Ecommerce KPIs Explained: The 12 Metrics That Actually Drive Growth

Every ecommerce KPI in plain English — CVR, AOV, LTV, CAC, contribution margin and more. Benchmarks, formulas, and how to act on each one.

Open any ecommerce analytics dashboard and you'll see a wall of acronyms: CVR, AOV, LTV, CAC, ROAS, MRR, CLV, GMV. To a new store owner it reads like a secret handshake. To a tired operator it reads like noise. But strip out the jargon and ecommerce KPIs are just twelve numbers, each answering one specific question about whether your store is healthy and whether it will be healthy six months from now.

This guide walks through every KPI that actually matters for an online store in 2026 — what it means in plain English, the formula, what a good value looks like, and crucially what to do when the number goes the wrong way. By the end you'll be able to read any ecommerce report and know exactly which lever to pull.

How to think about ecommerce KPIs

Before the list, a framework. Every ecommerce KPI fits into one of four buckets, and a healthy store has at least one metric tracked in each bucket. Stores that obsess over a single bucket — usually acquisition — fail because the others quietly rot.

  • Acquisition: how efficiently you bring new visitors and customers (sessions, CAC, ROAS, channel mix).
  • Conversion: how well your site turns visitors into customers (CVR, add-to-cart rate, checkout completion rate).
  • Value: how much each customer is worth (AOV, LTV, repeat rate, gross margin).
  • Profitability: whether the whole machine actually makes money (contribution margin, payback period, blended ROAS).

If you only track acquisition, you'll scale a money-losing business. If you only track conversion, you'll improve a site nobody visits. Balance is the point.

1. Conversion Rate (CVR)

The most quoted ecommerce KPI. Out of every 100 visitors, how many buy?

CVR = (Orders / Sessions) × 100

Benchmarks: 1.5-2.5% is typical, 3%+ is strong, below 1% almost always indicates a problem with tracking, traffic quality, or the site itself. Always segment by device — mobile CVR is typically 40% lower than desktop, and a sudden drop in mobile-only CVR is usually a checkout regression.

Levers: page speed, product photography, social proof, simplified checkout, payment method variety, mobile UX.

1a. Funnel-stage conversion rates

The single CVR number hides where revenue is leaking. Decompose into: view-product rate (homepage → product page), add-to-cart rate (product → cart), checkout-start rate (cart → checkout), checkout-completion rate (checkout → purchase). When CVR drops, the funnel tells you which step broke.

2. Average Order Value (AOV)

AOV = Revenue / Orders

How much the average customer spends in a single transaction. The most under-leveraged KPI in ecommerce — a 15% increase in AOV often produces more profit than a 15% increase in traffic because you don't pay acquisition costs twice.

Benchmarks vary wildly by category. Apparel: €60-€120. Beauty: €40-€80. Furniture: €300-€800. Compare AOV to your category and to your own trend, not across categories.

Levers: bundles, free-shipping thresholds set 10-20% above current AOV, post-add-to-cart upsells, volume discounts, complementary product recommendations.

3. Revenue Per Visitor (RPV)

RPV = Revenue / Sessions = CVR × AOV

The composite metric that captures both conversion and value in a single number. Underused because it's not as instantly readable, but it's the single best metric to optimize against — improving either CVR or AOV improves RPV, so it can't be gamed.

Benchmarks: €0.50-€2.00 for general ecommerce, €3-€8 for premium or considered purchases. The metric to obsess over if you only pick one.

4. Customer Acquisition Cost (CAC)

CAC = Total marketing spend / New customers acquired

How much you pay to bring in one new customer. The number that decides whether scaling is profitable. Compute it two ways: blended (all marketing spend, all new customers) for what your P&L sees, and channel-level for tactical decisions.

Don't make the rookie mistake of including only ad spend. True CAC includes media, agency fees, software, affiliate commissions, content production and a fair allocation of marketing salaries. Stores that report CAC on ad spend alone routinely understate it by 50%.

There's no universal 'good' CAC — it only makes sense relative to LTV (next KPI).

5. Customer Lifetime Value (LTV / CLV)

LTV = (Average order value × Orders per year × Gross margin %) × Average customer lifespan in years

The total profit a customer generates across their relationship with your store. Critical because it tells you how much you can afford to spend acquiring them. For most ecommerce businesses, calculate 12-month and 24-month LTV — anything longer is too speculative for decision-making.

Benchmarks: a healthy LTV/CAC ratio is at least 3 (you generate €3 of customer profit for every €1 spent acquiring them). Below 3, you're working hard for thin margins. Above 5, you're under-investing in growth.

6. CAC Payback Period

How many months it takes to recover the cost of acquiring a customer through their purchases. Under 6 months is excellent for direct-to-consumer ecommerce; 6-12 months is acceptable; over 12 months means you need outside capital to grow because you can't fund acquisition from cash flow.

Payback dominates LTV/CAC for cash-constrained businesses. A 5x LTV/CAC ratio means nothing if the payback period is 18 months and you run out of cash at month 12.

7. Repeat Purchase Rate

Repeat Rate = Customers with 2+ orders / Total customers

The percentage of customers who come back. The single best long-term predictor of store health. A store at 35% repeat rate compounds; a store at 12% is on a treadmill of constant new acquisition.

Benchmarks: 20-30% is typical, 35%+ is strong. Below 15%, either your product isn't repeat-purchase by nature (furniture, mattresses) or there's a retention problem worth investigating.

Levers: email and SMS flows, loyalty programs, subscription options, post-purchase experience, product quality, packaging.

8. Cart Abandonment Rate

Cart Abandonment = 1 − (Completed checkouts / Started checkouts)

The industry average is 70% — meaning seven out of ten customers who add to cart never complete the purchase. Under 60% is excellent; above 80% indicates a serious checkout problem worth fixing this week.

Top causes (in order): unexpected shipping costs at checkout, forced account creation, slow page load, limited payment options, security concerns, complicated checkout forms. Abandoned-cart email flows recover 5-15% of these regardless.

9. ROAS (Return on Ad Spend)

ROAS = Revenue from ads / Ad spend

Revenue generated per euro of ad spend. The tactical efficiency metric for paid acquisition. Expressed as a ratio (4x means €4 revenue per €1 spent).

Critical caveat: ROAS reported by Meta and Google is now systematically inflated by 30-100% due to modeled conversions. Trust your GA4 ROAS or server-side attribution, not platform-reported numbers, when making budget decisions.

And ROAS is not profit. A 4x ROAS on a 30% margin product nets roughly zero. Always pair ROAS with contribution margin (KPI #11) before scaling spend.

10. Gross Margin

Gross Margin = (Revenue − COGS) / Revenue × 100

What's left of each euro of revenue after the cost of the product itself (manufacturing, materials, direct labor). The fundamental constraint on every other ecommerce decision.

Benchmarks: 50%+ for apparel and accessories, 30-50% for general retail, 60-80% for digital products or premium brands, 15-30% for low-margin verticals like consumer electronics. If you don't know your gross margin to the percentage point, every other KPI in this list is theoretical.

11. Contribution Margin

Contribution Margin = Revenue − COGS − Variable costs (payment fees, shipping, fulfilment, marketing)

The most honest profitability metric for ecommerce. What's left after every variable cost — not just the product but the fees, shipping, fulfilment and marketing it took to deliver it.

Targets: 20-30%+ at the contribution-margin level means a healthy business. Single-digit contribution margin means you're operating a logistics company that happens to sell products and one bad month wipes out the year.

12. Traffic Mix and Channel Concentration

Not a single number but a structural KPI: what share of your revenue comes from each acquisition channel. A store with 70% of revenue from one paid campaign is structurally fragile — one platform algorithm change or one CPM spike and the business is in trouble.

Healthy mix: no single channel above 40-50% of revenue, organic + email + direct combined above 40% (these are durable, not dependent on rented audiences), at least three channels contributing meaningfully.

Diagnostic patterns: what combinations of KPIs reveal

Individual KPIs are signals. Patterns across KPIs are diagnoses.

  • Rising sessions, flat conversions: site problem (CVR is failing under load, or traffic quality declined).
  • Falling AOV with rising orders: promotional dependence — customers buying smaller orders chasing discounts.
  • Rising CAC with stable ROAS: ad platforms inflating attributed revenue while real CAC climbs. Run incrementality.
  • High first-order CVR, low repeat rate: acquisition is fine, product or retention is the problem.
  • Strong LTV/CAC but long payback period: profitable in theory, cash-constrained in practice. Borrow or slow down.
  • Falling cart abandonment but falling revenue: checkout improved, but you've lost top-of-funnel traffic.

Which KPIs matter at which stage

Stage matters. A €10k/month store and a €10M/month store have different priorities.

Under €30k/month revenue: obsess over CVR, AOV, and gross margin. Acquisition channels are noisy at small scale. Get the fundamentals right before scaling spend.

€30k-€300k/month: add CAC, ROAS, repeat rate, and contribution margin. This is where most stores die — they scale acquisition without contribution-margin discipline.

Above €300k/month: full stack of 12 KPIs plus cohort-level LTV by channel, payback period, and traffic mix monitoring. At this scale, marginal decisions on channel allocation become the most valuable analytical work.

Cadence: how often to check each KPI

  • Daily: revenue, orders (operational visibility, not for decisions).
  • Weekly: CVR, AOV, RPV, blended ROAS, cart abandonment, top landing pages.
  • Monthly: CAC, LTV, repeat rate, channel mix, contribution margin.
  • Quarterly: payback period, cohort retention, LTV by channel, full P&L reconciliation.

Resist the urge to check monthly metrics weekly — short windows have too much variance to drive decisions on long-term metrics like LTV.

Tools that calculate these KPIs for you

Most of these KPIs can be calculated by hand from a Shopify export and GA4. The work is tedious — pulling, cleaning, joining, charting — but free. Tools shorten the loop.

  • Looker Studio: free, native GA4 connector. Best for visual dashboards. Plan a few hours of setup.
  • Shopify's built-in analytics: accurate for revenue, AOV, repeat rate; weak on attribution and channel KPIs.
  • Triple Whale, Polar Analytics, Daasity: paid ecommerce analytics platforms with these KPIs prebuilt. Pricing scales with revenue.
  • Metriko and similar AI tools: compute the same KPIs and deliver them as plain-English daily briefs without you building dashboards.

Metriko reads your Google Analytics and delivers a daily brief covering every KPI in this guide — CVR, AOV, CAC, ROAS, repeat rate — in plain English, with the action you should take.

Common KPI mistakes to avoid

  1. Comparing KPIs across categories. AOV benchmarks for apparel are useless for furniture.
  2. Trusting in-platform ROAS. Always cross-reference with GA4 and store data.
  3. Calculating LTV on small or new cohorts. Need 6-12 months of cohort data for meaningful LTV.
  4. Optimizing one KPI in isolation. Pushing CVR by 0.5% through aggressive discounting drops AOV and gross margin by more.
  5. Reporting revenue without margin. A 50% revenue increase at a 30% margin reduction is a loss disguised as growth.
  6. Ignoring mobile-specific KPIs. Mobile is 60-70% of sessions for most stores; tracking only blended numbers hides mobile-specific failures.

Keep reading

Want to go deeper? These related guides build on what you just read: [the complete ecommerce analytics guide](/guides/ecommerce-analytics), [how to increase ecommerce conversions](/guides/improve-ecommerce-conversion-rate), [how to measure marketing ROI](/guides/how-to-measure-marketing-roi), [the simple marketing dashboard](/guides/simple-marketing-dashboard).

FAQ

Which ecommerce KPI should I look at first?

Start with conversion rate (CVR) and average order value (AOV). Together they form revenue per visitor (RPV), which is the most actionable single number. Improving both by 20% each nearly doubles revenue at the same traffic level — without spending anything more on acquisition.

What's a good CVR for ecommerce in 2026?

1.5-2.5% is typical for general ecommerce, 3%+ is strong, and category leaders often see 4-5%. Below 1% indicates a tracking issue, a traffic quality problem (paid traffic going to the wrong page), or a real site problem worth investigating immediately.

How do I calculate LTV correctly?

Start with a 12-month window: average revenue per customer across all orders in their first 12 months, multiplied by gross margin %. For more accuracy, segment by acquisition channel — channels often differ by 2-3x in LTV even at the same CAC. Avoid lifetime windows longer than 24 months; the data gets too speculative.

What's a healthy LTV to CAC ratio?

At least 3:1 — meaning you generate €3 of customer profit for every €1 spent acquiring them. Below 3, you're working hard for thin margins. Above 5, you're likely under-investing in growth. Always pair the LTV/CAC ratio with payback period — a healthy ratio with an 18-month payback is still a cash-flow problem.

Should I track ROAS or CAC?

Both, for different purposes. ROAS is the tactical metric for daily campaign decisions. CAC is the strategic metric for monthly budget allocation and capital planning. ROAS without CAC leads to scaling unprofitable campaigns; CAC without ROAS leaves easy optimization on the table.