vep/scripts/p0/sample_data/README.md

VEP P0 — data templates

Drop real GST exports here as CSV with these exact columns. The harness (run.py)

reads them and emits KILL/KEEP/REVISE. Until files exist (or columns are empty),

the harness reports INSUFFICIENT — it never invents numbers.

signups.csv columns: email, intent_bucket, converted_to_foundation, date cohort_jan2026.csv columns: attendee, bought_ageless, upgraded_tier2 subscribers.csv columns: sub_id, tier, practice_signal, continued, purchase_recency_days capped_clients.csv columns: client_id, revenue_stream, converted_to_scalable

Where to get these (GST-owned data; not available to the assistant):

• signups.csv : export the 190 7-Day Revival signups; tag each with the intent bucket they pick in the H1 test email; mark who bought Foundation.
• cohort_jan2026.csv : from the delivered Movement Lab cohort — who bought Ageless / Tier 2.
• subscribers.csv : export Tier 1/Tier 2 subscribers + any practice/view signal + churn.
• capped_clients.csv : live-class / private-session clients; did they buy video/equipment?

Run: python scripts/p0/run.py --data-dir scripts/p0/sample_data H1_THRESHOLD=0.10 python scripts/p0/run.py # e.g. need >=10% conversion to build