Blog

Jesús Avendaño
·
Revenue is the least informative number on a merchant's dashboard when it stands alone. It tells you what customers bought, not what you kept. Consider returns: a Loop report covered by Morningstar analyzed more than 13 million returns from 4,000 Shopify merchants and found that return fees have become standard practice. A completed sale doesn't settle the economics.
Revenue tells you half the story. You still need to account for product costs, advertising, fulfillment, payment fees, discounts, refunds, and the work required to keep customers buying. A busy store can generate more orders while leaving its owner with less money.
The most useful metrics for ecommerce form a profitability stack, not a collection of disconnected dashboard tiles. Start with net profit and gross margin. Then examine acquisition efficiency through ROAS, MER, and CAC. Use AOV and conversion rate to diagnose order economics and purchasing friction. Finally, evaluate repeat purchasing, lifetime value, and the balance between new and returning revenue.
Operating principle: Improve the metric that constrains profit, not the metric that makes the weekly presentation look better.
Each KPI below includes a formula, a hypothetical worked example, and the conditions under which it helps or misleads. The examples illustrate calculations, not industry benchmarks.
1. Net Profit
Net profit is what remains after the store pays its costs. It's the starting point because every other metric should eventually explain why this number changed.
Formula: Net profit = Net revenue − COGS − Advertising spend − Shipping and fulfillment − Fees − Operating expenses.
Use net revenue after discounts and refunds. Include return handling and inventory write-downs in the appropriate cost categories, but don't subtract the same expense twice.
Follow the money beyond the order
Say a store generates $200,000 in monthly net revenue, but a best-selling bundle loses money after product costs, shipping subsidies, advertising, and return handling. Repricing the bundle could improve profit even if it reduces order volume.
For a worked calculation, subtract each recorded cost category from that revenue. The important operating step is tracing the bundle's loss to a controllable expense rather than assuming its sales rank justifies more promotion.
Use it for pricing, budget allocation, and evaluating promotions. Review profit alongside revenue weekly, with a later reconciliation against accounting records.
When it misleads: Missing variant costs make profit look better than it is. Allocating shared overhead arbitrarily can also make a healthy product appear unprofitable.
Separate product contribution from fully allocated net profit. Use contribution to make merchandising decisions, then verify that total contribution covers the business's fixed expenses.
2. Gross Margin
A product can sell easily and still leave too little money to acquire the next customer. Gross margin measures the share of revenue left after the cost of goods sold.
Formula: Gross margin % = (Net revenue − COGS) ÷ Net revenue × 100.
Keep the revenue basis consistent with your profit reporting. Landed product costs belong in COGS; outbound delivery, payment processing, and advertising usually need separate treatment.
For a hypothetical skincare product with a 70% gross margin, the gross-margin-only break-even ROAS is approximately 1.4×, calculated by dividing one by the margin expressed as a decimal. That's an advertising ceiling before other expenses, not permission to scale.
Set promotion limits before launching
Use a profit margin calculation to establish how much room a discount leaves. Calculate it at the promotional price, not the normal selling price.
Recalculate costs: Update variant costs when suppliers change prices or inbound freight changes.
Protect paid promotion: Favor products with enough margin to absorb acquisition and fulfillment expenses.
Check bundles: Calculate the combined margin instead of assuming a larger basket is more profitable.
When it misleads: Gross margin excludes expenses that can decide whether an ecommerce order makes money. A seemingly attractive margin can disappear after shipping subsidies, processing fees, and customer acquisition.
Use gross margin to screen pricing decisions. Use contribution margin and net profit to approve them.
3. Return on Ad Spend (ROAS)
ROAS measures attributed revenue per unit of advertising spend. It's useful for comparing campaigns, audiences, and creative treatments within a consistent measurement setup.
Formula: ROAS = Attributed revenue ÷ Advertising spend.
In a hypothetical account, a campaign reports 5.0× ROAS, while store-level revenue divided by advertising spend produces 2.6×. That difference warrants an attribution review, but it doesn't prove the campaign report is wrong: the calculations use different revenue scopes.
Check attribution windows, view-through credit, refunds, and whether multiple platforms claim the same purchase.
Give ROAS a profit threshold
Use a break-even ROAS calculator before increasing budgets. For a practical threshold, account for variable costs beyond COGS rather than treating gross margin as spendable cash.
When it misleads: Platform attribution measures assigned credit, not necessarily additional sales caused by advertising. Retargeting existing buyers can look efficient without proving those customers needed an ad to purchase.
A 2025 Upcounting benchmark reported by Zembula put average ecommerce ROAS at 2.87, with a median of 2.04 across 14 industries. Neither figure determines your store's acceptable return.
Compare Meta and Google using consistent windows and cost definitions. Approve scaling against actual contribution, not a platform's green performance indicator.
4. Marketing Efficiency Ratio (MER / Blended ROAS)
MER answers a broader question than campaign ROAS: How much store revenue did the business generate relative to its total marketing spend?
Formula: MER = Total store revenue ÷ Total marketing spend.
Define the denominator before comparing periods. If agency retainers, influencer payments, affiliate commissions, and creative production are included, keep including them.
Say channel dashboards report 3.5× ROAS, but the store's MER is 2.1×. Those hypothetical results aren't interchangeable: platform ROAS uses attributed revenue and media spend, while MER uses store revenue and your chosen marketing-cost scope.
MER avoids adding together overlapping platform revenue claims. It doesn't prove that marketing caused every order.
Use it as a budget guardrail
Track MER alongside channel performance during sales and budget increases. If spend rises faster than revenue, investigate whether additional customers are becoming more expensive to acquire.
Few teams fully trust their attribution, and that uncertainty is a reason to reconcile multiple views, not replace attribution with another unquestioned ratio.
When it misleads: Existing customers, organic demand, seasonality, and wholesale orders can support MER without reflecting current advertising effectiveness.
Use controlled holdouts when feasible to estimate incremental sales. For everyday operations, pair MER with new-customer volume and contribution after marketing, then investigate disagreements rather than averaging them away.
5. Customer Acquisition Cost (CAC)
If you don't know what you can afford to pay for a customer, increasing spend is a financing decision disguised as a marketing decision.
Formula: CAC = Acquisition costs ÷ New customers acquired.
Include the acquisition expenses relevant to the decision: advertising, agency support, creative production, and allocated tools or labor. Discounts already deducted from net revenue shouldn't also appear as an acquisition expense.
Compare acquisition cost with available profit
Suppose a merchant spends $30,000 acquiring 1,000 new customers. CAC is $30. If first-order gross profit is $42, the merchant has $12 left before fulfillment, processing, other variable expenses, and overhead.
That's not a comfortable profit merely because the order value exceeds CAC. Calculate first-order contribution before deciding whether acquisition is self-funding.
Separate paid and blended CAC: Organic customers can make paid acquisition look cheaper when everything is pooled.
Compare channels carefully: A low-CAC channel may acquire customers who buy once and return expensive products.
Track payback: Identify when cumulative customer contribution recovers acquisition cost.
When it misleads: Short reporting windows can mismatch advertising expenses with purchases that happen later. Customer identification errors can also classify returning buyers as new.
Review CAC monthly by acquisition cohort and channel. During active scaling, monitor directional changes more frequently, but don't treat daily fluctuations as settled economics.
6. Average Order Value (AOV)
AOV is an order-size metric, not a profit metric. Larger baskets help only when the additional merchandise contributes more than the discounts and servicing costs required to sell it.
Formula: AOV = Net order revenue ÷ Number of orders.
Use a consistent treatment of taxes, delivery charges, refunds, and canceled orders. Otherwise, changes in reporting definitions can look like changes in customer behavior.
Test the basket, not just the threshold
Suppose a merchant raises a free-shipping threshold from $50 to $75, and AOV rises from $58 to $71. The hypothetical increase is $13 per order, but the test succeeds only if contribution improves without an unacceptable loss of orders.
Inspect what customers added. A full-price accessory may improve economics; a heavily discounted, bulky item may not.
Use it to evaluate bundles, product recommendations, quantity offers, and shipping thresholds. Segment new and returning customers because their purchasing patterns may differ.
When it misleads: AOV can increase because low-value shoppers stop buying. The average improves while total contribution falls.
Pair every AOV experiment with order count, conversion rate, and contribution per session. That combination helps distinguish better baskets from a smaller group of unusually large buyers.
7. Conversion Rate
Before blaming traffic quality, check whether customers can complete their purchases.
Formula: Purchase conversion rate = Orders ÷ Sessions × 100.
Choose sessions or visitors and stick to that denominator. For a session-based metric, use orders and sessions from compatible reporting systems and time zones.
In a hypothetical comparison, a paid landing page converts at 0.8%, while the site overall converts at 2.5%. That gap identifies a page worth investigating, not proof that rebuilding it will recover the difference.
Diagnose the step that fails
Campaign visitors may have different intent from returning customers. Compare like-for-like device and traffic segments, then inspect product view, add-to-cart, checkout start, payment submission, and order confirmation.
Baymard's checkout usability benchmark found that 65% of benchmarked sites had mediocre-or-worse checkout UX. Checkout deserves its own diagnosis rather than being folded into a marketing explanation.
Mobile failure: Check address entry, payment buttons, browser compatibility, and error messages.
Cost-related exits: Show delivery charges and total costs before customers commit to checkout.
Campaign mismatch: Align the landing page's product, price, and promise with the advertisement.
When it misleads: Heavy discounting or a shift toward high-intent returning traffic can improve conversion without improving profitability. Judge experiments against contribution as well as completed orders.
8. Repeat Purchase Rate
A customer placing another order is more useful evidence of retention than opening a promotional email. Repeat purchase rate measures how many customers buy again.
Formula: Repeat purchase rate = Customers with multiple purchases ÷ Customers in the defined cohort × 100.
Always attach a measurement window. Compare acquisition cohorts after equal time has elapsed, rather than giving older customers more opportunity to return.
Find the products that earn another order
In a hypothetical supplement store, subscription-eligible buyers have a 90-day repeat rate of 38%, compared with 12% for one-off buyers. That difference supports testing subscription education after purchase, but it doesn't establish that subscriptions caused the higher repeat rate.
Those customers may already have stronger replenishment intent. Compare product mix, acquisition source, and first-order discount exposure before changing strategy.
A DTC analysis cited in Shopify's retention guide found an average repeat-purchase rate of 18.8% across more than 156,000 customers. Treat it as context, not a universal target.
When it misleads: A store selling durable furniture shouldn't evaluate repurchasing like a replenishable skincare brand. Forced subscription renewals can also obscure dissatisfaction.
Track second-order timing and repeat contribution alongside the rate. Send replenishment reminders around observed buying intervals, and judge retention campaigns on profitable additional purchases rather than discount redemptions alone.
9. Customer Lifetime Value (LTV)
LTV determines how much future customer value you're willing to fund today. That makes its assumptions more important than its apparent precision.
For acquisition decisions, use contribution-based value rather than revenue.
Formula: Observed contribution LTV = Cumulative cohort contribution before acquisition costs ÷ Customers originally acquired.
Keep acquisition cost outside that calculation when comparing LTV with CAC. If your report subtracts CAC inside profit LTV, label it clearly so you don't subtract acquisition twice.
Separate observed value from forecasts
Suppose a January cohort reaches $95 in contribution LTV after six months, against $40 CAC. Its observed contribution after acquisition is $55 per acquired customer, before any shared overhead excluded from the calculation.
That supports further investigation of higher spending, not unlimited expansion. New audiences may behave differently, and the store still has to finance the wait for repeat purchases.
Use customer lifetime value measurement to build acquisition-month and channel cohorts. Compare cohorts at equal ages and show actual results separately from predictions.
Budget discipline: Don't pay today for repeat purchases that only exist in an optimistic forecast.
When it misleads: Revenue LTV ignores the cost of serving customers. Predictive LTV can overstate value when retention weakens or returns arrive late.
Monitor second-order behavior as an early signal, but retain conservative cash assumptions. A profitable customer relationship can still create a cash shortage if payback takes too long.
10. New vs. Returning Revenue Share
This metric shows whether revenue comes from customers making their first purchase or customers placing subsequent orders.
Formula: Returning revenue share = Revenue from repeat orders ÷ Total revenue × 100.
Calculate new-customer revenue share using first-order revenue as the numerator. Define the classification at order level so a customer's later purchase doesn't retroactively turn their original order into returning revenue.
Read the share alongside absolute revenue
Say returning revenue rises from 22% to 34% over six months after a merchant launches a loyalty email series. In that hypothetical scenario, the increase is encouraging only if returning revenue and contribution also grow.
The share could rise because acquisition collapsed. Equally, a strong acquisition push can reduce returning share even while existing customers spend more.
Keep repeat purchasing and revenue performance apart. That distinction matters because larger purchases from a small customer group can conceal weaker retention across the wider customer base.
Track both amounts: Show new and returning revenue in currency as well as percentages.
Check cohort health: Confirm that recent customers return, rather than relying exclusively on older loyal buyers.
Compare margins: Returning orders aren't automatically profitable if they require repeated discounts and expensive servicing.
When it misleads: There's no universally healthy revenue split. Replenishment businesses, gift brands, and durable-goods stores have different purchasing patterns.
Use the split to guide investigation into acquisition dependence, not to force every store toward the same retention target.
The 10 Ecommerce Metrics at a Glance
Metric | Category | Formula | Best For | Watch Out For |
|---|---|---|---|---|
Net Profit | Profitability | Net Revenue - COGS - Ad Spend - Fees - Operating Expenses | Assessing true profit and identifying cost leakage | Missing costs or assumptions when allocating shared expenses |
Gross Margin | Profitability | (Revenue - COGS) / Revenue x 100 | Pricing, promotion, and product-level profitability decisions | Excludes marketing and overhead; depends on accurate COGS |
Return on Ad Spend (ROAS) | Advertising | Ad-attributed revenue / Ad spend | Comparing campaign and creative revenue performance | Attribution overlap and lack of profit context |
Marketing Efficiency Ratio (MER / Blended ROAS) | Marketing | Total Revenue / Total Marketing Spend | Measuring overall marketing efficiency across channels | Does not identify channel performance and can include organic revenue |
Customer Acquisition Cost (CAC) | Customer Acquisition | Total acquisition cost / Number of new customers | Evaluating the cost and sustainability of customer growth | Timing mismatches and blended figures can obscure channel costs |
Average Order Value (AOV) | Sales | Total revenue / Number of orders | Evaluating bundles, upsells, and shipping thresholds | Does not show profitability and can hide falling order volume |
Conversion Rate | Conversion | Orders / Sessions x 100 | Diagnosing store and checkout performance | Averages can hide differences; discounts may boost conversion but hurt profit |
Repeat Purchase Rate | Retention | Customers with 2+ orders / Total customers in the period or cohort | Measuring customer loyalty and retention | Depends on measurement window, product type, and customer identification |
Customer Lifetime Value (LTV) | Customer Value | Total customer value over the relationship; use profit-based LTV for decisions | Setting acquisition budgets and prioritizing retention | Forecasts can be inaccurate, and revenue-based LTV can overstate value |
New vs. Returning Revenue Share | Revenue Mix | Returning revenue / Total revenue; new share is the inverse | Tracking reliance on new acquisition versus existing customers | Customer misidentification and seasonal changes can distort the split |
Turning the Metrics Into Decisions
A dashboard becomes useful when someone changes a decision because of it. Start with the current constraint, not a plan to optimize every KPI simultaneously.
If revenue is growing but cash feels tighter, investigate net profit, gross margin, and acquisition payback. If acquisition costs look stable but orders fall, inspect conversion by device and checkout step. If first orders look healthy but growth requires constant advertising increases, examine repeat purchasing and cohort LTV.
Choose a small operating focus
Pick two or three metrics that match the bottleneck. Give each an owner, a review cadence, and a decision it can trigger.
Profitability constraint: Review profit and margin weekly. Decide which products, discounts, or costs need intervention.
Acquisition constraint: Review CAC and MER weekly during active spending changes. Check whether additional investment produces acceptable contribution.
Retention constraint: Review mature cohorts monthly. Compare repeat behavior at equal cohort ages before changing lifecycle offers.
Keep the wider stack available for diagnosis. A falling MER might reflect weaker conversion rather than worse creative, while an improving AOV might conceal fewer completed orders.
Before trusting any profit number, complete the cost data. Enter variant-level product costs, reconcile refunds, capture shipping subsidies and payment fees, and include recurring software and operating expenses. Mark missing costs explicitly instead of treating them as zero.
Build one consistent reporting view
Use the same currency, time zone, revenue definition, and customer classification across reports. Separate preliminary operational figures from reconciled financial results, especially when returns and invoices arrive after the sale.
Cifra is a Shopify-native analytics platform that consolidates sales, advertising spend, product costs, fees, and operating expenses into a frequently refreshed profit view. Its dashboards, cohort reports, plain-language Ask bar, and scheduled delivery through email and Slack let teams work with the same definitions instead of maintaining disconnected exports.
The practical benefit isn't having fewer tabs. It's being able to connect a change in profit with the products, campaigns, discounts, or customer groups behind it, then decide what to investigate.
At the next review, select one proposed action and write down its expected effect. Specify the metric you expect to improve, the profit safeguard you won't compromise, and when you'll evaluate the result. That discipline turns metrics for ecommerce into a repeatable decision process rather than a weekly collection of numbers.
Cifra brings Shopify sales, advertising costs, fees, and customer performance into one profit-focused reporting view. Visit Cifra to connect your store and build dashboards around the profitability, acquisition, conversion, and retention metrics that guide your next decision.
Jesús Avendaño
·