Starting Users
The number of active users at the start of a reporting period; equals the prior period's closing user count.
◆ Count
Formula
Built from
What it measures
The active-user headcount you carry into a period before any of that period's acquisition or churn. It is a pure snapshot of one moment — close of the prior period — not an average and not a live count. Trials, paid, or seats are in scope only if your Total Users definition includes them; Starting Users inherits whatever scope Total Users uses.
Why it matters
Starting Users is the denominator that makes user-motion math honest. Churn rate, retention, and net-new-user growth all divide by the base you opened with, so a wrong Starting Users silently corrupts every one of them. Choosing a period-start snapshot (rather than an average or mid-point) means everyone reading the deck agrees on exactly how many users were "at risk" this period — no ambiguity, and numbers that compare cleanly across cohorts and against peers.
How to read it
Starting Users should never surprise you — it is last period's closing number, locked. Read it as the anchor, then compare it to Total Users at period end: end above start means you grew net, equal means inflows and outflows cancelled (not that nothing happened), below means you shrank. A Starting Users figure that does not match the prior period's Ending Users is a data-integrity flag, not a business signal — chase the reconciliation before you trust any churn or growth rate built on it.
What good looks like
Good
Starting Users matches the prior period's Ending Users exactly, period after period — the carry-forward chain is unbroken and downstream churn and retention can be trusted.
Watch
Starting Users diverges from the prior period's Ending Users without a documented restatement, hinting at backfill, a cutoff-time mismatch, or methodological drift.
Bad
Starting Users disagrees with the prior period's Ending Users chronically or swings unexpectedly with no restatement logic behind it — a calculation error that quietly corrupts every churn and growth figure built on it.
Watch-outs
- Confusing it with Total Users. Total Users is this period's live count; Starting Users is the prior period's close, frozen and read-only. Mix them up and you compare a period to itself.
- Double-counting the base. Starting Users already contains your existing users — never add it into your new-user total or you inflate the base you are measuring against.
- Period misalignment. June's Starting Users is the end of May, not a few days into June and not a rolling average — a sloppy cutoff time shifts every downstream churn number.
- Restating without carrying it forward. If you recompute a prior period's Total Users, the next period's Starting Users must reflect the new figure, or your churn and retention drift away from reality.
Worked example
Hypothetical
You close April with 10,000 Total Users. In May you add 2,000 new users and lose 500 to churn, closing at 11,500. June's Starting Users is therefore 11,500 — April's close is locked, and May's close becomes June's anchor. May's churn rate reads as 500 / 10,000 = 5%, and May's net-new-user growth as 1,500 / 10,000 = 15%.
Variants & windows
The same metric re-expressed by a mechanical transform — a trailing window, a growth rate, a per-unit scaling, or a book/segment cut. Each is computed from Starting Users above.
- Starting Contracted Users Starting · Contracted book
- Starting Live Users Starting