Active customers
How many customers actually bought from you in a defined window.
01 What it is
Why anyone looks at this number
How many customers actually bought from you in a defined window.
Revenue can hold up while the customer base shrinks, for a while. This is the number that shows it first, and it is the base every retention measure is built on.
02 The formula
How it is worked out
Active customers = count(distinct customers with at least one purchase in the window) grain : rolling 12 months × segment unit : count source : Billing, joined to the customer master
The count depends entirely on customer resolution. The same buyer under three account numbers is three customers until someone matches them, and then the trend line changes shape.
03 Worked example
The same number, with real inputs
| Customers with an order in 12 months | 3,180 |
| Same period last year | 3,340 |
| Calculation | 3,180 vs 3,340 |
| Result | −4.8% |
Revenue grew while the base shrank by 160 customers. Growth is coming from fewer, larger accounts, which is a concentration risk nobody has priced.
04 What moves it
Four things that actually change this number
Acquisition
New customers entering the base.
Churn
Customers who stopped, whether or not they ever said so.
Definition of active
A 12-month window and a 3-month window tell different stories.
Account structure
Group hierarchies, sites and buying entities.
05 Where the number lives
The system, the record and the fields
| System of record | Key record | Fields you need |
|---|---|---|
| Billing, joined to the customer master | Customer, resolved to a golden record | customer_id, parent_id, first_order_date, last_order_date, order_count |
The count depends entirely on customer resolution. The same buyer under three account numbers is three customers until someone matches them, and then the trend line changes shape.
06 How it goes wrong
Three ways this metric misleads people
No customer resolution
Duplicate accounts inflate the count and destroy every retention measure built on it.
Fix: Resolve to a golden record first; everything downstream depends on it.Window chosen to flatter
Widening the window from 6 to 12 months makes the base grow with no change in behaviour.
Fix: Fix the window, publish it, and keep it.Ignoring group structure
Twelve buying sites of one group can look like twelve customers or one.
Fix: Report both entity and group level, deliberately.08 Questions
Frequently asked
What window should define "active"?
Roughly one and a half times your typical purchase interval. If customers normally buy quarterly, a 6-month window is meaningful and a 12-month one hides churn for half a year.
Why does customer resolution matter so much here?
Because every retention, frequency and lifetime-value number divides by this count. Duplicates make churn look worse and value per customer look better than they are.
One definition, everywhere it is used
SCIKIQ stores this metric once and serves it to every dashboard, board pack and agent that asks.