E-Commerce Analytics: Metrics That Matter for Stores

An online store generates an extraordinary amount of data: every click, scroll, add-to-cart, and abandoned checkout leaves a trace. The problem is rarely a lack of numbers. It is knowing which of them actually predict revenue and which are simply noise dressed up as insight. A store owner who chases every figure ends up reacting to everything and improving nothing.

This guide cuts through the clutter. It walks through the e-commerce metrics that consistently matter, explains where each one sits in the customer journey, and shows how to turn each into a concrete action. It pairs naturally with our broader thinking on turning analytics into actionable decisions and sits within our pillar guide on data analytics for growing businesses.

The funnel is the map

Every useful store metric belongs somewhere in the purchase funnel: attracting visitors, converting them into buyers, growing the value of each order, and keeping them coming back. Organising your analytics around these four stages stops you from staring at isolated numbers and starts you asking the more useful question of where, exactly, customers are slipping away.

Once you see the funnel as a map, prioritisation becomes obvious. A store with heavy traffic but a weak conversion rate has a very different problem from one that converts well but cannot attract anyone. The metrics tell you which problem you actually have, so you spend effort where it will move revenue most.

The funnel also keeps teams honest about cause and effect. A spike in traffic that does not move sales is not automatically a success, and a dip in orders is not automatically a crisis if it follows a deliberate cut in low-quality traffic. By placing every metric in its funnel stage, you can read changes in context rather than reacting to each wobble in isolation.

Conversion rate: the headline number

Conversion rate, the share of visitors who complete a purchase, is the single most revealing store metric because it summarises how well everything between landing and checkout is working. A small improvement here multiplies across all your traffic, which is why optimising it is often cheaper than buying more visitors.

Read it by segment, not in aggregate

A blended conversion rate hides the truth. Mobile usually converts very differently from desktop; paid traffic behaves differently from organic; first-time visitors differ from returners. Breaking the number down by these segments is almost always where the actionable insight appears. Our guide to what makes a website convert digs into the design factors behind these gaps.

Pair it with intent

Not all traffic deserves the same conversion expectation. A visitor from a high-intent product search is far closer to buying than someone arriving from a broad awareness campaign. Judge conversion against the intent of the source, not against a single universal benchmark.

Average order value and revenue per visitor

Conversion rate tells you how often people buy; average order value tells you how much they spend when they do. Multiplying the two gives revenue per visitor, arguably the cleanest single measure of how much each session is worth. Improving either lever lifts the same outcome, and the better opportunity is not always the obvious one.

Levers for order value

Thoughtful product recommendations, bundles, free-shipping thresholds, and well-timed upsells all nudge order value upward without needing a single extra visitor. Because they apply to people already buying, the returns can be quick. Our conversion optimisation checklist covers many of these tactics in detail.

Why revenue per visitor prevents tunnel vision

Focusing on conversion rate alone can quietly hurt the business. A discount or a stripped-down checkout might lift the share of visitors who buy while shrinking what each one spends, leaving total revenue flat or lower. Revenue per visitor catches this trade-off because it folds both effects into one figure. When you optimise against it, you are forced to weigh frequency and value together, which is almost always the healthier objective.

Cart abandonment: where revenue leaks

A large share of shoppers add items to their cart and never complete the purchase. Each abandoned cart is a customer who wanted to buy but was stopped by something: an unexpected cost, a forced account creation, a slow or confusing checkout, or a payment option they did not trust. The abandonment rate, and the reasons behind it, point directly at fixable friction.

Diagnose before you discount

The reflex to win back abandoners with a discount can be expensive and habit-forming. Before reaching for a coupon, investigate why people leave. Often a clearer cost breakdown, a guest checkout option, or a faster page solves more abandonment than any promotion, and it does so without eroding margin.

Watch the checkout as a mini-funnel

The checkout itself is a funnel in miniature, with its own stages: cart, contact details, shipping, payment, and confirmation. Measuring drop-off at each step turns a vague abandonment figure into a precise diagnosis. If most shoppers vanish at the shipping step, the problem is probably cost or delivery options; if they leave at payment, it may be a trust or technical issue. Instrumenting these micro-steps is one of the highest-return analytics projects a store can undertake, because every recovered step flows straight to revenue.

Retention: the metric founders underrate

Acquiring a new customer almost always costs more than keeping an existing one. Yet retention metrics, repeat purchase rate, customer lifetime value, and time between orders, are routinely neglected in favour of flashier acquisition figures. A store that retains well can spend more confidently on acquisition because each customer is worth more over time.

Lifetime value reframes everything

When you measure what a customer is worth across their whole relationship rather than a single order, your entire strategy shifts. Channels that look unprofitable on first purchase can be strongly profitable over a year. This is why connecting retention data to acquisition spend, a theme in our piece on measuring marketing return on investment, changes which decisions look smart.

Cohorts reveal whether you are improving

A single retention number is hard to interpret in isolation. Grouping customers into cohorts by the month they first bought, then tracking how each cohort behaves over time, shows whether recent changes are actually making customers stickier. If newer cohorts return faster and spend more than older ones, your product and experience are improving. If they fade quicker, something has slipped, and the cohort view catches it long before a blended average would. This longitudinal lens is what separates stores that compound from those that simply churn through buyers.

Traffic quality and acquisition metrics

Before a visitor can convert, they have to arrive, and not all arrivals are equal. Two stores with identical visitor counts can have wildly different revenue simply because one attracts buyers and the other attracts browsers. That is why acquisition metrics deserve as much attention as the conversion numbers further down the funnel. Looking at sessions in isolation tells you almost nothing; looking at what those sessions do tells you everything.

Judge channels by outcome, not volume

A channel that sends a flood of cheap traffic can look impressive on a dashboard while quietly losing money, because the visitors it brings rarely buy. Conversely, a small stream of highly relevant visitors can be your most profitable source. Evaluate each channel by the revenue and the customers it produces, not by raw clicks. Engagement signals, such as the share of sessions that view a product or reach the cart, give an early read on quality long before the sales data fully matures.

Watch the cost of every acquired customer

Acquisition cost is the counterweight to lifetime value, and the two only make sense together. A customer who costs a great deal to acquire can still be a bargain if they buy repeatedly for years, while a cheap customer who never returns may not cover the cost of winning them. Tracking acquisition cost by channel, and comparing it against the value each channel's customers go on to generate, is one of the most consequential analyses a growing store can run.

Building a store dashboard that earns its place

A dashboard should answer questions, not decorate a screen. The best store dashboards are deliberately sparse: a single primary metric for each funnel stage, a counter-metric or two to guard against blind spots, and trend lines long enough to separate signal from seasonal noise. Everything else can live a click away, surfaced only when a headline number prompts a deeper look.

Review on a rhythm, act on a trigger

Set a regular cadence to review the dashboard, weekly for fast-moving stores, and pair each metric with a threshold that triggers action when crossed. This combination of routine and rule keeps the team responsive without drowning them in daily noise. The aim is a dashboard that quietly does its job in the background and speaks up only when a real decision is due.

Turning store metrics into action

The point of all this measurement is movement. For each metric, decide the action it would trigger: a weak mobile conversion rate prompts a checkout review, a high abandonment rate prompts a friction audit, a low repeat rate prompts a retention programme. Without a pre-agreed response, even the cleanest dashboard just describes a problem you never solve. The full sequence of store fixes is laid out in our ecommerce optimization guide, and the discipline of acting on findings is covered in common analytics mistakes to avoid.

Track a small, funnel-aligned set of metrics, read them by segment, and attach an action to each. Do that consistently and your analytics stop being a report card and start being a steering wheel. The stores that grow fastest are rarely those drowning in dashboards; they are the ones that have chosen a handful of metrics they trust and built the muscle of acting on them, week after week.

Frequently asked questions

References

1. Baymard Institute, research on checkout usability and cart abandonment, baymard.com
2. Google Analytics Help, documentation on e-commerce reporting, support.google.com

Want help applying these metrics to your store? Browse our resources on data analytics, or get in touch to discuss your goals.