KPIs for Ecommerce: The Numbers a Store Owner Should Know

You look up KPIs for ecommerce and get a list. Sometimes it has 12 numbers, sometimes 78, and usually a company that sells software wrote it.

I read the 18 text pages in Google’s top 20 results for “KPIs for ecommerce”. Thirteen gave formulas, and six of those got at least one wrong. Not one mentioned WooCommerce, or said where WooCommerce shows a number.

Store owners answering each other on Reddit sound different. Of the 32 owners whose posts I read, the longest list anyone gave was four numbers. One put it plainly: “Sales, margin, traffic, conversion rate. Everything else is just supplementary”.^[1] The complaint they wrote most about was that the numbers did not agree.

Before RuleInside, I founded EarthMeals, a plant-based ready-meals brand that sold online. Every customer cost more in money, time and attention than the business could sustain. Today I build managed AI voice and chat agents for WordPress stores. I lead every deployment personally, and my team runs the infrastructure. Nothing in this guide needs my product, and it recommends no tool.

KPI stands for key performance indicator. A metric is anything your store can count: visits, orders, emails, refunds. A KPI is a metric you have tied to a goal and to a decision. When it moves, you already know what you will look at next.

So KPIs for ecommerce are a short list chosen from a long one. People search for “key performance indicators for ecommerce” and “ecommerce metrics” as if they were the same thing. They overlap, but the difference decides what you watch. A store can track hundreds of metrics. It should watch only a handful of KPIs.

The pages Google ranks first for kpis for ecommerce cannot agree on how many. ThoughtSpot, which sells analytics software, advises: “Focus on 8-12 KPIs at most to avoid analysis paralysis”.^[2] Gorgias, which sells a help desk, says: “Track 3-5 KPIs per goal, 10-15 total across your business”.^[3] Another list in the top ten runs to 78 entries.

Owners are stricter than any of them. Among 32 owners in 22 Reddit threads I read, sales or revenue came up most often, named by 8. Profit or margin came next, named by 5. Nobody listed more than four numbers. Every list of six or more came from a vendor, an agency or an account that names no store.

The short list is not laziness. A KPI earns its place only if you would act when it moves. If you cannot name the action, it is a metric you can look up later.

The pages Google ranks first agree on five core KPIs for ecommerce, and then they split. Here is how often each KPI appears across the 18 pages I read.

The gap at the bottom matters most. The top-ranking pages stop at gross margin, before the ads are paid. Yet 5 of the 32 owners named profit or margin as a number they watch.

So here is my short list, in two steps. Start with four KPIs for ecommerce, close to the ones owners name themselves:

1. Net sales: what you sold, after returns and coupons.
2. Profit after ad spend: sales minus product cost, shipping, fees and ad spend.
3. Conversion rate: the share of visits that end in an order.
4. Average order value: net sales divided by orders.

Those four tell you whether you sold, whether you kept money, whether visits turned into orders and how big each order was. Once they are steady, add three that explain why they move:

1. Acquisition cost against lifetime value: what a new customer costs, and what they are worth.
2. Repeat purchase rate, with returns: whether customers come back and keep what they buy.
3. Conversations per 100 orders: how many questions each order creates.

Conversion rate is orders divided by visits, times 100. If your store had 20,000 sessions last month and 300 orders, your conversion rate was 1.5%.

The catch is the word “visits”. Google Analytics 4 gives you two different rates. The session key event rate is “number of sessions in which a key event happened / total number of sessions”. The user key event rate divides by people instead.^[4] One shopper who visits three times is three sessions but one user. So the user rate usually reads higher.

A “key event” is GA4’s name for what it used to call a conversion.^[5] Google marks purchase as a key event by default.^[6] If someone also marked “add to cart” as a key event, your key event rate stops being a purchase rate. Check that before you trust it.

Also, “Session and User metrics are calculated through an estimation”.^[7] GA4 estimates your visits. Your orders, in WooCommerce, are exact.

Littledata, which sells analytics tracking for Shopify stores, publishes one of the clearest benchmarks. Across 426 Shopify stores, the median converts 1.4% of sessions into orders. Above 2.7% puts a store in the top 20%.^[8] The window was 6 July to 3 October 2026, and sectors differ widely: 2.8% in beauty, 0.9% in home and furniture.

Returning visitors convert better. Contentsquare, an analytics vendor, measured 2.9% for returning visitors against 1.7% for new ones, across industries, from late 2024 to late 2025.^[9] So a month with more first-time traffic can lower your rate without anything going wrong.

WooCommerce defines three sales figures (gross, net and total), and its average order value is net sales divided by orders.

Sales is the number owners name first. One owner’s daily view: “Daily sales, week to date, month to date, last 7 days , last 14 days. Sales sales sales sales only sales”.^[1] Yet the top-ranking pages barely define it. WooCommerce does, in four lines.^[10]

So WooCommerce’s average order value leaves out tax and shipping. It is also net of refunds and coupons. If net sales were $60,000 from 520 orders, your average order value was about $115.

Which orders count matters too. By default, “Pending payment, Cancelled and Failed order statuses are excluded while Processing, On hold, and Completed order statuses are included”. And “The Refunded status cannot be excluded”.^[10] WooCommerce also dates each order by when it was paid, unless you change the setting.

Littledata’s median is $107 across the same 426 Shopify stores.^[11] Its figure is purchase revenue from GA4 divided by orders, while WooCommerce’s starts from net sales. Check whether both count tax and shipping the same way before you compare.

Your own last 90 days beat any benchmark. A median of other people’s stores tells you little about your products and prices.

Cart abandonment is the share of shoppers who add something to the cart and do not buy. Littledata defines it per session: “Of the sessions that added to cart, the share that did not purchase”.^[12] Across 423 Shopify stores, the median was 79%.

The figure you see quoted most is 70.22%, from the Baymard Institute. It is “an average calculated based on 50 different studies”, with different stores, years and definitions.^[13] Both numbers are fair. They just measure different things, so say which one you mean.

Checkout is a separate step, and worth its own number. Littledata’s median checkout completion was 52% across 425 stores: about half of the checkouts started did not finish.^[14] A leak in the cart and a leak in the checkout have different causes.

A high rate is partly normal. Baymard found that 42% of US online shoppers had abandoned a cart because “I was just browsing / not ready to buy”.^[13] Some of your carts were never going to be orders.

WooCommerce does not report this rate. Measuring it needs events from GA4, such as add to cart and checkout started. The Google Analytics for WooCommerce extension can send both.^[15]

What to do when it rises: look at what shoppers needed to know before they left. A shopper who left no email or phone number cannot be reached later. The question has to be answered while they are still on the page.

Two KPIs for ecommerce tell you whether growth pays. Customer acquisition cost, or CAC, is your marketing spend divided by the new customers it brought. Harvard Business School Online names two numbers: what “your business spent on marketing” and “how many new customers you acquired”.^[16]

Lifetime value, or LTV, is what a customer is worth to you over the whole relationship. HBS Online writes it as m times T. Here m is the yearly revenue from an average customer, minus what it costs to serve them. T is how many years they stay.^[16] Note that it uses margin, not revenue.

What ratio is healthy? HBS Online says the benchmark “is generally considered at least three”. Below one, “you’re losing money”.^[16] Treat three as a rule of thumb. The page’s worked example is a subscription software company, not a store.

Acquisition is also getting dearer. Contentsquare reports that “the cost of a visit rose +9% YoY (up +30% over the past 3 years)”.^[17] That is the cost of a visit, not of a customer. But if your conversion rate is flat, a dearer visit means a dearer customer.

In WooCommerce, the Orders report’s Customer Type filter has two options, New and Returning. Use it to count new customers in a period.^[18] Your ad spend comes from the ad platforms and your other marketing bills.

Profit after ad spend is what each sale leaves once its variable costs and the ads are paid. Its formal name is contribution margin after ad spend. Not one of the 18 top-ranking pages on KPIs for ecommerce lists contribution margin. Owners ask for it anyway. One owner who runs several stores wrote: “Personally, the only main kpi I care about is profit”.^[19]

Contribution margin is what each sale leaves after its variable costs. The US Small Business Administration gives the formula: price minus variable cost per unit, divided by price.^[20] For a store, variable costs are the product, shipping, packaging and payment fees.

Then subtract the ads. Return on ad spend, or ROAS, is revenue from ads divided by ad spend. A blended version, sometimes called MER, divides all your revenue by all your ad spend. Neither one is profit. Another owner asked the right question: “But wouldn’t Profit from ad spend be a better metric?”^[21]

Profit on paper rarely matches the bank at first. One owner wrote that “my final profit never quite matches what’s in my bank account”. The gap, they added, is usually 10% to 20%.^[22] Look first at fees, refunds and payout timing, and reconcile against the bank once a month.

The next two KPIs for ecommerce show what happens after the first order. Repeat purchase rate is the share of customers who have ordered more than once. Divide the customers with two or more orders by all your customers. Use a long window, such as the last 12 months, or the rate will look low.

WooCommerce can give you both counts. Its Customers report has a “Number of orders” filter. It also joins guest orders by email. “When a guest places multiple orders with the same email address, WooCommerce associates those orders with the same customer record”.^[23] The summary under the table shows average orders and average lifetime spend.

I found no neutral benchmark for this one. The only figure I found came from an agency quoting its own clients. So start from your own number. One owner wrote: “Not tracking this yet, 75% of my sales are one time only, so I’m researching the other 25%”.^[24]

Return rate is returns divided by gross sales, and both sit in WooCommerce’s Revenue report. Mind the dates. “Refunds appear as a negative number on the date that the return occurred (not the date the order was placed)”.^[10] So a busy month’s returns can land in the next month.

The National Retail Federation estimated that 19.3% of online sales would be returned in 2025.^[25] That is a forecast from US retailers with over $500 million in revenue, not stores your size.

Most lists of KPIs for ecommerce skip this one. Every order brings questions: before the sale, about delivery and about returns. Count them. Add up the emails, chats and calls your store answered in a month. Divide by that month’s orders and multiply by 100.

The result is conversations per 100 orders. Because it is per 100 orders, a busy month and a quiet month compare fairly. Gorgias, which sells a help desk, publishes it by category. “An Electronics brand generates about 46 support tickets for every 100 orders”, it writes. A food and beverage brand creates about 20.^[26] Those figures come from brands at about $10 million a year in sales.

Some of those conversations are sales in progress. Gorgias reports that “1 in 9 support inquiries is a pre-purchase question”.^[27] That is a shopper deciding whether to buy.

Track your median first reply time as well, from your own inbox. Gorgias publishes a median for large brands and then writes: “That number describes no one”.^[26] Your own trend is the useful figure.

Do not expect a neat link to sales. Gorgias says plainly that a conversion curve by response time “is not derivable from platform data without circularity”.^[27] Our guide to customer support for ecommerce covers what else to measure. If the count keeps rising, read our guide on whether to outsource customer service.

No page in Google’s top 20 says where a WooCommerce owner finds these KPIs for ecommerce. Here they are, with the source of truth for each.

None of WooCommerce’s reports counts visits. Its order attribution “cannot be used to track your visitors across sessions”, and it uses “a last-click attribution model”.^[28] So visits come from GA4 or another analytics tool.

This is the complaint owners wrote most about. Nine of the 32 said a number they relied on was wrong, or that they could not get a reliable one. One saw “26 new users” from paid social in GA4. “Yet I’ve had 155 clicks on meta ads”.^[29]

Some of the gap is by design. WooCommerce warns that GA4 needs consent, so “you may notice a discrepancy between your store and Google Analytics/Ads data”.^[15] GA4 also records a purchase only when the payment gateway sends the shopper back to the thank-you page. GA4 will usually show less revenue than WooCommerce. That is expected.

WooCommerce has its own delay. On the scheduled setting, which WooCommerce recommends for most stores, it “Automatically updates analytics data every 12 hours”.^[10] Today’s figure may not be final yet.

My recommendation is to check net sales daily, most other KPIs for ecommerce weekly or monthly, and lifetime value quarterly. The top-ranking pages disagree on this. Four of them give a schedule, and one says to check conversion rate daily, another weekly, another monthly. Of the 8 owners who said how often they look, the two who check daily look at sales or profit. Two more had stopped checking daily.

So the schedule below is my recommendation, not a finding. It suits a store without an analyst.

A daily look at the other numbers mostly shows noise. One owner stopped checking daily because the platform’s numbers were “wildly inaccurate”.^[1] A weekly or monthly view smooths out the swings.

Write down what you changed and when. A KPI only teaches you something if you know what happened before it moved. As one owner who has run a store for eight years put it: “But fundamentally you need to understand your numbers”.^[30]