What are the key KPIs in ecommerce

In ecommerce, it is very easy to fall into the trap of measuring everything while at the same time understanding almost nothing. Dashboards are full of numbers, reports show dozens of charts, and teams regularly talk about traffic, sales, ROAS, or conversion rate, yet in practice many companies still cannot answer the most important question: which indicators really tell us whether the business is growing in a healthy way, and which merely create the impression of control. That is why the conversation about KPIs in ecommerce should not begin with a list of metrics. First, it is necessary to understand what decisions the company wants to make on the basis of them and which part of the business model these indicators are actually supposed to describe.

This is especially important today, when ecommerce is no longer just a sales channel, but part of a much larger system involving marketing, margin, logistics, retention, pricing policy, promotions, customer service, and integrations with other tools. In such an environment, KPIs cannot be a random set of data from Google Analytics. They should form a logical model for assessing the condition of the business. Good KPIs show not only what happened in the store, but also why it happened and what needs to be done next. That is exactly what determines whether analytics supports growth or is merely reporting on the past.

Why observing sales alone is not enough

Many companies begin ecommerce performance analysis with revenue. That is natural, but from the perspective of managing the business it is definitely not enough. Revenue growth alone does not yet say whether the store is operating efficiently, whether customer acquisition is profitable, whether the traffic is of good quality, whether the purchasing process is not losing customers along the way, and whether the growth is not in fact being bought at the expense of profitability. After all, revenue can be increased through strong discounts, aggressive paid campaigns, and broad exposure of the offer, while at the same time margin deteriorates, customer acquisition cost rises, and customer value over time weakens.

For this reason, the key KPIs in ecommerce must always be analyzed across several layers at the same time. The first layer concerns sales scale, the second the quality of traffic and the effectiveness of the purchase path, the third the economics of the business, and the fourth retention and customer value over a longer horizon. Only by combining these perspectives do we get a picture that is suitable for making management decisions.

Sales KPIs that are really worth starting with

In every ecommerce analysis, the foundation remains revenue, number of transactions, and conversion rate. The problem is that these three indicators are often treated too superficially. Revenue shows scale, the number of orders indicates volume, and conversion rate helps us understand how effectively the store turns traffic into purchases. However, only when these data points are compared with other metrics does it become clear whether growth results from a better offer, higher-quality traffic, a more effective checkout, greater product availability, or simply increased advertising spend.

A very important indicator is also average order value, or AOV. In practice, it shows how much one customer leaves on average during a single purchase. This is a particularly important metric wherever the business works with bundles, cross-sell, upsell, free shipping thresholds, or promotional policy. Growth in AOV often makes it possible to improve performance without the need to increase traffic proportionally. Equally important is revenue per session, because it combines traffic quality and store effectiveness. If this indicator is growing, it usually means that the organization is monetizing each visit better, rather than simply inflating the number of users.

In practice, conversion rate should be analyzed very carefully. The number itself has little value without context. A good conversion rate looks different in a store with everyday products, different in a premium category, and different again in a B2B model, where the decision-making path is longer and more complex. That is why conversion should be read not as a universal benchmark, but as an indicator of change over time and as a result for specific segments: channels, devices, countries, categories, campaigns, or customer groups. Google Analytics 4 provides ecommerce metrics and such indicators as average purchase revenue, but meaningful interpretation still depends above all on the business model and proper data segmentation.

Funnel KPIs show where sales are really being lost

One of the most common mistakes in ecommerce is analyzing the final result without understanding at which stage of the purchase process the problem appears. A store may have good traffic and an attractive offer, yet still lose sales between the product page and the cart. It may also effectively collect add-to-cart actions, but have a weak checkout, unclear delivery costs, or too long a form. That is exactly why purchase funnel KPIs are of key importance.

The most important of these include the transition rates between product view, add to cart, begin checkout, and purchase. This is not only about the classic cart abandonment rate, but about the entire logic of the user moving through the subsequent steps. If the number of view_item events is high, but add_to_cart is low, the problem may lie in price, product presentation, availability, or the fit of the offer to the traffic intent. If begin_checkout is high, but purchase drops noticeably, the causes usually need to be sought in the finalization process itself. In GA4, these stages can be analyzed, among other ways, through Purchase Journey and Checkout Journey, which show the number and percentage of users dropping off at successive funnel stages.

This is one of those areas in which data starts to become truly useful only when it is connected with a product or UX decision. A drop between add_to_cart and purchase is not yet an insight in itself. Insight appears only when the store is able to connect that drop with a specific source of the problem: the mobile version, payment method, shipping costs, discount codes, a technical error, or too many steps. In mature ecommerce, KPI analysis does not end with reading the number. Its purpose is to indicate the place where something needs to be simplified, improved, or tested.

Traffic and customer acquisition KPIs tell you whether sales have healthy sources

Sales in ecommerce should never be analyzed separately from traffic sources. Two stores may achieve the same revenue, but one will do it thanks to organic traffic, a base of returning customers, and a good margin, while the other will do it thanks to expensive paid campaigns that will stop delivering results the moment the budget is switched off. That is why the key KPIs should always include indicators related to user acquisition and channel quality.

At this level, it is worth observing the number of users, number of sessions, share of new and returning customers, conversion rate by channel, revenue by traffic source, and customer acquisition cost, or CAC. It is also very useful to look at blended ROAS or MER, because they make it possible to assess marketing effectiveness more broadly than at the level of a single campaign. Especially in companies that run many activities in parallel, analyzing only one channel can lead to incorrect decisions. Google Analytics 4 distinguishes between Traffic acquisition and User acquisition reports, where the former shows session sources, while the latter focuses on the sources of new users. This distinction matters, because the effectiveness of current traffic is analyzed differently from the real acquisition of new customers.

In practice, good channel analysis does not come down to the question of which channel delivers the highest number of transactions. Much more important is which channel delivers the most profitable growth, which attracts customers with the highest AOV, which delivers returning customers, and which works best at a specific stage of the purchase path. A channel that has a lower last-click share in sales may at the same time play a very important role in building demand and initiating the first visit. That is why channel KPIs need to be analyzed in the logic of the entire path, and not only on the basis of a single attribution report.

Margin, promotion costs, and returns – without them KPI analysis is incomplete

One of the most underestimated topics in ecommerce is the difference between sales and profitability. Many organizations report revenue in great detail, but monitor margin, discounting levels, the share of shipping costs, return costs, or the impact of specific categories on the final result much less effectively. Yet from the perspective of business management, these indicators often determine whether ecommerce is actually building value.

That is why the key KPIs should also include gross margin, margin after discounts, share of marketing costs in revenue, refund rate, return rate, and contribution margin, if the organization is able to calculate it. In multi-channel or multi-market companies, it very quickly turns out that high sales in one segment do not necessarily mean a healthy result. In fact, the opposite may be true – the greatest volume is often where the highest promotion, logistics, and after-sales service costs accumulate. For this reason, sales and marketing KPIs should always be compared with data from ERP, warehouse systems, and finance. Web analytics alone will not show the full picture of profitability, even if it measures user behavior very well.

This is the moment when it becomes clear why mature ecommerce needs not only a good analytics tool, but also a coherent data architecture. If the team is unable to connect traffic, transaction, margin, return, and marketing cost data, then management is looking at several different truths at the same time. And this usually ends with decisions being made on the basis of an incomplete picture.

Retention and customer value over time are more important than many companies think

Growth in ecommerce does not consist solely in continuously bringing new users to the store. At a certain point, the most important question becomes whether customers return, how often they purchase again, how much they are worth over a longer period, and which segments generate the highest value for the business. That is exactly why the key KPIs should include repeat purchase rate, customer lifetime value, share of returning customers in revenue, average time between purchases, and cohort retention.

These metrics are particularly important wherever acquisition costs are rising, price competition is strong, and further growth can no longer rely solely on increasing media budgets. In such situations, the business starts to be won not by the one who buys the most traffic, but by the one who can better monetize the relationship with the customer over time. GA4 offers retention reports and cohort analysis, which help assess how many users return after the first visit or first purchase, but the real value appears only when these data are connected with the CRM, orders, and the logic of customer segmentation.

In practice, retention changes the way of thinking about the entire ecommerce business. Instead of asking only how much the campaign cost and how many orders it generated, the company starts asking what kind of customer the campaign brought in and whether that customer returned after 30, 60, or 90 days. This is a much more mature model for evaluating effectiveness, because it shows not only the cost of the first sale, but the real quality of growth.

How to analyze KPIs so as not to draw incorrect conclusions

The biggest problem is usually not the lack of data, but incorrect interpretation of the data. KPIs in ecommerce should be analyzed in layers and in context. First, the overall result needs to be examined, then the segments, and then the relationships between the indicators. A drop in conversion does not always mean there is a problem with the store. Sometimes it results from a change in the channel mix and a higher share of cold traffic. A drop in AOV is not always bad if the company is at the same time increasing purchase frequency. Revenue growth is not always a success if it comes at the cost of lower margin and higher acquisition cost.

Good KPI analysis therefore begins with business questions. Do we want to improve profitability? Are we trying to increase the number of new customers? Do we want to improve the effectiveness of the mobile checkout? Is our goal to grow sales in a specific market? Each of these questions sets a different set of priority indicators. There is no single perfect dashboard for everyone. There are, however, well-defined decisions to which the appropriate metrics need to be matched.

Consistency of methodology is also very important. If a company calculates conversion once on sessions and another time on users, if it attributes revenue once to the day of purchase and another time to the day the traffic was acquired, or if different teams use different definitions of an active customer, conclusions very quickly stop being comparable. That is exactly why ecommerce analytics should not only be available, but also standardized and understandable across the entire organization.

Why correct data measurement is just as important as the KPIs themselves

In practice, ecommerce cannot be analyzed well if the measurement itself has been implemented incorrectly. In Google Analytics 4, ecommerce data does not appear automatically – it requires the implementation of the appropriate ecommerce events, such as view_item, add_to_cart, begin_checkout, or purchase. Only when they are sent correctly do they feed purchase reports and make it possible to analyze user behavior at the subsequent stages of the path. GA4 is also based on an event model, and the most important business actions can be marked as key events, meaning events that are particularly important to the success of the business.

This matters enormously, because many organizations try to draw precise conclusions from incomplete or inconsistent data. A common problem is the lack of correctly transmitted transaction value, incorrect event structure, duplicate purchases, inconsistent campaign tagging, or discrepancies between analytics data and the back office. It is also important to remember that reports do not always show data in real time – Google describes this area as data freshness, meaning the time needed to process the data and make it available in reports. If the team does not understand these limitations, it becomes very easy to misjudge the current situation.

In the case of engagement metrics, it is also worth knowing the definitions. In GA4, engagement rate refers to the percentage of engaged sessions, and such a session is one that lasts longer than 10 seconds, has a key event, or includes at least two page or screen views. Without understanding this logic, it is easy to treat engagement rate as a simple replacement for older traffic quality metrics, while in practice its meaning depends on the structure of visits and the user’s intent.

How to build a KPI set that will actually be useful for ecommerce

The best KPI set is not the longest one, but the most useful one. In practice, a well-functioning ecommerce reporting model is most often based on a limited number of indicators assigned to specific business goals. Management usually needs a different level of view than the performance team, CRM, or ecommerce manager. That is why sensible analytics does not consist in showing everything to everyone, but in building several reporting layers: strategic, operational, and experimental.

At the strategic level, the company should see revenue, margin, number of orders, AOV, CAC, share of returning customers, and growth rate. At the operational level, conversion by device and channel, purchase funnel performance, campaign effectiveness, refund rate, or product availability become more important. At the experimental level, specific hypotheses related to UX, price, product exposure, checkout, or personalization are analyzed. Only such a structure makes KPIs stop being dead numbers and start supporting the daily management of ecommerce.

Key KPIs in ecommerce should lead to decisions, not only to reports

That is why the most important KPIs in ecommerce are not those that look best on a dashboard, but those that make it possible to notice a problem faster, understand its cause better, and decide more accurately what to do next. In practice, this means the need to look simultaneously at sales, the purchase funnel, traffic sources, acquisition costs, margin, and retention. Only then does the organization see whether the store is truly growing, or merely increasing turnover without building lasting value.

From our perspective, this is exactly where a mature approach to ecommerce begins. Not with fascination for a single indicator and not with a random set of reports, but with a coherent analytical model that connects data with the business objective. And when such a model is well built, KPIs become one of the strongest growth tools – they help teams react faster, plan better, and run ecommerce in a truly predictable way.