Quick answer: Ecommerce KPIs are the numbers that show whether your store is healthy: conversion rate, average order value (AOV), revenue per visitor, customer acquisition cost, gross margin after shipping and discounts, and repeat purchase rate. Traffic alone is not enough. A short weekly scorecard beats a dashboard with forty vanity charts nobody reads.
Start with outcomes, then pick metrics
Every store exists to produce contribution margin sustainably. KPIs are instruments on that path. If a metric cannot change a decision, demote it. The list below is a practical core for small and mid-size merchants in 2020, whether you run Shopify, WooCommerce, Magento or another platform. Software choice changes where you click (Shopify vs Magento), not which questions matter.
1. Conversion rate
Orders divided by sessions (or users; pick one definition and keep it). Segment by device, channel and landing page. A blended rate can hide a strong desktop site and a weak mobile checkout. When the leak sits in purchase steps, prioritize product page clarity and checkout optimization.
2. Average order value (AOV)
Revenue divided by orders. Raise it with bundles (product bundles), thoughtful upsells and free-shipping thresholds that still protect margin. Discounts that slash AOV and margin together are not wins even if conversion ticks up.
3. Revenue per visitor (RPV)
Conversion rate times AOV for the same visit definition. RPV helps compare landing pages and campaigns when one page converts less often but sells higher-ticket items.
4. Customer acquisition cost (CAC)
Marketing spend attributed to new customers divided by new customers acquired. Include ads, agency fees and tools you buy for growth. Compare CAC to first-order contribution margin and to longer-term value if customers repeat.
5. Gross margin after variable costs
Revenue minus product cost, payment fees, shipping subsidies you absorb and attributable discounts. Busy stores with negative unit economics still fail. Track this by channel when paid social looks “cheap” on CAC but heavy on discounting.
6. Repeat purchase rate and purchase frequency
Percentage of customers who buy again within a window (90 or 180 days) and how often they return. Retention improves the math on CAC. Email and SMS flows, quality products and sane returns processes influence this more than a single pop-up.
7. Cart and checkout abandonment
How many start checkout versus finish. Pair the KPI with qualitative tools and form analytics. Fix address validation pain, surprise shipping costs and slow pages before buying more top-of-funnel traffic.
8. Refund and chargeback rates
High refunds signal product truth problems, sizing issues or shipping delays. Chargebacks signal fraud or broken expectations. Both erase apparent gains from conversion spikes.
9. Site performance indicators that affect revenue
Server errors during campaigns, p95 page load on PDPs and mobile CLS-style instability can mute every other KPI. Tie engineering work to revenue risk when you prepare for a traffic surge.
10. Inventory health (sell-through and stockouts)
Merchandising KPIs belong on the same scorecard as marketing. Ads that sell out SKUs you cannot replenish waste spend and disappoint customers. Sync realities between store and warehouse matter; Reliable inventory sync between store and warehouse reduces nasty surprises.
Build a scorecard, not a museum
Weekly sheet columns that work:
- Sessions
- Conversion rate (desktop / mobile)
- Orders
- AOV
- Revenue
- Gross margin $
- Blended CAC
- Email revenue share
- Refund rate
Add a notes column for launches, outages and promotions. Google Analytics annotations plus this sheet keep memory honest.
Channel-level views
Look at organic, paid search, paid social, email and direct separately. Organic may show lower conversion but better margin after you stop paying per click. Email often shows strong RPV because the audience already knows you. Do not starve a channel because its conversion rate differs without reading CAC and margin.
UTM discipline keeps channels honest (UTM parameters). Messy tagging creates fake “direct” and fake “referral” buckets.
Leading vs lagging indicators
Lagging: monthly profit, repeat rate. Leading: add-to-cart rate, checkout start rate, email click rate, PDP views per session. Leading indicators let you act mid-month. Still judge success on lagging money metrics.
Targets without fantasy
Borrow benchmarks cautiously. A specialty high-ticket store will not match a low-cost impulse shop’s conversion rate. Set targets from your own trailing three months, then aim for incremental gains. A 10% relative lift in conversion with stable AOV and margin is concrete. “We want to 10x” without a model is not a KPI plan.
Common reporting mistakes
- Celebrating traffic records while conversion falls
- Ignoring device splits
- Mixing gross sales with net after refunds in the same sentence without labels
- Attributing all revenue to the last click forever without discussion
- Optimizing ROAS while quietly increasing discounts
FAQ
How many KPIs should leadership see?
Five to nine on the primary scorecard. Analysts can dig deeper. Executives glaze over at forty tiles.
Should we track Domain Authority or similar SEO scores as ecommerce KPIs?
Not as primary store health metrics. Organic revenue, non-brand organic conversions and category rankings for money terms matter more.
What tools do we need?
Your platform reports, Google Analytics, ad platforms and a spreadsheet can cover the core. Add BI tools when volume justifies them.
How do offline events fit?
If you run pop-ups or wholesale, track those revenues separately so online KPIs stay interpretable.
Cohort views for subscription and replenishment brands
If you sell refillable goods or memberships, add cohort retention charts: what share of January buyers purchase again in months two and three. Blended repeat rate can hide a worsening new-customer quality problem while older loyal buyers still carry revenue. Cohorts also show whether a brutal discount cohort ever becomes profitable.
Connecting KPIs to experiments
Every A/B test or major UX change should name the primary KPI and guardrails. Example: primary = checkout conversion; guardrails = refund rate and AOV. If conversion rises because you removed a friction field that later spikes support tickets, you did not win. Write the hypothesis in one sentence before launch.
Sharing KPIs with agencies and freelancers
Give partners the definitions, not only the targets. If “conversion rate” means orders divided by sessions excluding bot filters in one tool and users in another, debates will waste the retainer. Hold a thirty-minute metrics alignment meeting at the start of every engagement. Revisit after major tagging changes.
When paid media agencies optimize to platform-reported purchases, reconcile weekly with your store’s net orders after cancellations. Platform pixels and reality diverge; your scorecard should prefer store truth for money decisions.
Offline and wholesale caveats
If part of your revenue is purchase orders outside the cart, keep a separate line for offline sales so online conversion rate stays meaningful. Mixing wholesale bulk orders into the same AOV chart without a filter will confuse every optimization meeting.
Weekly operating rhythm
- Update the scorecard every Monday
- Pick one red metric to investigate
- Ship one fix or test
- Annotate the change
- Review again next week
That rhythm turns ecommerce KPIs into management, not decoration.
If you want help instrumenting analytics and reporting that ties to store profit, see our Turn visitors into customers ecommerce services.
