How much does your infrastructure actually cost at scale?

You're running on a free tier. Or a $20/month plan. Or maybe $200/month with three services. The product works, users are growing, and the bill is manageable. But someone asks: what happens at 1,000 users? At 10,000? And you don't know — because nobody modeled it.

Infrastructure costs don't scale linearly. They have cliffs. A database that costs $15/month at 100 users might cost $400/month at 5,000 because you've hit the connection limit and need a bigger instance. A serverless function that's practically free at low traffic becomes the largest line item at scale because of cold starts and per-invocation pricing. A CDN that you're not even thinking about becomes necessary when your API latency starts losing users.

The problem with "we'll figure it out when we get there"

By the time you get there, the architecture is set. Migrating a database is a project. Switching from serverless to containers is a project. Adding a caching layer after the fact is always harder than designing for it from the start.

The infrastructure cost audit answers the question now, before architecture decisions become architecture constraints.

Eight reports, from topology to recommendations

The /infrastructure-cost-audit pipeline produces eight documents:

Service Topology (Report 01) — Maps every service the application depends on: databases, hosting, CDN, email, authentication, storage, search, monitoring, CI/CD. For each: the dependency relationship, data flow direction, failure mode, and current tier. This is the foundation — you can't model costs without knowing the surface area.

Cost Projections (Report 02) — Per-service pricing across four scaling tiers: MVP (current), 100 users, 1K users, 10K users. Not estimates — actual pricing from the provider's published tiers, factoring in connection limits, storage caps, bandwidth tiers, and per-unit pricing. Shows where the cliffs are.

LOC/Complexity Costing (Report 03) — Correlates codebase size and complexity with development cost. LOC counts by language, complexity scoring by module, estimated development hours, and a replacement cost estimate. This isn't about infrastructure — it's about understanding the investment the code represents.

Technology Roadmap (Report 04) — Stage-by-stage progression from MVP through growth to scale. What infrastructure changes are needed at each stage, why, and the cost delta. This is the migration plan you'll wish you had when you hit a scaling cliff.

Executive Report (Report 05) — Synthesizes reports 01–04 into a narrative with charts. The document you hand to a CTO or investor. Current monthly cost, projected cost at each tier, major cost drivers, and the three infrastructure decisions that matter most.

Technology Requirements Spec (Report 06) — Vendor-agnostic capability requirements. Instead of "use AWS RDS," it specifies "relational database with connection pooling, automated backups, read replicas, and point-in-time recovery." This lets you evaluate alternatives without being locked to a specific provider's product names.

Recommended Stack (Report 07) — The opinionated counterpart to report 06. Given the requirements, here's the specific stack that balances cost, performance, and operational complexity. Includes alternatives for each layer.

Report 00 is the executive summary — key metrics, total monthly cost at each tier, top three cost risks, and the recommended next infrastructure investment.

When to run it

Before committing to a provider. The vendor-agnostic requirements spec (report 06) gives you evaluation criteria that aren't written by a vendor's marketing team.

Before fundraising. Investors want to know your burn rate at scale. Report 05 gives them a credible projection instead of a guess.

When the bill surprises you. If this month's infrastructure cost was higher than expected, the service topology and cost projection reports show you exactly why and what's coming next.

Before a major architecture decision. The roadmap (report 04) shows you the infrastructure consequences of your current architecture at each growth stage.

---

Resources

Pipeline reference: /infrastructure-cost-audit — 8 reports, service topology to recommended stack.

Related reading:

• 03. Documenting projects that don't document themselves — technical specification feeds into infrastructure analysis
• 04. How defensible is your product, really? — cost structure affects competitive positioning

Download: Full toolkit (252KB) — all 16 commands, all 11 skills, installs in 30 seconds.

---

Part of the Claude Skills Library.