Trials¶

Trial-to-paid conversion rates, funnel analysis, and monthly trends from the Tidemill engine.

How Stripe models trials¶

A subscription with a trial period has:

• trial_start — Unix timestamp when the trial begins
• trial_end — Unix timestamp when the trial expires
• status = "trialing" — during the trial window

When the trial ends, Stripe automatically attempts to charge the customer. If payment succeeds, status transitions to "active" (converted). If payment fails or the customer cancels before trial end, the sub moves to "canceled" or "incomplete_expired" (expired/not converted).

Tidemill's trial funnel — cohort-based¶

A trial is attributed to the period of its trial_started event, and its eventual outcome rolls up to that same cohort no matter when the outcome event arrives. A January-cohort trial that converts in March counts toward January's conversion rate — and will update January's number when the March conversion lands.

Metric	Definition
started	Trials whose trial_started fell in the period
converted	…of those, that eventually converted to paid
expired	…of those, that eventually expired without pay
conversion_rate	converted / started

In [7]:

from tidemill import reports
from tidemill.reports.client import TidemillClient

reports.setup()

START, END = "2025-07-01", "2026-04-30"
tm = TidemillClient()

from tidemill import reports from tidemill.reports.client import TidemillClient reports.setup() START, END = "2025-07-01", "2026-04-30" tm = TidemillClient()

1. Overall Trial Conversion Rate¶

In [8]:

rate = tm.trial_rate(START, END)
print(
    f"Overall trial-to-paid conversion rate: {rate:.1%}"
    if rate is not None
    else "No trial data available"
)

rate = tm.trial_rate(START, END) print( f"Overall trial-to-paid conversion rate: {rate:.1%}" if rate is not None else "No trial data available" )

Overall trial-to-paid conversion rate: 96.7%

2. Trial Funnel¶

Breakdown of trial outcomes: started, converted to paid, and expired.

In [9]:

funnel_data = reports.trials.funnel(tm, START, END)
reports.trials.style_funnel(funnel_data)

funnel_data = reports.trials.funnel(tm, START, END) reports.trials.style_funnel(funnel_data)

Out[9]:

Started	Converted	Expired	Conversion Rate
30	29	1	96.7%

In [10]:

reports.trials.plot_funnel(funnel_data)

reports.trials.plot_funnel(funnel_data)

3. Monthly Cohort Conversion Rate¶

One row per started-in-month cohort with the eventual outcomes of its trials. Recent months may still show pending trials that haven't converted or expired yet — their conversion rate will update as those trials reach their terminal state.

In [11]:

trial_df = reports.trials.timeline(tm, START, END)
reports.trials.style_timeline(trial_df)

trial_df = reports.trials.timeline(tm, START, END) reports.trials.style_timeline(trial_df)

Out[11]:

	started	converted	expired	conversion_rate
period
Jul 2025	6	5	1	83%
Aug 2025	4	4	0	100%
Sep 2025	4	4	0	100%
Oct 2025	4	4	0	100%
Nov 2025	2	2	0	100%
Dec 2025	5	5	0	100%
Jan 2026	5	5	0	100%

In [12]:

reports.trials.plot_timeline(trial_df)

reports.trials.plot_timeline(trial_df)

In [ ]: