Twelve metrics are enough to run an online store: definition, formula, worked example and common misreading for each KPI.
Twelve metrics are enough to run an online store. Sessions, conversion rate and cart abandonment show the path from visit to purchase. Net revenue, average order value and revenue per session show what a visit is worth. ROAS, customer acquisition cost and contribution margin per order show whether advertising pays for itself. Return rate, repeat purchase rate and customer lifetime value show whether customers stay.
None of these numbers means much on its own. A high conversion rate with a falling contribution margin is a problem, not a win. Read metrics in pairs and always over time. The most useful comparison is your own previous month and the same month last year, not an average from the internet.
The numbers must be collected cleanly before you act on them. Shopware, Shopify and Google Analytics calculate the same metric in different ways. Pick one source and one formula per metric. If you would rather not run the store yourself: with the solution25 shop subscription (Scale 299 euros per month, Pro 499 euros per month, Enterprise on request, 24-month minimum term each, no setup fee) our team builds and operates the store, with a monthly report in the Pro package.
This article is for managing directors, ecommerce leads and marketing leads who want to read their own store numbers and base decisions on them. For each metric you get the definition, the formula, a worked example, what it is good for and the most common misreading. All worked examples are assumptions based on an imaginary store. They are not industry figures.
A metric is any number your store produces. A KPI (key performance indicator) is a metric you tie a decision to. The difference is practical: page views are a metric. Contribution margin per order is a KPI, because it decides how much you can spend to win a new customer.
| Metric | Formula | Question it answers |
|---|---|---|
| Sessions | Number of visits in the period | Are enough visitors coming? |
| Conversion rate | Orders ÷ sessions × 100 | How many visits end in a purchase? |
| Cart abandonment rate | (1 − orders ÷ sessions with a cart) × 100 | Where do buyers drop out? |
| Net revenue | Gross revenue − sales tax or VAT − discounts − refunds | What is really left from sales? |
| Average order value | Net revenue ÷ orders | How big is a cart? |
| Revenue per session | Net revenue ÷ sessions | What is a visit worth? |
| ROAS | Revenue from ads ÷ ad spend × 100 | What does one ad euro return in revenue? |
| Customer acquisition cost | Marketing cost ÷ new customers | What does a new customer cost? |
| Contribution margin per order | Order value − variable costs | Does an order make money? |
| Return rate | Returned orders ÷ orders × 100 | How much revenue comes back? |
| Repeat purchase rate | Returning customers ÷ all customers × 100 | Do customers come back? |
| Customer lifetime value (CLV) | Contribution margin per order × orders per customer | What is a customer worth over time? |
The order follows the money: first visits, then purchases, then costs, then retention. Search metrics such as clicks, impressions and rankings are left out on purpose. They are covered in the article on the SEO roadmap for online stores.
Definition: A session is one continuous visit. Google Analytics describes it as a group of user interactions that take place within a given time frame. By default a session there ends after 30 minutes without activity. A person who drops by in the morning and again in the evening creates two sessions.
What it is good for: Sessions are the denominator for almost every other metric. Split by channel (search, ads, newsletter, direct) they show where demand comes from.
Common misreading: More sessions are not automatically better. 10,000 extra visits from a broadly targeted campaign can push the conversion rate down without adding a single order. Also, only visitors who consent to tracking are counted. Shopify states this explicitly for sessions in its help center. A new cookie banner can therefore change the number while visitor behavior stays the same.
Definition: The share of sessions that end with an order. Shopify divides sessions that completed checkout by sessions. Google Analytics calls the metric session key event rate, the percentage of sessions with a key event. A key event there is an action you mark as important to your business. In a store that is the purchase.
Formula: Orders ÷ sessions × 100. Worked example (assumption): 800 orders from 40,000 sessions give 2.0 percent.
What it is good for: It measures how well the store turns visits into purchases. Split by device and channel it shows where the store loses buyers.
Common misreading: The conversion rate across all channels is a blend. Regular customers from the newsletter buy more often than first-time visitors from an ad. When the mix shifts, the rate changes even though the store is the same. How to improve the rate is covered in the article on conversion optimization for online stores.
Definition: The share of sessions with a filled cart that end without an order. Google Analytics counts the events add_to_cart and begin_checkout for this. Shopify reports the share of sessions with cart additions and the share that reached checkout.
Formula: (1 − orders ÷ sessions with a cart) × 100. Worked example (assumption): 3,200 sessions with a cart and 800 orders give 75 percent abandonment.
What it is good for: It shows whether the problem sits before the cart (range, price, product page) or after it (shipping costs, payment methods, required fields). Split the path into steps: cart, checkout started, order.
Common misreading: Many visitors use the cart as a wish list. A high value is therefore normal and not an alarm by itself. The change is what counts: if abandonment rises after a change to shipping costs or payment methods, you have found the cause.
Definition: The revenue left after sales tax or VAT, discounts and refunds. Shopify separates gross sales (before discounts and sales reversals) from net sales (gross sales minus discounts minus sales reversals). Google Analytics subtracts refunds from purchase revenue.
Formula: Gross revenue − sales tax or VAT − discounts − refunds. Worked example (assumption): 800 orders bring 64,000 euros of net revenue after these deductions.
What it is good for: It is the basis for order value, contribution margin and customer lifetime value. If you calculate with gross figures here, every metric that follows looks better than it is.
Common misreading: The revenue on the store dashboard is usually order intake. According to the documentation, the dashboard in the Shopware administration shows two statistics, one on orders and one on sales. Cancellations and returns that arrive weeks later are not included in the month of the order. Check the monthly figure against your accounting before you pass it on.
Definition: The mean value of an order, often shortened to AOV. Shopify calculates gross sales minus discounts, divided by the number of orders. Shopware Analytics lists the metric as average order value.
Formula: Net revenue ÷ orders. Worked example (assumption): 64,000 euros ÷ 800 orders = 80 euros.
What it is good for: It shows whether tiered prices, bundles, accessory recommendations or a free shipping threshold work. Together with the conversion rate it explains revenue completely.
Common misreading: An average hides the distribution. A few large orders from business customers lift the value while the typical cart shrinks. Look at the median as well, or separate customer groups. And a higher order value that comes from discount campaigns can lower the contribution margin.
Definition: The net revenue an average visit brings. Mathematically it is conversion rate times average order value.
Formula: Net revenue ÷ sessions. Worked example (assumption): 64,000 euros ÷ 40,000 sessions = 1.60 euros.
What it is good for: It is the most honest number for tests on the store. A change that lifts the conversion rate and lowers the order value may bring nothing in the end. Revenue per session shows that in one number. It also tells you the most a click may cost.
Common misreading: The value is revenue, not profit. If you take in 1.60 euros per session and pay 1.20 euros per click, nothing is left after cost of goods and shipping.
Definition: ROAS stands for return on advertising spend, meaning revenue per ad euro. Google Ads Help calculates conversion value divided by cost, times 100. The example there: 5 euros of revenue for 1 euro of ad spend gives 500 percent.
Formula: Revenue from ads ÷ ad spend × 100. Worked example (assumption): 32,000 euros of ad revenue for 8,000 euros of ad spend give 400 percent. Read the other way round, advertising takes this share of the revenue it brings: 8,000 ÷ 32,000 = 25 percent.
What it is good for: It compares campaigns and channels with each other and shows where the next budget should go.
Common misreading: ROAS measures revenue, not profit. Whether 400 percent is enough depends on your margin. With a margin of 36.25 percent before advertising (29 of 80 euros) the break-even point is about 276 percent (1 ÷ 0.3625). Below that, every order costs money. Attribution adds to this: ad platforms claim purchases that would have happened without an ad, for example searches for your brand name.
Definition: What you spend to win one new customer, often shortened to CAC.
Formula: Marketing cost ÷ new customers. Worked example (assumption): 8,000 euros of ad spend and 500 new customers give 16 euros per new customer.
What it is good for: It sets the upper limit for your ad budget. Together with customer lifetime value it shows whether growth pays off.
Common misreading: The numerator often holds click costs only. The cost also includes the agency, tools, vouchers for first orders and working time in your own team. Decide what counts and stick to it. Second mistake: putting all orders in the denominator instead of new customers only. Then the cost seems to fall because regular customers are counted too.
Definition: What is left of an order after all costs that arise only because of this order.
Formula: Net order value − cost of goods − shipping and packaging − payment fees − share of return costs − ad cost per order. Worked example (assumption): 80 euros − 40 euros − 6 euros − 2 euros − 3 euros = 29 euros before advertising. After 10 euros of ad cost per order (8,000 euros ÷ 800), 19 euros remain.
What it is good for: It is the number that decides. Rent, salaries, store operation and profit must be paid from it. Every other metric in this article is a lever on it.
Common misreading: The store and the analytics tool do not know your costs. This metric is therefore almost always built in a spreadsheet or in the ERP system, not on the dashboard. Many retailers never calculate it and steer by revenue. Payment fees differ noticeably between providers. The comparison of payment providers for Shopware helps with the calculation.
Definition: The share of orders that come back in full or in part. It can be calculated by orders, by items or by value. The three results differ.
Formula: Returned orders ÷ orders × 100. Worked example (assumption): 96 of 800 orders come back. That is 12 percent.
What it is good for: It shows how much revenue does not stay and which products, sizes or channels cause that.
Common misreading: Returns arrive with a delay. If you divide the returns of a month by the orders of the same month, a strong previous month gives you a value that is too high. Calculate by order month. Causes, costs and countermeasures are covered in the article on reducing the return rate.
Definition: The share of customers in the period who have ordered before. Shopify divides returning customers by all customers who placed orders. Shopware Analytics reports customers and new customers separately, and the share can be derived from that.
Formula: Returning customers ÷ all customers × 100. Worked example (assumption): 750 customers order in the month, 250 of them not for the first time. That is 33.3 percent.
What it is good for: It shows whether product, delivery and service convinced people. Regular customers cost nothing to acquire, so every repeat purchase lifts the contribution margin.
Common misreading: The rate depends on the period and the product range. A store for mattresses has a different repeat purchase rate than a store for coffee without being worse. It also drops when you win many new customers, which is then a good sign. Guest orders without a customer account distort the count when the same person shows up as a new customer.
Definition: The contribution margin a customer brings over the whole relationship. Google Analytics lists a related metric as lifetime value, but there it is revenue from purchases, not contribution margin.
Formula (simple): Contribution margin per order before advertising × orders per customer in the period you look at. Worked example (assumption): 29 euros × 2.5 orders in 24 months = 72.50 euros.
What it is good for: It answers how much a new customer may cost. In the example, 72.50 euros of lifetime value stand against 16 euros of acquisition cost. The store could invest more in new customers, as long as the repeat purchases really happen.
Common misreading: Lifetime value is an estimate about the future. If you calculate it with revenue instead of contribution margin, or assume ten years, you can justify any ad budget. Use a period you can back up with your own data, such as 12 or 24 months. One way to more repeat purchases is the newsletter in Shopware.
Three systems deliver metrics, and all three are right, just for different questions.
The Shopware administration shows how orders and sales have developed on the dashboard. The Shopware Analytics app adds more under Dashboard > Analytics. The documentation lists, among others, total revenue, orders, average order value, customers, new customers, payment methods, shipping methods, discounts and promotions, product sales and revenue by sales channel. For the storefront it adds page views, unique visitors, conversion rate and the most popular search terms.
Three points from the documentation are worth knowing:
Shopify documents its formulas openly: conversion rate as sessions that completed checkout divided by sessions, average order value as gross sales minus discounts divided by orders, returning customer rate as returning customers divided by customers. Shopify counts sessions only when visitors consent in the cookie banner.
Google Analytics 4 works with events. For a store these include view_item, add_to_cart, begin_checkout, purchase and refund. If one of these events is missing, or the purchase is sent twice, every derived metric is wrong. Purchase revenue there is the sum of purchases minus refunds, provided the refund event is sent as well.
Set exactly one source per metric. Revenue and orders from the store, sessions and channels from the analytics tool, costs from accounting. A gap between store and analytics tool is to be expected. What matters is that it stays stable. If it jumps, the cause is usually tracking or the cookie banner, not the market. The legal side of tracking under German and EU law is covered in the article on data protection in online stores.
The following example is an assumption. The store is imaginary and the numbers are chosen so that the math is easy. They are not reference values for your industry.
Starting data for one month: 40,000 sessions, 3,200 of them with a cart. 800 orders from 750 customers, including 500 new customers. 64,000 euros of net revenue. 8,000 euros of ad spend, 32,000 euros of revenue from ads. 96 orders come back. Per order: 40 euros cost of goods, 6 euros shipping and packaging, 2 euros payment fee, 3 euros share of return costs.
| Metric | Calculation | Result |
|---|---|---|
| Sessions | counted | 40,000 |
| Conversion rate | 800 ÷ 40,000 | 2.0% |
| Cart abandonment rate | 1 − 800 ÷ 3,200 | 75% |
| Net revenue | from the store, checked against accounting | €64,000 |
| Average order value | 64,000 ÷ 800 | €80 |
| Revenue per session | 64,000 ÷ 40,000 | €1.60 |
| ROAS | 32,000 ÷ 8,000 | 400% |
| Customer acquisition cost | 8,000 ÷ 500 | €16 |
| Contribution margin per order | 80 − 40 − 6 − 2 − 3 − 10 | €19 |
| Return rate | 96 ÷ 800 | 12% |
| Repeat purchase rate | 250 ÷ 750 | 33.3% |
| Customer lifetime value (24 months) | 29 × 2.5 | €72.50 |
The store earns 19 euros per order, so 15,200 euros of contribution margin per month (19 × 800). Fixed costs come off that. Three levers are visible:
The calculation shows what the twelve numbers are for: you can compare measures in euros before you spend money on them. Which measure really works only shows in a test in your own store.
The most common question about metrics is: is my value good? The web is full of averages, for conversion rate or cart abandonment for example. We deliberately quote none in this article. The reasons:
If you do use an external value, check the source, the survey year, the country, the industry and the formula. If one of these is missing, the value is not fit to be a target.
For retailers in the United States: Calculate every metric with revenue excluding sales tax. Sales tax differs by state and is usually handled through a tax service integration, so take the tax amount from there and not from a flat rate. All amounts in this article are in euros, including the prices of the shop subscription. Most Shopware partners, solution25 included, are based in Europe. Meetings usually take place in the morning, Eastern Time.
Looking at twelve metrics every day leads to busywork. Daily values fluctuate at random. A split by pace works well:
| Rhythm | Metrics | Purpose |
|---|---|---|
| Daily, two minutes | Orders, revenue | Spot faults: checkout, payment method, tracking |
| Weekly | Sessions by channel, conversion rate, cart abandonment, ROAS | Adjust campaigns and store |
| Monthly | Net revenue, order value, revenue per session, acquisition cost, contribution margin, return rate | Set budget and priorities |
| Quarterly | Repeat purchase rate, customer lifetime value | Review strategy: range, retention, channels |
A usable monthly report fits on one page: twelve rows, each with the value, the previous month, the same month last year and one sentence on the cause. The sentence is the important part. A number without an explanation leads to no decision.
Metrics need operation. After every update someone has to check that tracking still runs, that the cookie banner is integrated correctly and that test orders are filtered out. In small teams this often waits until a number looks odd. Loading time belongs here too, because a slow store pushes the conversion rate down. More on that in the article on optimizing Shopware 6 performance.
If you want to hand over build and operation, the solution25 shop subscription is one option. Our team builds and operates the store, with Shopware, WooCommerce or Shopify. Scale costs 299 euros per month, Pro 499 euros per month, and Enterprise is available on request. The minimum term is 24 months in each case and there is no setup fee. According to the package description, Pro includes a dedicated contact person and a monthly report. Ad spend and the assessment of your margin stay with you in any case, because only your accounting knows the cost side.
Contribution margin per order. It shows whether an order makes money after cost of goods, shipping, payment fees, returns and advertising. All other metrics explain why it rises or falls.
There is no general answer. The rate depends on product range, price, channel mix, consent rate and calculation method. A usable yardstick is your own value in the previous month and in the same month last year, split by device and channel.
The store counts every order. Google Analytics only counts visitors who consented to tracking, and only purchases where the purchase event arrives. A gap is therefore normal. It only becomes a warning sign when it changes suddenly.
Not at the start. A spreadsheet with twelve rows, updated once a month, is enough. A tool pays off when several people need the same numbers regularly or when several sales channels come together.
Orders and revenue daily, to spot faults. Conversion rate, cart abandonment and ROAS weekly. Net revenue, order value, acquisition cost, contribution margin and return rate monthly. Repeat purchase rate and customer lifetime value once a quarter.
solution25 is a Shopware Platinum Partner from Bremen, Germany, and has worked on online stores since 2017. With metrics we can help in three places, and in one we cannot.
What we cannot do: know your margin. Cost of goods, shipping costs and return costs come from your accounting. Without these figures every analysis stops at revenue. We cannot promise specific results, because metrics show problems. They do not solve them.
If you would like to go through your numbers together, write to us using the contact form. The values of the last three months and a short description of how you measure today are helpful.

Besar Bilalli is co-founder and Co-CEO of solution25, the Shopware agency based in Bremen. Before founding the agency in 2017, he and Besnik Bilalli built their own online store without any startup capital and grew it to more than 500 orders a day. Since then, he has guided merchants through Shopware projects, from platform selection to day-to-day operations. He’s happy to answer questions about this article personally: get in touch.
Responsible for content and updates: Besar Bilalli. Last updated: November 2026. Next content review: February 2027.