Customer Purchase Frequency (PF)
Purchase Frequency measures how often the average customer buys from you in a period.
Divide total orders by unique customers. That is Purchase Frequency: are customers coming back, and how often? It matters because it is the growth lever most brands forget. There are only three: more customers, bigger orders, more frequent orders.
Signal Type
Metric Role
Metric Type
The Formula
Worked example
Benchmarks and interpretation
Where the number actually comes from
When to use it
Common Mistakes
In the REACT framework
Purchase Frequency sits in the Talk phase. Rhythm makes revenue predictable, and frequency times order value times retained customers is the maths behind LTV. It shows whether you are building habits or processing transactions.
Related metrics
Frequently Asked Questions
What is Customer Purchase Frequency?
Purchase Frequency measures how often the average customer buys from you in a period. It shows whether customers are buying on rhythm or by accident.
How do you calculate Customer Purchase Frequency?
Use the formula: Purchase Frequency = Total Orders Processed ÷ Total Unique Customers. Keep both inputs in the same reporting period and avoid mixing users, sessions, events, or customers unless the formula calls for it.
What data do you need for PF?
You need total orders and unique customers, pulled from the relevant connected sources and computed for the same period.
What mistakes should you avoid with PF?
Do not blend one-time buyers into the base without noticing, and do not buy frequency with discounts so deep that the habit only exists on sale days.
When should marketers use PF?
Use Purchase Frequency when deciding where post-purchase effort goes. It tells you whether the growth lever to pull is repeat behaviour, and which segments already have the habit worth reinforcing.
What is a good PF?
Good depends on category rhythm: groceries and software renew on different clocks. Judge against your natural purchase cycle, and read rises as habit forming, not just promotion response.
Sources and methodology. PF formula and definition derive from standard marketing analytics practice and platform reporting conventions. REACTIQ360 harmonises source data from CRM / billing / ecommerce, Customer lifecycle systems, Harmonisation layer and applies a consistent same-period computation methodology.
One screen.
One story. One move.
One story. One move.
REACTIQ360 harmonises CAC and 129 other digital marketing metrics into one decision-ready intelligence layer. Powered by proprietary metrics and agentic AI. Beta access is invite-only.
Join the beta waiting list
