Build vs Subscribe: Personal Software Calculator

Niche workflows no vendor supports: your exact pipeline, your exact data shape, zero feature bloat.
Glue between tools you already own — replacing "integration tier" subscriptions is the highest-ROI build.
Long horizons: build cost is paid once; subscriptions compound forever and get price hikes.
Learning dividend: each app you build makes the next one faster, dropping your effective build hours.

Compliance and audits: HIPAA, SOC 2, payroll, taxes — liability is worth paying a vendor for.
Team software: multi-user auth, permissions, and support tickets multiply maintenance fast.
Anything security-critical (password managers, payments) where a bug is catastrophic.
Constantly-moving targets: apps built on third-party APIs that change monthly quietly rot.

Subscriptions grow linearly; building is an upfront step plus a gentle maintenance slope. Where the lines cross is your break-even month.

Subscribe (cumulative)Build (cumulative)

Does the subscription hold data hostage? If exporting is painful, factor migration hours into the build column.
Is your workflow stable? Build for processes you have repeated for 6+ months, not ones you are still designing.
Count the honest hourly rate: use what your leisure or billable time is actually worth, not aspiration.
Budget maintenance realistically — a common rule of thumb is 15-20% of build effort per year for software you rely on daily.
Start with the smallest app: a single replaced $10/month tool that takes an evening to build pays back in weeks and teaches you the stack.

Subscribe side: seat creep, plan upgrades forced by feature gating, annual price increases of 5-15%, and the "one more tool" tax when apps do not talk to each other.
Build side: hosting ($0-10/month for most personal apps), a domain, backups, and the occasional dependency or API break that eats a weekend.
Both sides: switching costs. Re-entering data and re-learning a UI is real time — amortize it over the years you will use the choice.

Pre-AI, a simple CRUD app was 40-120 hours of work; AI app builders and coding assistants routinely compress that to 2-15 hours for personal-scale tools.
The bottleneck moved from writing code to specifying what you want — which you, the user of your own workflow, know better than any vendor.
Maintenance also shrinks: pasting an error message into an assistant usually produces the fix, so the maintenance-hours slider keeps drifting down each year.

Build It Yourself vs. Keep Subscribing