💳 by @davehappyminion
Check Subscriptions
Find every recurring charge hitting your accounts. Surface the ones you forgot about, calculate the total damage, and suggest what to cut.
1. Scan Email for Subscription Receipts
Search the user's email for the last 60 days using queries like:
- "receipt" OR "invoice" OR "subscription" OR "renewal" OR "billing"
- "your payment" OR "charge of" OR "auto-renewed"
- Filter to known subscription senders (Apple, Google, Amazon, Spotify, Netflix, Adobe, etc.)
Extract from each match:
- Service name
- Amount charged
- Billing frequency (monthly, annual, weekly)
- Date of last charge
- Email/account associated
2. Scan Bank and Credit Card Statements
If statement data is available (CSV exports, financial tool integrations), scan for:
- Recurring charges with the same merchant and similar amounts
- Charges that repeat on a predictable cycle (monthly, quarterly, annual)
- Small charges that are easy to miss ($1-5 range — often trials that converted)
Cross-reference with email findings to build a complete picture.
3. Categorize Each Subscription
Group every subscription into one of these buckets:
- Active & Used — user clearly uses this regularly
- Active & Questionable — no recent usage signals, or overlaps with another service
- Forgotten / Unknown — no engagement in 60+ days, or user may not recognize it
- Trial Converting Soon — free trial with an upcoming first charge
4. Flag Anomalies
Call out anything unusual:
- Price increases since the last charge
- Duplicate services (e.g., two cloud storage plans, two music services)
- Annual renewals coming up in the next 30 days (chance to cancel before the charge)
- Subscriptions tied to an old email or payment method
5. Present the Report
SUBSCRIPTION AUDIT — [Date]
==================================
MONTHLY BURN: $[total]/mo ($[annual_total]/yr)
SERVICE COST CYCLE LAST CHARGE STATUS
---------------------------------------------------------------
[Service Name] $XX.XX monthly [date] Active & Used
[Service Name] $XX.XX annual [date] Questionable
[Service Name] $XX.XX monthly [date] Forgotten
...
FLAGGED FOR REVIEW:
- [Service] — No usage detected in 90 days. Last charge $X.XX on [date].
- [Service] — Annual renewal of $XXX.XX coming up on [date].
- [Service] — Price increased from $X.XX to $Y.YY since last cycle.
- [Service] + [Service] — Possible overlap (both are [category]).
SUGGESTED CANCELLATIONS:
- [Service] — $X.XX/mo — [reason]
- [Service] — $X.XX/mo — [reason]
POTENTIAL SAVINGS: $[amount]/mo ($[annual]/yr)
6. Deliver the Summary
Send the report to the user via their preferred messaging channel. If annual renewals are approaching within 14 days, mark the message as high priority.
Notes
- NEVER cancel, pause, or modify any subscription automatically. This workflow is read-only. The user decides what to act on.
- NEVER log into banking portals or payment systems. Only use data already available through email, exported statements, or authorized integrations.
- Some subscriptions bill annually — normalize everything to a monthly equivalent for the burn calculation, but show the actual billing cycle in the table.
- If a service can't be identified from the merchant name, flag it as "Unknown" and include the raw merchant string so the user can investigate.
- Re-running this weekly builds a history that makes it easier to spot new charges or price creeps over time.
