Most developers I know pay for way more subscriptions than they actually use. I knew I was guilty of this too, so last month I did a proper audit: went through every bank statement, categorized every recurring charge, and made a decision on each one.
The result was messier than I expected.
What I Found in Month One
My starting list was 23 active subscriptions. Some were obvious — GitHub Pro, a VPS, Spotify — but others I had completely forgotten about:
- A Figma plan I signed up for during a side project in Q3
- Two different cloud backup services running in parallel (neither fully set up)
- A "lifetime deal" for a tool I used exactly twice
- A podcast app I switched away from when Spotify added their podcast feed
Total monthly spend: $287/month. More than I thought.
The breakdown by category:
| Category | Monthly Cost | Actually Used |
|---|---|---|
| Dev tools (IDEs, hosting, APIs) | $94 | Mostly yes |
| AI/productivity tools | $71 | Mixed |
| Media/entertainment | $52 | Yes |
| Storage/backup | $38 | Partially |
| "Miscellaneous" | $32 | No |
The miscellaneous category was the clearest win: $32/month in services I had no memory of using.
The Audit Process
I did this manually because I wanted to understand why I had each subscription, not just whether I was using it.
Step 1: Statement review. Three months of bank statements, categorized in a spreadsheet. This took about 90 minutes.
Step 2: Usage check. For each service, I checked: last login date, actual usage this month, and whether the workflow would break if I canceled. Some tools you think you use are actually just "nice to have" with no real impact if they disappear.
Step 3: Overlap audit. Several services were solving the same problem. I had both Notion and Obsidian active. I had three tools with AI writing assistance as a side feature. I had GitHub Copilot AND another AI coding tool — useful during evaluation, kept both way too long.
Step 4: Decision matrix. For each subscription: keep / cancel / downgrade / defer (try to use it properly for 30 more days before deciding). I defaulted toward cancel for anything in the "defer" category after the first audit.
What I Cut (And What I Kept)
Cut immediately ($89/month saved):
- Duplicate backup service ($14)
- Figma paid plan → free tier covers what I do ($15)
- Podcast app ($5)
- Lifetime deal tool ($0, but mental overhead cleared)
- Productivity app with no recent usage ($12)
- AI writing tool #2 — Copilot covers this ($15)
- Cloud API subscription I was "planning to use" ($28)
Kept but downgraded ($7/month saved):
- VPS: dropped to smaller instance, will upgrade if needed
Net reduction: $96/month — more than I expected, mostly from tools I had genuinely stopped using but never bothered to cancel.
The Bigger Problem: Subscription Drift
The more interesting finding was how this happened.
Each individual subscription decision made sense at the time. The Figma plan was $15/month during a project where it made sense. The AI writing tool was $15/month when I was writing a lot.
The problem is that these decisions are made individually and never reviewed collectively. Each tool passes the "is this worth $X/month?" check on its own. But when you see the full list, the tradeoffs look different: Am I getting $71/month of value from AI productivity tools? Probably not.
The subscription trap is that canceling any individual one feels like losing something. But the question is whether the aggregate spend makes sense given what you actually use.
What I'd Do Differently
Audit quarterly, not annually. Monthly spend is easy to forget when each payment is small.
Set calendar reminders when signing up for trials. I use a simple note now: tool name, monthly cost, review date (90 days out), and a one-line note on what it's for.
Default to annual billing only for tools I've used for at least 3 months. The savings are real but the lock-in is also real.
Use free tiers more aggressively. Many tools have generous free tiers that cover 80% of what I actually needed.
For the streaming and SaaS subscriptions I kept, I also started checking whether shared or family plans were available. The price differences are bigger than you'd think — Netflix drops from $22.99 to under $7/month on a shared plan, and similar math applies to Spotify, YouTube Premium, and even ChatGPT Plus. I found SubSaver useful for comparing these options across services, and they have a detailed walkthrough of the audit approach that's more structured than what I did here.
The total: $287/month → $191/month after the audit. Not everything that stayed is earning its keep, but at least I now know what I'm paying for and why.
The audit took about three hours total. Worth it.
Top comments (0)