eCommerce Analytics: What It Is, Types, Metrics and Tools

eCommerce analytics is the process of collecting, analyzing and interpreting data from every source that affects your online store, so you understand customer behavior, sales performance and marketing results. It follows the whole customer journey, from the first visit and the first order to repeat purchases and referrals, and it exists for one reason: to help you decide what to do next.

With purchase history, on-site behavior and campaign data in one view, you can improve conversion rates, keep more customers, lower acquisition costs and personalize the shopping experience. This guide explains what eCommerce analytics is, the four types, the metrics that matter, the tools to combine and how to start. Omniconvert Commander (formerly Nexus), the AI eCommerce growth engine, covers the customer side of that picture: segments, lifetime value and profit per customer.

What is eCommerce analytics?

An online store produces data at every step. Your analytics tool records sessions, sources and page views. Your store platform records orders, products, refunds and customer accounts. Your ad and email platforms record spend, clicks and opens. Surveys and reviews record what customers think. eCommerce analytics connects these sources so you can answer questions that no single tool answers alone, such as "which channel brings customers who buy a second time?"

It helps to separate three ideas that people often mix up. Reporting shows numbers. Business intelligence organizes those numbers into dashboards so teams can monitor performance; our guide to business intelligence in eCommerce covers that layer in detail. Analytics goes one step further: it explains why a number moved and what you should do about it. A store needs all three, but only analytics changes what you do on Monday morning.

eCommerce analytics also includes customer analytics, the part that looks at people rather than sessions: who your customers are, how often they buy, how much they are worth and when they stop buying. For most stores this is the least developed part, and it is where most of the profit is decided.

Why eCommerce analytics matters

McKinsey's DataMatics survey found that companies that make intensive use of customer analytics are 23 times more likely to clearly outperform their competitors in new-customer acquisition, and nine times more likely to surpass them in customer loyalty [McKinsey DataMatics, 2013]. In daily practice, analytics helps an online store in these ways:

• Guide customers through the funnel. Behavior data shows where visitors hesitate or leave, so you know which step to fix first.
• Run better upsell and cross-sell campaigns. Purchase history shows which customers respond to which offers.
• Make better product recommendations. Browsing and buying patterns show which products customers view and buy together.
• Build a data-driven pricing strategy. Demand and competitor data support a more flexible pricing strategy, and large retailers already use data to set prices, as The Atlantic's report on online pricing describes.
• Improve customer retention. Trends in the repeat purchase rate show whether customers come back, and when they stop.
• Improve the user experience. Bounce rates, heatmaps and recordings show which areas of a page get attention and which get ignored.
• Fund the channels that bring customers. Channel data shows which sources drive first orders, and which messages work on each one.
• Segment and target better. Customers are not all alike, and they expect offers that fit them.
• Retarget more precisely. Customer data lets you build narrower retargeting audiences from behaviors you have already analyzed, a point BigCommerce's guide to customer analytics also makes.
• Get a better return on marketing. Watching how your KPIs move after each campaign shows where to cut and where to invest.
• Identify your best customers. Customer lifetime value combines the number of orders, the average order value and the time a customer stays, so you know who deserves the most attention.

Types of eCommerce analytics

You can sort eCommerce analytics by the question it answers. This is the most useful lens, because it tells you how mature your analysis is and what the next step is.

Most stores live in descriptive analytics. The move from describing the past to predicting and prescribing is the subject of our essay on data-driven retail, and AI analytics in retail shows how AI speeds up the last two steps.

You can also sort eCommerce analytics by the data it uses:

• Customer analytics: who buys, how often, how much and for how long. Demographics, purchase history, segments, retention and lifetime value.
• Sales and product analytics: what sells, at what margin, with what return rate, and which products bring repeat buyers.
• Marketing and acquisition analytics: traffic sources, campaign results, acquisition cost and return on ad spend.
• Behavioral analytics: how visitors move through the site, where they click, scroll and leave.
• Financial analytics: margins, shipping and refund costs, and profit per order and per customer.

eCommerce analytics metrics to track

You can track hundreds of numbers. The ones below decide whether the store grows profitably, and every other metric usually explains one of them. Our list of important eCommerce metrics goes deeper on each one.

A short worked example shows why these metrics belong together. A store has 50,000 sessions in a month and 1,000 orders, so its conversion rate is 1,000 ÷ 50,000 = 2%. Revenue is $65,000, so the average order value is $65. If 600 of those orders came from new customers and acquisition cost $24,000, CAC is $40. Each new customer's first order brings $65 of revenue, but at a 40% gross margin only $26 of gross profit. The first order does not pay back the $40 CAC. The store makes money only if customers buy again, which is why the customer retention rate and lifetime value matter as much as traffic.

Return on ad spend has the same trap: it measures revenue, not profit. Read what ROAS is before you judge a channel on it alone, and look at the churn rate next to retention, since one is the mirror of the other.

eCommerce analytics tools

Most eCommerce platforms come with built-in analytics. They are a good start, but they rarely show the full picture or connect your data sources. That is why most stores add one tool or a mix of the platforms below, which give more detailed eCommerce reports on sales, customer behavior, margins, lifetime value and RFM segments. For a wider comparison, see our roundup of eCommerce analytics and attribution tools.

Google Search Console

Google's tools are free and hard to avoid. Google Search Console helps you optimize an eCommerce store for search: it reports errors that stop pages from being indexed, missing information and pages that do not work properly, and it alerts you when it detects a problem. Since organic search is how many shoppers find a product, this is where search analytics starts. To use it, verify that you own the site, submit your sitemap, and ask Google to recrawl your URLs when you change important pages. From then on you have access to the indexing reports and the performance reports on queries, clicks and positions.

Google Analytics 4

Google Analytics is the most widely used analytics tool for online stores. Its current version, Google Analytics 4 (GA4), replaced Universal Analytics: standard Universal Analytics properties stopped processing data on July 1, 2023. Older guides, such as Google's legacy Universal Analytics setup guide, describe the previous version. In GA4 you create a property, add a web data stream for your store and install the Google tag, either directly in your page code, through your store platform's integration or through Google Tag Manager. Data starts to appear only after the tag is live; GA4 does not collect anything from before that point.

Instead of the old Universal Analytics custom reports, GA4 uses Explorations, where you combine the dimensions (such as city or device) and metrics (such as sessions, engagement rate or purchases) that matter to you.

eCommerce tracking in Google Analytics

Universal Analytics had a standard eCommerce report and an enhanced eCommerce report, both documented in the legacy Universal Analytics eCommerce guide. GA4 replaces both with recommended eCommerce events, such as view_item, add_to_cart, begin_checkout and purchase. When your store sends these events with product and order details, GA4 shows product performance, revenue, average purchase value and the checkout funnel, so you can see at which step shoppers drop out and which segments drop out most.

Campaign tracking with UTM parameters

To measure campaigns, add UTM parameters to the links in your ads, emails and posts. They record the source (the site, advertiser or platform), the medium (such as CPC, email or banner) and the campaign name, and utm_term can carry the paid search keyword. The utm_content parameter separates versions of the same ad or links in the same email: if an email has two call-to-action links, give each a different utm_content value to see which one converts better. Google's guide to collecting campaign data with custom URLs explains the parameters, and the same logic applies in GA4. Use one naming convention across the team, or the same channel will show up under three names.

Hotjar

Analytics is not only numbers. Hotjar, now part of Contentsquare, shows how visitors actually experience your site, which traffic reports cannot do (a clear value proposition, if you ask us). It offers four main features:

• Heatmaps highlight the most clicked or tapped areas of a page and how far visitors scroll. Clicks and scrolls show what visitors care about and which paths they take; see our list of heatmap tools for alternatives.
• Recordings replay real sessions, so you see what visitors click, what they skip and where they get stuck. They are the fastest way to find usability problems.
• Feedback widgets let visitors react to your copy, design or features on the page itself.
• Surveys ask specific questions to visitors who meet a behavior trigger. Place them on the pages that heatmaps and recordings flagged, to learn the reason behind the behavior.

Omniconvert Commander

Traffic and behavior tools stop at the session. Omniconvert Commander's customer intelligence starts at the customer. Omniconvert built it for customer value optimization: it reads your store's order and customer data, groups customers into RFM segments, calculates customer lifetime value and True Profit per order and per customer, and pushes segments directly to Meta Ads, Google Ads and Klaviyo. Instead of building these views by hand in Google Analytics and spreadsheets, you get them as ready reports and use them in your campaigns. The next section explains what this customer layer measures.

Customer analytics: the layer most stores miss

A store that knows its traffic but not its customers optimizes for the first order. A store that knows its customers optimizes for profit, because acquisition costs keep rising and retention is where margin comes from. When you focus on your most valuable customers, you improve your eCommerce profit margin and the return on your marketing.

The usual starting point is RFM analysis, one of the main customer segmentation models. It scores each customer on three values, and the RFM score combines them to identify your best customers:

• Recency: the time since the customer's last order.
• Frequency: how often the customer orders, or the average time between orders.
• Monetary value: the customer's spend, as a total or as an average per order, which reflects their purchasing power.

On top of RFM, a complete customer view adds buying behavior (orders, products, dates, spend, AOV and the channels that brought each customer), customer value (lifetime value, retention rate and profit per customer) and customer feedback. Satisfaction data, such as the Net Promoter Score, which sorts customers into promoters, passives and detractors, tells you why a segment behaves the way it does. Our guide to customer intelligence in eCommerce shows how these sources fit together.

How to start with eCommerce analytics

1. Define your goals and decisions

   Write down the three to five decisions analytics should support, such as where to spend acquisition budget or which customers to target for a second order. Each decision points to the metrics you need.
2. Choose your KPIs

   Pick a short list from the metrics table above and define each one in writing, including what counts as revenue and what counts as a customer. Our guide to KPI tracking helps you keep the list short.
3. Set up and test tracking

   Install GA4 with eCommerce events, tag every campaign with UTMs and connect your store data. Then test it: configure and test eCommerce tracking in Google Tag Manager, place a test order, and compare reported purchases with real orders for the first days.
4. Unify your data

   Merge orders, customers, products and campaigns into one dataset, so every order belongs to a customer and every customer to a source. Without this step, customer analytics is not possible.
5. Build one dashboard, then the reports

   Put the KPIs on one eCommerce dashboard and make each chart answer a question; actionable dashboards show how. Add detailed reports from the list of top eCommerce reports only when a dashboard number raises a question, and keep simplified reporting for the wider team.
6. Act, test and compare

   Turn each finding into an action or an A/B test, and measure it with the same metric. Compare your numbers with the market in the real-time eCommerce benchmark report to see whether a change is yours or the market's. Our guide to data-driven decision making covers the full process.

The most common eCommerce analytics failure is trusting a revenue number that does not match the store. A tag fires twice on the thank-you page, or some payment methods send buyers to an external page and they never return, so GA4 reports 15% more or 20% fewer purchases than the back end. The team notices only after weeks of reports, stops trusting every dashboard and goes back to decisions by opinion. Check the gap every week, and if transactions are missing, follow a guide to debugging missing Google Analytics transactions before you build anything on top of the data.

Real-world example: from analysis to results

Analytics shows where a store loses customers; experiments show which fix works. AliveCor used Omniconvert to run a structured A/B testing program and achieved a 21 percent lift in conversion rate, a 5 percent lift in revenue per visitor, and 94 percent statistical relevance across their experiments [Omniconvert, AliveCor case study]. The lesson for any analytics setup is the same: each insight should end in a hypothesis, a test and a number you can compare with the baseline.

The same applies to customer analytics. When you know which segment buys again and which one leaves after one order, you can test a different offer for each one and measure the result in repeat purchase rate and lifetime value, not only in first-order conversion.

eCommerce analytics FAQs