Back to the subscription dictionary

Trial conversion and the free-to-paid ratio

The free-to-paid ratio compares free and paying customer accounts on one date. Trial conversion instead follows how many accounts from one defined starting cohort become paying within a fixed observation window. Both need a consistent account unit and cannot be used interchangeably.

Prepared with AIContent updated

A snapshot of customer accounts

Assume 10,000 free customer accounts and 1,000 paying customer accounts on the same date. The free-to-paid ratio is 10,000 ÷ 1,000 = 10:1. Both sides count accounts, rather than people on one side and businesses on the other.

Accounts can have been created at different times. The ratio therefore does not identify how many accounts from one particular starting group converted. Users and paid seats are separate populations.

Two different questions

Illustrative, separate datasets; do not combine them into one calculation.

Same date: 10,000 free accounts / 1,000 paying accounts
10:1
Cohort: 25 of 100 trial accounts start paid by day 30 from trial start
25%

Both datasets count unique customer accounts. The ratio is a snapshot; conversion follows each trial account for thirty full days from trial start.

Follow one trial cohort through a full window

For example, follow 100 new trial accounts for thirty full days from each account’s trial start. If 25 start a qualifying paid subscription within their windows, conversion is 25 ÷ 100 × 100 = 25%.

Define the qualifying paid event and handle cancellations consistently. Accounts without the full observation window remain incomplete observations rather than automatic failures. A window starting after trial expiry is a different method and needs its own label.

Interpret with cost and retention

A high free-to-paid ratio can be intentional in a freemium model. High trial conversion can reflect a small, already committed trial audience. Neither measure establishes profitability. Track contribution, CAC and retention in the relevant groups.

Use consistent account and time definitions

State free plan versus time-limited trial, any card requirement, account identifier, observation start and end, and qualifying paid event. Count each account once and document exclusions. Linked user counts do not themselves change the number of accounts.

A rate for every trial start and a rate only for activated accounts have different denominators. Report them separately rather than changing the denominator to improve the apparent rate. Show late conversions as separate follow-up of the original cohort.

Questions and answers

Is 10:1 the same as 10% conversion?

No. The 10:1 example compares 10,000 free accounts with 1,000 paying accounts on one date. Conversion follows the same 100 trial accounts for thirty full days from trial start. Snapshot populations and cohort history are different bases.

What is a good level?

There is no universal target. Compare the same product model, cohort and observation window, and also measure later customer contribution and retention.

Can inactive trial accounts be excluded?

Only in a separately defined measure. A rate for every trial start differs from one only for activated accounts. Document account and activity criteria and do not change the denominator during the measurement.

Sources and documentation

We update our content regularly and do our best to keep it accurate. If you find an error or think something is missing, please tell us. We are always looking for ways to improve the content.

Found an error? Let us know