Skip to main content
Beton
posthog_freemium_to_paid posthog_freemium_to_paid_mvp_v1

Freemium-to-paid

Freemium SaaS where the wedge into paid is running up against free-tier limits. Limit-reached events feeding modals and trial starts are the signal chain.

Value metric

Free-tier limit hits that drive paid conversion

Success event

trial_started

entity_type: account

Scale

  • 280 accounts
  • 4 users/account (mean)
  • 8 sessions/user (mean)
  • 30 days of history

Research metrics proxied

  • — Time-to-limit
  • — Limit-hit frequency
  • — Conversion funnel drop-off

Signal paths

Positive paths end in trial_started. Negative paths do not. Every generated event_id belonging to a path is recorded in ground_truth.json.

Positive signals (2)

limit_modal_trial ×80
limit_reached → upgrade_modal_shown → trial_started
cohorts: high_intent, medium_intent, power_user
blocked_feature_trial ×45
feature_blocked → upgrade_modal_shown → trial_started
cohorts: high_intent, medium_intent

Negative signals (2)

limit_modal_dropoff ×60
limit_reached → upgrade_modal_shown
cohorts: medium_intent, low_intent, lurker
repeated_feature_blocks ×40
feature_blocked → feature_blocked
cohorts: low_intent, noisy_bot_like

Generate this dataset

Config file: configs/posthog_freemium_to_paid_mvp.yaml

Quickstart 
# Dockerized Postgres (recommended for inspection)
docker compose up -d

uv run dryfit \
  -c configs/posthog_freemium_to_paid_mvp.yaml \
  --dsn postgresql://dryfit_writer:dryfit_writer@127.0.0.1:54329/dryfit \
  --print-summary

# Or local Postgres
./scripts/generate-local -c configs/posthog_freemium_to_paid_mvp.yaml --print-summary

Full setup instructions are in the repo's README — including local Postgres, Grafana inspection, and dataset restore.

Noise parameters

DryFit injects realistic noise on top of the generated signal paths. These probabilities are per-event. Noise never touches rows referenced by ground_truth.json — your scoring logic can trust the truth file is exact.

missing event probability
5.0%
duplicate event probability
2.0%
out of order probability
3.0%
null property probability
3.0%
anonymous actor probability
2.0%
weird property probability
1.0%

Other scenarios

Combined coverage (all models)

→ upgrade_clicked

Union of event types across all business models

Contact / record-based SaaS

→ contact_created

Contacts, leads, subscribers, accounts managed

Credits / token-based

→ credits_purchased

Credits consumed, tokens used, compute units

Event-volume SaaS

→ custom_event_tracked

Events tracked, data points ingested, log lines
Browse all scenarios →

Benchmark your detector against Freemium-to-paid

Clone the repo, run the config, check your agent's output against ground_truth.json.

View on GitHub