Subscription Tracking

​

Overview

Money Tracker’s Subscription Tracking feature uses AI-powered analysis to automatically identify recurring transactions in your transaction history. It detects patterns, calculates intervals, and predicts when your next payment is due.

​

How Subscription Detection Works

The system analyzes your transaction history to find recurring patterns:

​

Subscription Candidates

Detected subscriptions are called “candidates” and must meet these criteria:

• Minimum 2 occurrences: At least 2 transactions from the same merchant
• Minimum 65% confidence: Confidence score ≥ 65%
• Same currency: All transactions in the same currency

These thresholds ensure only legitimate recurring charges are displayed.

​

Subscription Card Information

Each subscription shows:

​

Subscription Status

Money Tracker determines status based on the next estimated date:

​

Active Subscription

• Next date is in the future or recent past
• Grace period: Up to 7 days after expected date

​

Inactive Subscription

• Next estimated date has passed by more than the grace period
• No new charges detected

​

Unknown Status

• Insufficient historical data
• Irregular intervals

​

Grace Period

Subscriptions in the grace period show a warning badge: Hover over the warning icon to see grace period details.

​

Filtering and Sorting

​

Filter by Status

​

Sort Options

• By Status (default): Active first, then inactive, then unknown
• By Next Date: Soonest upcoming payment first
• By Merchant Name: Alphabetical order
• By Amount (High to Low): Most expensive subscriptions first
• By Amount (Low to High): Least expensive subscriptions first
• By Confidence: Highest confidence first

​

View Modes

​

List View

​

Grid View

Toggle between views using the buttons in the header.

​

Frequency Types

Money Tracker detects these recurring patterns:

​

Confidence Score Explained

Confidence is calculated based on:

​

High Confidence (75%+)

• Consistent intervals (low standard deviation)
• Stable amounts (minimal variation)
• Many occurrences (5+)
• Always from the same email source

​

Medium Confidence (50-74%)

• Mostly consistent intervals
• Some amount variation
• 3-4 occurrences
• Usually from the same source

​

Low Confidence (under 50%)

• Irregular intervals
• Significant amount changes
• Only 2 occurrences
• Variable email sources

​

Subscription Insights

​

Total Spending

Each subscription card shows:

• Average Amount: Typical charge amount
• Min Amount: Smallest charge detected
• Max Amount: Largest charge detected

This helps identify:

• Price increases over time
• Promotional pricing that ended
• Variable usage charges

​

Interval Analysis

• Average Interval: Mean days between charges
• Interval Standard Deviation: How consistent the billing is

Low standard deviation = predictable billing High standard deviation = irregular charges

​

Common Subscription Types

​

Streaming Services

• Netflix, Spotify, Disney+, etc.
• Usually monthly with consistent amounts
• High confidence scores

​

Software as a Service (SaaS)

• Adobe, Microsoft 365, GitHub, etc.
• Monthly or annual billing
• Very predictable intervals

​

Utilities

• Electric, water, internet, phone
• Monthly billing with variable amounts (usage-based)
• Medium confidence due to amount variation

​

Memberships

• Gym, clubs, professional organizations
• Monthly or annual
• Consistent amounts, high confidence

​

Ride-Sharing / Food Delivery

• Uber, DoorDash, Grubhub
• High frequency but irregular intervals
• Lower confidence, may not appear unless used very regularly

​

Managing Subscriptions

​

Reviewing Subscriptions

​

Identifying Forgotten Subscriptions

Look for:

• Inactive subscriptions you didn’t intentionally cancel
• Services you no longer use but are still paying for
• Duplicate subscriptions (e.g., multiple streaming services)

​

Tracking Price Increases

Compare avg_amount, min_amount, and max_amount to spot:

• Gradual price increases
• Promotional pricing that expired
• Unexpected charges

​

Limitations

​

Not a Subscription Manager

You must cancel subscriptions through the service provider directly.

​

Detection Accuracy

Subscription detection depends on:

• Having sufficient transaction history (at least 2 occurrences)
• Consistent merchant names across transactions
• Regular billing intervals

Irregular or one-time charges won’t be detected as subscriptions.

​

Best Practices

1. Connect your primary email: Use the Gmail account that receives subscription receipts
2. Import historical emails: Run Email Import to analyze past transactions
3. Review regularly: Check the Subscriptions page monthly for new patterns
4. Verify merchant names: Ensure AI extracts consistent merchant names
5. Filter by status: Focus on active subscriptions to track current spending

​

Troubleshooting

​

Subscription Not Detected

If a known subscription isn’t showing:

• Ensure you have at least 2 transactions from that merchant
• Check if merchant name is consistent across transactions
• Verify confidence score is ≥ 65%
• Try running Email Import to get more historical data

​

Wrong Frequency

If frequency is incorrect:

• The AI calculates frequency from actual intervals
• Manual subscription tracking is not currently supported
• Irregular billing (e.g., usage-based) may show as “unknown”

​

Inactive Subscription Still Active

If a subscription shows inactive but you’re still paying:

• Check the last date - there may be a pending charge
• Verify your email is receiving receipts (check Gmail connection)
• Wait for the next billing cycle to update status

​

Related Features

• Transactions - View individual subscription charges
• Gmail Integration - Automatic receipt detection
• AI Extraction - How transactions are identified
• Multi-Currency - Track subscriptions in different currencies