Ecommerce Metrics: The Complete Formula Guide

Why most metric guides steer you wrong

Most ecommerce metric lists hand you a wall of formulas and stop. They rarely tell you the one thing that decides whether a "good" number is actually good: your margin. A 4.0 return on ad spend looks great until you learn the product only keeps a quarter on the dollar after costs — then it is a loss.

So this guide is organized around profit, not vanity. We define every metric with an exact formula, walk a numeric example you can follow line by line, and then show the identities that tie the metrics together. Wherever a real-world benchmark appears, it links to its source. Wherever you see a worked number, it comes from arithmetic you can check.

Think of this page as the map. Each metric family links to a deeper article in the cluster when you want to go further on one number.

Start with per-order profit, not revenue

Before any ratio, you need the profit of one average order. Everything downstream — break-even, target ROAS, LTV — depends on it.

Say a print-on-demand shirt sells for a $40 order with $16 of product cost (the blank plus printing). Gross profit = $40 − $16 = $24, and gross margin = ($40 − $16) ÷ $40 × 100 = 60%. That is the profit after the product alone.

But an order costs more than the product. Subtract shipping, payment processing, and pick-and-pack — say $5 + $1.60 + $1.40 = $8 of other variable costs. Now contribution margin before ads = $24 − $8 = $16, a margin ratio of $16 ÷ $40 = 0.40.

Then subtract the ad spend that order carried. At a 4.0 ROAS the order absorbs $40 ÷ 4 = $10 of ads, so contribution margin after ads = $16 − $10 = $6, or $6 ÷ $40 = 0.15 of revenue. That $6 — not the $40 top line — is what the order truly leaves behind. Getting this stack right is the whole game, and it is exactly the "true per-order profit" that tools like PodVector exist to compute.

Acquisition and efficiency metrics

These answer one question: is your paid growth paying for itself?

ROAS (return on ad spend)

ROAS = revenue from ads ÷ ad spend. Say ads produced $40,000 ÷ $10,000 = 4.0. It is a revenue ratio, so it flatters you — it says nothing about margin.

Break-even ROAS is the ROAS at which margin dollars exactly cover the ad spend: break-even ROAS = 1 ÷ margin ratio. On a contribution margin of 0.40, that is 1 ÷ 0.40 = 2.5; on a thin 0.20 it climbs to 1 ÷ 0.20 = 5.0, so a "great-looking" 4.0 would lose money. The lower your margin, the higher the ROAS you must clear just to avoid bleeding — this is the most useful identity in paid media, and the break-even ROAS formula breaks it down in full.

Target ROAS is break-even plus a profit buffer. If you want to keep 0.15 of revenue on a 0.40 margin base, you need roughly 1 ÷ (0.40 − 0.15) = 4.0. You can size that number with a target ROAS calculator instead of guessing.

POAS (profit on ad spend)

POAS puts profit in the numerator instead of revenue: POAS = ROAS × margin ratio. On a 0.60 gross margin, 4.0 × 0.60 = 2.4. The corollary is elegant — POAS = 1 exactly when ROAS hits break-even. So POAS > 1 means the campaign truly profits; POAS < 1 means it loses no matter how healthy the ROAS looks. Same 4.0 ROAS, but 4.0 × 0.60 = 2.4 is healthy while 4.0 × 0.20 = 0.8 is a loss.

MER (marketing efficiency ratio)

MER = total revenue ÷ all marketing spend, store-wide. Say $40,000 ÷ $12,500 = 3.2. It is attribution-free: because it never splits credit by channel, it cannot double-count the way per-platform ROAS does when Meta and Google both claim the same order. Use ROAS to tune a channel and MER to judge the whole engine. The marketing efficiency ratio guide and the deeper dive on what MER is both walk this out.

CAC, CPA, CPC

• CPC (cost per click) = ad spend ÷ clicks = $10,000 ÷ 20,000 = $0.50.
• CPA/CPO (cost per acquisition/order) = spend ÷ orders. It decomposes cleanly: CPA = CPC ÷ conversion rate = $0.50 ÷ 0.04 = $12.50.
• CAC (customer acquisition cost) counts new customers, not orders. Paid CAC = ad spend ÷ new customers = $10,000 ÷ 800 = $12.50; blended CAC adds tools and team, so $12,500 ÷ 800 ≈ $15.63.

The distinction matters: a returning buyer creates an order (feeding CPO) but not a new customer (not feeding CAC). Our how-to-calculate-CAC walkthrough covers the blended-versus-paid split in detail.

Customer value and retention metrics

Acquisition tells you what a customer costs. These tell you what they are worth.

AOV and LTV

AOV (average order value) = revenue ÷ orders = $40,000 ÷ 1,000 = $40.

LTV (lifetime value) is the whole relationship, not one order. On a margin basis: LTV = AOV × purchase frequency × lifespan × margin ratio = $40 × 1.6 × 2 × 0.60 = $76.80. Always state your basis — the revenue version of the same customer would read $40 × 1.6 × 2 = $128, which is why decks that mix a revenue-LTV against a profit-CAC overstate health badly.

LTV:CAC

LTV:CAC = LTV ÷ CAC = $76.80 ÷ $15.63 ≈ 4.9. The common health benchmark is a ratio of around three to one, a rule that traces back to SaaS economics and is explained by Harvard Business School Online. Below one-to-one you lose money on every customer; far above five-to-one you may be under-investing in growth. Two levers dominate the ratio: margin and lifespan.

Retention and churn

Retention rate = (customers at end − new) ÷ customers at start. Say (5,400 − 800) ÷ 5,000 × 100 = 92%. Churn is the complement: (1 − 0.92) × 100 = 8%, and customer lifespan ≈ 1 ÷ churn = 1 ÷ 0.08 = 12.5 periods — which is exactly the "lifespan" that feeds LTV. Cutting churn stretches lifespan and lifts LTV without touching CAC, which is why the customer retention rate formula is one of the highest-leverage numbers in the whole set.

Repeat purchase rate = customers with two or more orders ÷ total customers = 240 ÷ 1,000 × 100 = 24%. Unlike retention, it is cumulative rather than time-boxed.

On-site funnel metrics

Traffic is worthless until it converts. These measure the store itself.

Conversion rate (CVR) = orders ÷ sessions × 100 = 1,000 ÷ 40,000 × 100 = 2.5%. For context, the average ecommerce conversion rate sits in the low single digits — Dynamic Yield's benchmark data puts a typical figure near about 3%, though it swings widely by industry and device. Watch the denominator: per-session, per-visitor, and per-click all give different numbers.

Add-to-cart rate = cart-add sessions ÷ sessions = 4,000 ÷ 40,000 × 100 = 10%.

Cart abandonment rate = 1 − (orders ÷ carts). Say (1 − 1,000 ÷ 4,000) × 100 = 75%. That is close to the long-run industry norm: roughly seven in ten carts are abandoned, per the Baymard Institute, a figure that has barely moved in years. Small recovery wins here compound fast, which is why the cart abandonment rate deserves its own analysis.

Revenue per session (RPS) = revenue ÷ sessions = $40,000 ÷ 40,000 = $1.00. Neatly, RPS = CVR × AOV = 0.025 × $40 = $1.00, which is why a conversion win and an AOV win multiply rather than add.

Cost and margin metrics

These are the metrics the top-of-funnel guides always skip — and the ones that actually decide profit.

COGS is the direct product cost: $16 per order → $16 × 1,000 = $16,000.

Gross margin = (revenue − COGS) ÷ revenue = ($40 − $16) ÷ $40 × 100 = 60%.

Contribution margin subtracts all variable costs, not just COGS. Pre-ad (CM2): $40 − $16 − $8 = $16. Post-ad (CM3): $16 − $10 = $6. Gross margin tells you if a product is worth making; contribution margin tells you if it is worth selling through this channel at this CAC.

Net margin subtracts fixed costs too: total CM3 of $6 × 1,000 = $6,000, minus $4,000 of fixed cost = $2,000 net, or $2,000 ÷ $40,000 × 100 = 5%.

Markup vs margin trips up pricing constantly. Markup is over cost, margin is over price. The same order is ($40 − $16) ÷ $16 × 100 = 150% markup and ($40 − $16) ÷ $40 × 100 = 60% margin. Same gap, two numbers — never confuse them.

Break-even units = fixed costs ÷ contribution margin per order = $4,000 ÷ $6 ≈ 667 orders per month before profit begins.

How the metrics connect

The metrics are not a list — they are a web of identities. A few worth memorizing:

• CPA = CPC ÷ CVR. Cheaper clicks or better conversion both cut acquisition cost.
• Break-even ROAS = 1 ÷ margin ratio. Your margin sets the bar every campaign must clear.
• POAS = ROAS × margin ratio. Profitable exactly when POAS crosses one.
• RPS = CVR × AOV. CRO and merchandising compound.
• MER ≤ blended ROAS, always, because total marketing spend is at least ad spend.

When these identities disagree, it is almost always a definition problem — a margin used one way in break-even and another in LTV. Keeping one consistent margin basis across every ratio is what separates a real P&L from a pretty chart.

The pitfalls that quietly break the math

• Denominator drift. "Conversion rate" per session, per visitor, and per click are three different numbers. Standardize before comparing.
• Clicks vs link clicks. Meta's "all clicks" includes likes and taps; compute CVR on link clicks or landing-page views.
• Mixed-attribution double-counting. Summing each platform's claimed conversions over-counts shared journeys and inflates every channel's ROAS — the reason MER exists.
• Revenue basis vs profit basis. ROAS and revenue-LTV flatter; POAS and margin-LTV tell the truth. Keep numerator bases consistent in any ratio.
• Averages hiding the distribution. One "AOV $40" can mask many one-time buyers plus a few whales. Segment before acting.

Where PodVector fits

Most of the errors above share one root cause: the profit number lives in a spreadsheet while the revenue numbers live in the ad platforms, and nothing reconciles them per order.

PodVector connects Shopify, Meta Ads, Google Ads, Printify, Printful, and Stripe, and computes true per-order profit across all of them — so contribution margin, POAS, and break-even are grounded in real fees, real COGS, and real shipping instead of a rounded guess. PodVector is not a dashboard you have to read; Victor, its AI operator, analyzes your live data and proposes the moves, executing approved actions on the Shopify side. He reads your ad data to find what is working, but does not touch your ad account. You stay in control of every change.

See your true per-order profit with PodVector →

FAQs

What are the most important ecommerce metrics to track?

Start with contribution margin per order, then the ratios that depend on it: POAS, break-even ROAS, CAC, and LTV:CAC. Revenue-only metrics like ROAS and AOV are useful for tuning but can hide losses, so never judge a campaign on them alone.

What is the difference between ROAS and POAS?

Same denominator, different numerator. ROAS = revenue ÷ ad spend; POAS = profit ÷ ad spend = ROAS × margin ratio. ROAS is top-line and can look healthy on an unprofitable product; POAS is bottom-line and turns positive only when ROAS beats your break-even.

How do I calculate break-even ROAS?

Break-even ROAS = 1 ÷ your contribution-margin ratio. On a 0.40 margin that is 1 ÷ 0.40 = 2.5; on 0.25 it is 1 ÷ 0.25 = 4.0. Use contribution margin (net of shipping and fees), not gross margin, or you will set the bar too low.

What is a good LTV:CAC ratio for ecommerce?

Around three to one is the widely cited benchmark, per Harvard Business School Online. Below one-to-one you lose money per customer; well above five-to-one can signal you are under-spending on growth. Measure LTV on margin, not revenue, so it lines up with a real CAC.

Why don't my ad platform numbers match my analytics?

Because they count different universes. Ad platforms compute conversion on clicks and often over-claim credit for shared journeys; your analytics counts sessions. Never divide a platform's conversions by your analytics sessions — use a blended, attribution-free view like MER to judge whether the whole marketing engine is profitable.

Is revenue or profit the better metric to optimize?

Profit. Revenue metrics like ROAS and revenue-basis LTV can rise while your bank balance falls, especially on thin-margin products. Optimize channels on ROAS if you like, but decide what to scale on POAS and contribution margin.