Ecommerce KPIs: 15 Metrics Every Ecommerce Manager Should Track
Most ecommerce KPIs get tracked because a dashboard offered them, not because anyone decided they were worth watching. The result is a weekly report nobody acts on: forty numbers, no hierarchy, and no rule for what to do when one of them moves.
This is the set I actually use, organised so that each number tells you where to look next.
The short answer
A metric earns its place only if a change in it would change what you do on Monday.
Everything else is reporting. The fifteen ecommerce KPIs below sit in four tiers, and the tiers matter more than the list: a business number never moves on its own, so when revenue drops the answer is always somewhere in a lower tier.
Why revenue alone is not enough
Revenue is an outcome. It tells you that something happened, never what or why, and by the time it moves the cause is weeks old.
Two stores can post identical revenue while one is compounding and the other is dying. The first grew repeat purchases and held acquisition cost flat. The second bought its way to the same number and will need a larger budget next month to stand still. Revenue cannot distinguish them. The ecommerce KPIs underneath it can.
Worse, revenue is not even profit. A store can grow revenue 30% while contribution falls, because the growth came from a discounted line that costs more to ship than it earns. That is not a hypothetical, it is one of the more common things an audit turns up.
The ecommerce KPI pyramid
Arrange the metrics so that causation runs upward. Business outcomes at the top, the behaviour that produces them underneath.
Read it downward when something breaks and upward when you are planning. That single habit removes most of the arguing from a performance review, because the question stops being whose fault it is and becomes which tier moved first.
Tier 1: business ecommerce KPIs
- 1. Revenue. The headline. Useful only alongside the four tiers below it, and misleading on its own.
- 2. Gross profit. Revenue minus cost of goods. The first number that survives a discount-driven growth story.
- 3. Contribution margin. Gross profit minus fulfilment, payment fees and variable marketing. This is the number that tells you whether an order was worth having, and most stores do not track it per product.
- 4. Customer lifetime value. Contribution from a customer across their whole relationship with you. It sets the ceiling on what you can afford to pay to acquire one.
If you track only one thing from this tier, track contribution margin by product line. It reorders merchandising decisions faster than any other single metric.
Tier 2: acquisition ecommerce KPIs
- 5. Customer acquisition cost. Total acquisition spend divided by new customers, not by orders. Mixing repeat orders into the denominator flatters it badly.
- 6. LTV to CAC ratio. The single healthiest indicator in ecommerce. Below roughly 3:1 you are buying growth rather than earning it.
- 7. Blended ROAS. Across all channels, not per platform. Platform-reported figures each claim the same conversions and sum to more revenue than you actually made.
- 8. Traffic by source, with conversion attached. Volume without the conversion rate beside it is a vanity number.
On attribution: if Meta, Google and your platform each report the same order, your channel-level ROAS figures will never reconcile with your bank account. Use blended numbers for decisions about total spend and platform numbers only for decisions inside that platform.
Tier 3: conversion ecommerce KPIs
- 9. Conversion rate, split. By device, country and source. The blended figure hides everything worth knowing, and mobile usually hides the most. The ecommerce conversion rate calculator will show you what one percent at each funnel step is worth on your own numbers.
- 10. Add-to-cart rate. Isolates the product page from the checkout. If this is healthy and orders are not, the problem is downstream.
- 11. Checkout completion rate. Of those who start checkout, how many finish. Baymard Institute puts the average cart abandonment rate at 70.22% across 50 studies, so a poor figure here is normal but rarely irreducible.
- 12. Average order value. Watch the trend, not the number. Flat AOV that only moves when you discount is a merchandising problem.
Tier 4: customer ecommerce KPIs
- 13. Repeat purchase rate. Share of customers who order again. The lever that decides whether you own customers or rent them monthly from ad platforms.
- 14. Time to second order. More actionable than repeat rate, because it tells you when to intervene rather than only whether you failed to.
- 15. Review and return rates. Product truth. A rising return rate on one line is a merchandising or expectation problem that will show up in contribution two months later.
If this KPI drops, check these first
This is what turns a list of ecommerce KPIs into a decision tool. When a number moves, work the row rather than opening every dashboard you own.
| When this drops | Check in this order |
|---|---|
| Conversion rate | Traffic quality by source → product page → mobile → checkout → pricing and delivery cost |
| Add-to-cart rate | Product page information → delivery timing → variant clarity → price presentation → images |
| Checkout completion | Where shipping cost appears → guest checkout → form length → payment methods → errors on mobile |
| Average order value | Product mix → merchandising and ranking → bundles → free-shipping threshold → discount depth |
| Repeat purchase rate | Delivery experience → post-purchase communication → product satisfaction → replenishment timing → lifecycle flows |
| Contribution margin | Product mix → discounting → shipping cost per order → payment fees → returns |
| Blended ROAS | Channel mix → creative fatigue → landing page match → conversion rate → competitor pricing |
Each row is ordered by how often the cause turns out to live there, and by how cheap it is to check. Start at the left.
Working out which of these ecommerce KPIs is your actual constraint, and what fixing it is worth in euros, is what a diagnostic does. A fixed-price audit starts at €400.
Which ecommerce KPIs matter at your stage
Tracking all fifteen from day one is a good way to track none of them properly.
| Stage | The three that matter | Why |
|---|---|---|
| Under €250k a year | Conversion rate, AOV, gross profit | Volume is too low for reliable cohort or channel analysis. Fix the storefront economics first. |
| €250k to €1M | CAC, contribution margin, repeat purchase rate | Growth becomes a question of whether each customer is worth what you paid for them. |
| €1M and above | LTV to CAC, contribution by channel, time to second order | Channel and product mix now decide profitability more than any single page does. |
Vanity metrics versus decision metrics
A metric is vanity when it can improve while the business gets worse. Six that appear on most dashboards and change almost nothing:
- Sessions without conversion attached. Cheap traffic raises it and lowers everything else.
- Platform-reported ROAS alone. Each platform claims the same order.
- Bounce rate on its own. Highly variable by page type and traffic source, and rarely actionable without segmentation.
- Email list size. Deliverability and revenue per recipient decide whether it is an asset or a cost.
- Social followers. Unless you can trace revenue to them.
- Average time on site. Often rises when people cannot find what they came for.
The test is simple: if this number halved tomorrow, would you do anything differently? If not, stop reporting it.
What a usable dashboard looks like
One screen, four blocks matching the pyramid, each showing the current figure, the same period last year, and the direction. No more than fifteen numbers total.
Year-on-year rather than month-on-month, because ecommerce is seasonal and month-on-month comparisons generate false alarms and false comfort in roughly equal measure. Add a plain-language note each week saying what changed and why. In six months that note is worth more than the dashboard.
Before trusting any of this: compare last month’s orders and revenue in your platform admin against your analytics. A few percent apart is normal. Fifteen percent apart means every ecommerce KPI above is being read from fiction, and reconciling the data becomes the first job.
If the numbers point at a constraint you cannot name, the five growth levers explain how they interact, the traffic without sales diagnostic narrows a conversion problem to a stage, and the audit guide covers what a structured review adds.