Most metric roundups hand you a list of thirty definitions and stop there. That is fine if you have never sold anything. But you run a store — real orders, real ad spend, real supplier invoices — so you need to know which numbers to read together, and where the popular ones lie to you.
Throughout, we will use one example store so the math ties together. Say you run Riverbend Tees, a print-on-demand apparel shop doing 340 orders a month at a $31 average order value ($10,540 in revenue), spending $2,800/month on Meta ads, with a blank-plus-print cost of $12.40 per order. Every number below traces back to that store.
The four groups of ecommerce analytics metrics
You do not need all thirty. You need one or two from each of four groups, because each group answers a different question:
- Ad efficiency — are my ads bringing in customers profitably?
- On-site funnel — is my store turning traffic into orders?
- Profit — after every cost, am I actually keeping money?
- Retention — do customers come back, and is that worth what I paid to get them?
If you only track the first two, you are measuring activity, not profit. That is the gap this article closes. For the bigger picture of how these feed decisions, see our overview of ecommerce business intelligence.
Ad-efficiency metrics
ROAS (return on ad spend) is revenue divided by ad spend. Riverbend's blended ROAS is $10,540 ÷ $2,800 = 3.76. That looks healthy — and it can still be a losing number, which is the trap in the next section.
POAS (profit on ad spend) replaces revenue with profit in the numerator. On a gross-profit basis, POAS = ROAS × gross-margin ratio. At Riverbend's 60% gross margin: 3.76 × 0.60 = 2.26. POAS above one means the ads made money; below one means they lost it no matter how good the ROAS looked.
CAC (customer acquisition cost) is ad spend divided by new customers, not orders. If 280 of the 340 orders came from first-time buyers, CAC = $2,800 ÷ 280 = $10.00. This is different from cost per order (CPO), which counts every order including repeat buyers.
For context on what "good" looks like: the median Meta Ads ROAS for fashion and apparel brands sits around 2.18x according to Adamigo's 2026 benchmark, and the median ecommerce ROAS across all categories is roughly 2.0x per TrueProfit's 2026 data. Riverbend's 3.76 blended figure is strong for the category — but benchmarks are context, not a target.
The profit metrics roundups skip
This is where most "top ecommerce metrics" articles go thin. They define ROAS and stop, so you never learn the number that decides whether ROAS is good: your break-even.
Gross margin subtracts only product cost. For Riverbend: ($31 − $12.40) ÷ $31 = 60%.
Contribution margin (CM2) subtracts all variable costs before ads — shipping ($4.20), payment processing (about 3%, or $0.93), and pick/pack ($1.10). That is $6.23 more per order, so CM2 = $31 − $12.40 − $6.23 = $12.37 per order, a 40% margin ratio.
Break-even ROAS is the single most useful identity in paid media: 1 ÷ contribution-margin ratio. On Riverbend's 40% CM2, that is 1 ÷ 0.40 = 2.5. Any campaign below 2.5 ROAS loses money even while the platform reports a "profitable" 2.0.
Here is why that matters. A store with a 25% margin needs roughly 4x ROAS just to break even, while a 70% margin store is profitable below 2x — the exact point ClickZ makes in its 2026 ROAS guidance. Two stores can hit the same ROAS and one is thriving while the other bleeds. The number that separates them is margin, and it is the number most dashboards never show you.
Contribution margin after ads (CM3) finishes the picture. Riverbend's ad cost per order is $2,800 ÷ 340 = $8.24, so CM3 = $12.37 − $8.24 = $4.13 per order. That is the real per-order profit before fixed costs — the number our ecommerce performance analytics guide builds channel decisions on.
On-site funnel metrics
Conversion rate (CVR) is orders ÷ sessions. If Riverbend gets 13,600 sessions a month, CVR = 340 ÷ 13,600 = 2.5%. For comparison, the median apparel and accessories store converts at about 1.69% and the upper quartile near 3.24%, per Shogun data reported by Blend Commerce. So Riverbend converts above the apparel median — a real edge worth protecting.
Cart abandonment rate is the share of carts that never become orders. The long-run industry average is 70.22%, calculated by Baymard across fifty studies. If your rate is far above that, checkout friction — not traffic — is your problem.
Revenue per session (RPS) blends the two: revenue ÷ sessions. For Riverbend, $10,540 ÷ 13,600 = $0.78. It equals CVR × AOV (0.025 × $31 = $0.775), which is why a conversion win and an average-order-value win multiply rather than add.
Retention and lifetime value
Repeat purchase rate is the share of customers who buy more than once. It is the cheapest lever you have, because a repeat order carries no CAC.
LTV (customer lifetime value), on a margin basis, is AOV × purchase frequency × lifespan × margin ratio. Say Riverbend customers buy 1.6 times a year for two years at a 60% gross margin: $31 × 1.6 × 2 × 0.60 = $59.52.
LTV:CAC ratio compares that to acquisition cost: $59.52 ÷ $10.00 = about 6:1. When lifetime margin is several times your CAC, you have room to spend more to grow. When it drifts toward one-to-one, acquisition is underwater and no amount of ad tuning fixes it. For the forward-looking version of this analysis, see our note on predictive analytics for ecommerce.
How the metrics tie together
The metrics are not a checklist — they are a chain. Reading Riverbend's month top to bottom:
| Metric | Value | What it tells you |
|---|---|---|
| Blended ROAS | 3.76 | Revenue per ad dollar |
| Break-even ROAS | 2.5 | The floor ROAS must clear |
| POAS (gross) | 2.26 | Ads are profitable (above 1) |
| CM3 per order | $4.13 | Real per-order profit pre-fixed-cost |
| Conversion rate | 2.5% | Store turns traffic to orders |
| LTV:CAC | ~6:1 | Room to reinvest in growth |
The chain reads like this: your break-even ROAS (set by margin) tells you whether your actual ROAS is good; POAS confirms it in profit terms; CM3 tells you the dollars per order; and LTV:CAC tells you whether to push harder. Miss the middle two and you will scale a campaign that looks like a winner and shrinks your bank balance.
That reconciliation — pulling true per-order profit out of Shopify orders, supplier costs, and ad spend at once — is exactly the manual work that eats a POD operator's week.
Where PodVector AI fits
PodVector AI is the company; Victor is an AI employee that works over your live store data. Victor connects to Shopify, Meta Ads, Google Ads, Printify, Printful, Gelato, and Klaviyo, and computes true per-order profit — the CM2 and CM3 numbers above — instead of leaving you to rebuild them in a spreadsheet. Victor is not a dashboard you read; it is an employee you ask.
Victor delivers reports straight to your Google Drive, and can draft approval-gated customer-support emails that you approve before they send. Every write action Victor takes is approval-gated, so nothing executes without your sign-off. Put Victor to work on your store's numbers and get your real per-order profit without the reconciliation.
If you want a human-guided version of this analysis first, our ecommerce analytics consulting overview covers when that makes sense.
FAQs
Which ecommerce analytics metrics should an operating store track first?
Start with one metric from each of the four groups: ROAS for ad efficiency, conversion rate for the funnel, contribution margin for profit, and LTV:CAC for retention. Those four answer four different questions and stop you from optimizing one at the expense of the others. Add the rest once these are stable and trustworthy.
Why is ROAS misleading on its own?
ROAS measures revenue per ad dollar, not profit. A 3.0 ROAS is excellent on a 60% margin product and a loss on a 25% margin one, because break-even ROAS is 1 ÷ your margin ratio. Always pair ROAS with your break-even number or with POAS (profit on ad spend), which bakes margin into the calculation.
What is the difference between CAC and cost per order?
CAC counts new customers in its denominator; cost per order counts every order, including repeat buyers. For a store with real retention they diverge sharply — a returning customer generates an order (raising your order count) without being a new customer. Use CAC to judge acquisition and cost per order to judge fulfillment-side economics.
What is a healthy conversion rate for an apparel store?
The median apparel and accessories store converts at roughly 1.69%, with the upper quartile near 3.24%, according to Shogun's 2026 benchmark. Treat those as context, not goals — your traffic mix and price point move the number a lot. A worked store at 2.5% is comfortably above the category median.
How is contribution margin different from gross margin?
Gross margin subtracts only the cost of goods. Contribution margin subtracts every variable cost — product, shipping, payment fees, fulfillment, and (in CM3) ad spend. Gross margin tells you whether a product is worth making; contribution margin tells you whether it is worth selling through this channel at this acquisition cost.
Can I track all of this without a spreadsheet?
Yes. The blocker is usually that the numbers live in different places — orders in Shopify, ad spend in Meta and Google, product cost in your print supplier. Victor, PodVector AI's AI employee, connects those sources and computes true per-order profit directly, so the profit metrics above are read from live data rather than rebuilt by hand each month.