Foundation · 2 min
What you need to know
What customer metrics should ecommerce stores track?
Track new and returning customers, purchase frequency, time between purchases, cohort retention, net sales, contribution, returns and service needs. Use privacy-safe identity rules and state unmatched orders.
How many customers are new vs returning?
Classify a customer as new at the first known qualifying purchase and returning on later qualifying purchases. Reconcile guest, account and cross-device identity limitations before reporting the split.
What is purchase frequency?
Purchase frequency is the average number of qualifying orders per customer during a defined period. It should be interpreted with product buying cycles and customer tenure.
How is purchase frequency calculated?
Divide the number of qualifying orders by the number of unique eligible customers in the same observation window. State how cancellations, full refunds and guest identities are handled.
What is time between purchases?
Time between purchases is the elapsed time between consecutive qualifying orders for the same identified customer. Use a distribution or median as well as an average because long gaps can distort it.
How do I identify my most valuable customers?
Rank customers or cohorts by contribution over a consistent period, then examine repeat behavior, returns, discounts and service costs. Revenue alone can overvalue expensive relationships.
How do I identify customers becoming inactive?
Flag customers whose current gap exceeds the expected product-aware repurchase interval or their own prior pattern. Separate likely inactivity from one-time, seasonal and gift purchasing.
A quick store check
Reconcile customer identity across guest and account orders before reporting new, returning, frequency or value metrics.
Foundation · 2 min
Frequently asked questions
Can guest orders be included in customer analytics?
Yes, if identity matching is reliable and privacy-safe; otherwise report the known limitation.
Is the biggest spender always the most valuable customer?
No. Margin, returns, discounts, service costs and future behavior can change customer value.