A bounce rate by itself rarely gives you that answer.
One of the easiest mistakes to make with Ecommerce CRO metrics is treating every number as equally important across the entire site. It isn't. Bounce rate can tell you something useful on a landing page, while checkout completion rate is far more useful once someone has started paying.
At this stage, visitors are deciding whether the page matches what they came for. Look at bounce rate, average time on page, and scroll depth to understand whether people are engaging with the experience.
If bounce rate is high, check the traffic source and the promise made before assuming the page itself is the problem. Someone arriving from an ad that promises one thing and landing on a page that delivers something else may leave immediately, even when the page itself is technically fine.
Add-to-cart rate becomes more useful here, along with product-page exits and pages per session. These metrics can help you understand whether visitors are moving from browsing to showing genuine purchase intent.
If people are viewing products but aren't adding anything to their cart, look at pricing, product information, imagery, reviews, shipping details, and other trust signals. The problem may have less to do with traffic quality and more to do with what the product page is communicating.
This is where cart abandonment and checkout completion rate matter most. If plenty of visitors add products to their carts but far fewer complete checkout, the friction is probably happening later in the journey.
Don't stop measuring once someone buys. AOV, revenue per visitor, repeat purchases, and early customer lifetime value signals can tell you whether the conversion you are generating is actually valuable.
Upsells, retention, and post-purchase email capture can all affect that picture. The important part is matching the metric to the stage rather than putting every available number on the same dashboard.
There isn't one CRO metric that explains everything. The useful approach is to track a small group of metrics that show what is happening at different points in the customer journey.
Conversion rate is calculated as (Conversions ÷ Total Visitors) × 100. Ecommerce conversion rates often fall around 2.5% to 3%, although the number varies considerably by category, device, traffic source, and price point.
It remains one of the clearest metrics for understanding whether more visitors are completing the action you care about. Just make sure you are comparing like-for-like traffic and time periods before treating a change as meaningful.
Add-to-cart rate is calculated as (Add-to-Carts ÷ Product Page Visitors) × 100. Around 8% to 10% is often used as a rough reference point, but your own category benchmark matters more.
This metric is particularly useful because it helps separate product-page issues from problems further down the funnel. If visitors aren't adding products to their carts, improving checkout probably isn't where you should start.
Cart abandonment rate is calculated as (1- (Completed Purchases ÷ Carts Created)) × 100. Cart abandonment commonly sits around 70%, so a high number isn't automatically a sign that something is broken.
Instead, look at the trend and compare it with your own historical performance. A sudden increase after a site change, pricing update, or shipping change is often more useful than the absolute number itself.
Checkout completion rate is calculated as (Completed Checkouts ÷ Checkout Starts) × 100. It shows how effectively your checkout page turns shoppers who have started checkout into completed purchases.
A weak checkout completion rate can point towards friction around shipping, payment, forms, account creation, or technical issues. It also helps distinguish checkout problems from earlier hesitation around the product or cart.
AOV is calculated as Total Revenue ÷ Number of Orders. There isn't one universal AOV to aim for because a fashion store, luxury retailer, and beauty subscription business will naturally have very different order values.
Start by looking at your own historical data. Once you understand how AOV is changing over time, industry benchmarks can provide additional context.
Revenue per visitor is calculated as Total Revenue ÷ Total Visitors. It brings conversion rate and order value together, which makes it useful when conversion rate alone doesn't tell the whole story.
For example, a store with a lower conversion rate may still generate more revenue per visitor if the customers who do convert have significantly higher order values. That can change which CRO problem deserves attention first.
Bounce rate and exit rate are often treated as if they mean the same thing, but they answer different questions. Bounce rate looks at sessions where a visitor leaves without moving to another page, while exit rate looks at how often a particular page is the final page viewed in a session.
Neither metric is especially useful in isolation. A high bounce rate on a blog post may be completely normal, while the same pattern on a paid landing page could indicate a problem with the experience or message.
Customer acquisition cost is calculated as Total Acquisition Spend ÷ New Customers Acquired. It connects website performance with what you are spending to bring new customers through the door.
If conversion rate improves but CAC remains high, the issue may sit further upstream. Traffic quality, channel mix, targeting, and acquisition costs can all affect CAC independently of what happens on your website.
A simple CLV calculation is Average Order Value × Purchase Frequency × Average Customer Lifespan. CLV gives CAC some much-needed context because the value of acquiring a customer depends on what that customer is worth over time.
A $60 acquisition cost means something very different for a customer worth $70 over their lifetime than for one worth $400. Looking at both metrics together gives you a clearer view of acquisition efficiency.
The following table gives you a practical starting point for the main CRO metrics covered above. These figures are directional rather than universal targets, so use them alongside your own historical performance.
Use these numbers as reference points rather than targets carved in stone. A store selling $20 products will behave differently from one selling $2,000 products, and traffic source can change conversion performance significantly.
Your trailing 90-day performance is usually the better starting point, with industry numbers providing additional context.
CRO metrics can show where visitors are dropping out of the buying journey. Start by reviewing the funnel from landing page to purchase and look for the stage with the biggest or most unexpected drop-off. Compare key stages such as:
For example, a strong add-to-cart rate but low checkout completion may point to an issue with the cart or checkout rather than the product page. Unexpected costs, limited payment options, technical issues, or a complicated checkout can all contribute to this drop-off.
Segment your data by traffic source, device, landing page, product, customer type, location, and campaign. An overall conversion rate can look healthy while one segment, such as mobile visitors, performs much worse and points to a specific usability or technical issue.
Once you find the weak point, use session recordings, heatmaps, surveys, support questions, or usability testing to understand what is causing it. Then test a focused change and measure its impact on purchases, revenue per visitor, or profit.
Measuring a CRO test involves more than comparing conversion rates. You need enough data to determine whether the difference is meaningful or simply the result of random variation.
Before launching a test, define the primary metric, baseline conversion rate, meaningful improvement, target audience, test duration, and decision criteria. Choose one primary metric in advance to avoid selecting a favourable result after seeing the data.
Sample size depends on your baseline conversion rate, expected improvement, confidence level, statistical power, number of variations, and traffic split. Avoid stopping a test simply because one version takes an early lead. Set the required sample size before the test begins.
Statistical significance helps determine whether the difference between variations is likely to be more than random chance. However, a statistically significant result does not automatically mean the change is valuable, so also consider the size of the improvement, additional conversions, and its impact on revenue or profit.
User behaviour naturally varies because of factors such as traffic sources, promotions, seasonality, device mix, holidays, and technical issues. Run tests over a representative period, keep traffic allocation consistent, check tracking, and avoid making decisions based on short-term swings.
Early results can change significantly as more visitors enter the test. Repeatedly checking results and stopping as soon as one variation leads can increase the risk of a false winner. Define your decision rules upfront and treat tests without a clear result as inconclusive.
Once the test reaches its planned sample size, review the primary metric first, then check secondary effects such as AOV, revenue per visitor, refunds, cancellations, or lead quality. The goal is not just a statistically reliable result, but an improvement that creates meaningful business value.
The right CRO tool depends on what you need to understand. Some tools are designed to measure traffic and conversions, while others help explain visitor behaviour or run experiments.
Analytics platforms such as GA4 give you the basic picture: traffic, conversions, funnels, and drop-offs. For most stores, that's where the measurement setup should start because it provides the foundation for understanding what is happening across the site.
Heatmaps and session recordings from tools such as Hotjar and Clarity help explain what the numbers don't. If checkout completion falls, for example, a recording might reveal that visitors repeatedly struggle with a particular field or step.
These tools are particularly useful when you know that something is going wrong but can't explain why from analytics alone. They add behavioural context to the conversion data.
A/B testing platforms such as Optimizely and VWO help you test changes instead of relying entirely on assumptions. They can also make the statistical side of experimentation easier to manage.
The important part is having a clear hypothesis before you test. A testing platform won't make a weak hypothesis useful simply because it produces a statistically significant result.
You don't necessarily need an expensive CRO stack to get started. GA4 plus a free heatmap tool can give a smaller store a useful amount of information, provided the tracking is set up correctly.
A more developed setup with paid analytics, testing, and recording tools can easily cost $100 to $400 a month. Enterprise experimentation platforms and dedicated CRO teams can push that figure into the thousands.
The right level depends less on how impressive the tool looks and more on whether you have enough traffic and testing volume to use it properly. Paying for advanced experimentation software won't help much if your site doesn't generate enough traffic to run meaningful tests.
CRO metrics become useful when they're connected to an actual question. Which stage is losing people? Is the change you are seeing meaningful? And if you fix it, does that translate into more revenue?
Start with the metrics that match each funnel stage and benchmark them against your own historical performance. Industry numbers can provide context, but they shouldn't replace your own data.
Most importantly, don't let engagement numbers become the goal when the business outcome is revenue. More clicks, longer sessions, or more page views only matter when they contribute to the outcome you are trying to improve.
