Every month, a founder shows us a codebase they paid $5,000 for and asks what it would cost to "tidy up." The honest answer is usually: more than building it again.
This is not a hit piece on offshore developers or cheap freelancers. Some of the best engineers we work with live a long way from Israel and charge accordingly. The problem is not geography. The problem is what gets cut to hit $5,000.
What $5,000 buys, and what it does not
A competent freelancer working for $50/hour can deliver about 100 hours of work for $5,000. That is enough time to wire together a Next.js app, a database, and a few CRUD screens. It is not enough time for: meaningful UX research, an actual design system, multi-tenancy, billing, role-based access, observability, tests, accessibility, performance budgets, security hardening, or documentation. So none of that happens.
The compounding cost
You ship. The product is rough but it works. You start selling. Then the cost starts compounding.
- Every new feature costs 3× because the foundation cannot support it.
- Every onboarded customer takes 4 hours of manual work because no admin tools exist.
- Every outage takes 2 days to diagnose because there is no observability.
- Every churn meeting cannot point at why because no events were instrumented.
- Every prospective hire turns down the codebase after their tech interview.
The cheapest engineer is rarely the most expensive part. The cheapest decision usually is.
What good costs
A production-grade v1 of a SaaS — with real architecture, billing, auth, an admin console, instrumentation, and the documentation a future hire can actually onboard from — costs $30,000 to $60,000. That sounds expensive until you realize it is two months of one mid-level engineer's salary, and it compounds for years.
When cheap is the right answer
There are cases where a $5,000 prototype is exactly correct: you are validating a hypothesis, you genuinely intend to throw the code away, and you have the discipline to actually do that. We have built those. But know that is what you are buying.
The question to ask
Before you sign any SaaS development engagement, ask one question: "What is not included that we will need within twelve months?" If the answer is "we will figure it out then" — assume that is going to cost you 5× whatever you are saving today.