Ecommerce Analytics: A Complete Guide for Online Stores
A complete ecommerce analytics guide: acquisition, product, funnel, customer, revenue, retention, merchandising, UX and experimentation, plus data quality.
Quick answer
Track what answers your business questions. For most stores that means revenue, orders, average order value and margin from the commerce platform; sessions, conversion and the funnel from product view to purchase by channel and device from web analytics; new versus returning customers and repeat purchase for retention; and returns and stock-outs for operations. Write a measurement plan, implement standard ecommerce events, validate them against real orders, treat the commerce platform as the source of truth for revenue, and review data on a daily, weekly and monthly rhythm.
Start With Questions, Not Reports
Stores drown in dashboards because tracking starts with tools rather than questions. List the decisions you make regularly and what you'd need to know to make them better. The diagram above shows the cycle: questions shape the event plan, collected data is validated, reported and used to decide, and decisions raise new questions.
| Question | Metric | Source |
|---|---|---|
| Are we growing profitably? | Net revenue, gross margin, contribution after marketing | Commerce platform, finance |
| Which channels bring buyers? | Revenue, orders and conversion by channel | Web analytics, ad platforms |
| Where do shoppers drop off? | Stage-to-stage funnel rates by device | Web analytics |
| Which products sell and which don't? | Views, add-to-cart rate, sales, returns by product | Web analytics, commerce platform |
| Do customers come back? | Repeat purchase rate, cohort revenue | Commerce platform |
| Is search working? | Zero-result queries, search conversion | Web analytics, search tool |
Sources and What Each Is Good For
| Source | Best for | Limits |
|---|---|---|
| Commerce platform | Orders, revenue, refunds, customers, products | Limited view of pre-purchase behavior |
| Web analytics (e.g. GA4) | Sessions, journeys, funnel, channels, events | Consent and blocking gaps; modelled data |
| Search Console | Organic queries, clicks, indexing | Search only |
| Ad platforms | Spend, impressions, platform-attributed conversions | Each claims credit its own way |
| Email and CRM | Engagement, lifecycle, retention | Owned channels only |
| Heatmaps, recordings, surveys | Why people behave as they do | Samples, not totals |
Pro tip
Decide once which source is the truth for each number. Revenue and orders from the commerce platform; behavior from analytics. Most reporting arguments come from mixing them.
The Event Plan
Google documents a standard set of recommended ecommerce events for GA4 (Google Analytics developer documentation). Using them, rather than inventing names, makes the built-in ecommerce reports work.
- Every event carries an items array with item_id and item_name at minimum
- Events with revenue send value and currency
- purchase sends a unique transaction_id so duplicates can be removed
- Custom events only for questions the standard set can't answer (size guide opened, filter applied)
| Event | When it fires |
|---|---|
| view_item_list / select_item | A product list is seen / a product in it is clicked |
| view_item | A product page is viewed |
| add_to_cart / remove_from_cart | Items are added or removed |
| view_cart | The cart is opened |
| begin_checkout | Checkout starts |
| add_shipping_info / add_payment_info | Shipping and payment steps are completed |
| purchase / refund | An order is placed / refunded |
| view_promotion / select_promotion | A promotion is seen / clicked |
Validate Before You Trust
Tracking breaks quietly: a theme update removes a script, checkout moves to a new domain, an app fires a second purchase event. Validate after every release.
- Place test orders and check they appear once, with the right value and currency
- Compare analytics revenue with platform revenue weekly; investigate sudden changes in the gap
- Check sessions aren't split when shoppers move to checkout
- Confirm consent settings behave as intended
- Look for self-referrals from payment providers
- Keep a change log so data shifts can be explained
What to Track by Area
| Area | Core metrics |
|---|---|
| Acquisition | Sessions, conversion rate and revenue by channel, campaign and landing page; cost per acquisition |
| Funnel | Product view rate, add-to-cart rate, cart-to-checkout rate, checkout completion, by device |
| Merchandising | Product list click-through, views, add-to-cart rate and sell-through by product and category |
| Search and discovery | Search usage, zero-result queries, search exits, filter use |
| Customers | New vs returning revenue, repeat purchase rate, cohort revenue, lifetime value |
| Economics | Average order value, discount rate, gross margin, returns rate |
| Experience | Core Web Vitals, JavaScript errors, payment failures |
Conversion and Funnel Reporting
A single conversion rate hides where problems are. Report stage-to-stage rates by device and channel so you can see whether shoppers fail to find products, fail to add them, or fail to finish checkout. See ecommerce conversion rate and ecommerce conversion funnel.
Not sure your store's data can be trusted?
ZSpace audits ecommerce tracking against real orders and fixes the gaps before you base decisions on it.
Attribution: Useful, Not Exact
Every attribution model assigns credit by rules, and ad platforms each count conversions their own way, so their totals rarely add up to your actual orders. Use attribution to compare channels directionally, check platform claims against total revenue, and run holdout or incrementality tests for large budget decisions.
Customer and Retention Analytics
Conversion rate measures first purchases; the economics of most stores depend on the second. Track repeat purchase rate, time to second order and revenue per customer by acquisition cohort. See ecommerce cohort analysis.
Qualitative Evidence
Numbers show where; qualitative evidence shows why. Pair funnel data with session recordings, heatmaps, on-site surveys, support tickets, reviews and usability tests before deciding what to change. See ecommerce heatmaps and customer journey analytics.
Reporting Rhythm
| Cadence | Focus | Audience |
|---|---|---|
| Daily | Revenue vs expected, tracking health, errors, stock-outs | Ecommerce manager |
| Weekly | Channels, funnel by device, top products, search | Ecommerce, marketing, merchandising |
| Monthly | Cohorts, margin, returns, experiments, trends | Leadership |
Privacy and Consent
Respect consent choices, avoid sending personal data such as emails in event parameters, and document what's collected. Expect gaps: some visitors won't be tracked or will be modelled. Design reports around trends and ratios that tolerate gaps rather than exact counts.
The Analytics Domains of an Online Store
Ecommerce analytics isn't one report. It's a set of domains, each answering different questions for different teams. A complete practice covers all of them at the depth your business needs, built on shared definitions.
| Domain | Core questions | Deeper guide |
|---|---|---|
| Acquisition | Which channels bring profitable customers? | Attribution and channel reporting |
| Product | Which products and categories perform, and why? | Product analytics |
| Funnel | Where do shoppers drop off? | Funnel analytics |
| Customer | Who are our customers and how do they behave? | Segmentation, cohorts |
| Revenue | What drives revenue and margin? | Dashboards, driver trees |
| Retention | Do customers come back? | Cohorts, repeat purchase |
| Merchandising | Is the catalog organized to sell? | Merchandising metrics |
| UX | Where do interfaces cause friction? | Event tracking, research |
| Experimentation | Did a change cause an improvement? | Testing programmes |
How the Pieces Fit Together
This guide is the hub for a set of deeper articles. Analytics architecture covers where data comes from and how it's combined. Event tracking covers what to measure and how to name it. Funnel analytics covers drop-off diagnosis, product analytics covers product performance, KPI dashboards cover metric selection and dashboard design covers presentation. For segments and cohorts, see customer segmentation and cohort analysis.
An Analytics Maturity Path
| Stage | Characteristics | Next step |
|---|---|---|
| Basic | Platform reports, page views and purchases | Tracking plan with core ecommerce events |
| Structured | Ecommerce events, funnels, shared definitions | Product and category analysis, segmentation |
| Integrated | Orders, refunds, marketing and CRM joined | Cohorts, margin, experimentation programme |
| Advanced | Modelled metrics, forecasting, automation | Continuous optimization and governance |
Common Analytics Mistakes
- Tracking everything and deciding nothing
- Reporting analytics revenue as if it were accounting revenue
- Duplicate purchase events inflating conversion
- One site-wide conversion rate with no segments
- Summing ad platform conversions
- No change log, so data shifts can't be explained
Want analytics that lead to decisions?
Talk to ZSpace about CRO and analytics audits and tracking implementation.
Conclusion
Good ecommerce analytics answers specific questions with validated data. Start with the decisions you make, implement standard events, validate them against real orders, report by segment and funnel stage, and combine numbers with qualitative evidence. For Shopify's tools, see Shopify analytics; for turning metrics into a view leaders use, see ecommerce KPI dashboard.
For related guides, see customer analytics, ecommerce attribution and ecommerce data warehouse.
Common questions
Collecting and analysing data about how shoppers find, browse and buy from an online store, and about the customers and orders that result, so the business can make better decisions about marketing, merchandising, UX and operations.