Build, buy, or borrow a team: a founder's framework for getting software made
There are really only three ways to get software built, and most founders only seriously consider one of them. A clear-eyed look at the trade-offs of each.
Summarize this article with AI
ChatGPT
Claude
PerplexityKey takeaways
- There are three ways to get software built: build in-house, buy off the shelf, or borrow a team.
- Buy the generic, build the core once it's stable, borrow a team when it's urgent and still evolving.
- The mistake is committing to the most expensive, least reversible option before you've learned anything.
When a company needs software it doesn't have, it has three options: build it in-house, buy something off the shelf, or borrow a team to build it for you. Most founders default to one and never seriously price the others. Here's the honest version of each.
Buy: the off-the-shelf SaaS
Fastest and cheapest, when it fits. If a $50-a-month tool does the job, buy it and move on. The trap is the slow accumulation: five tools that almost fit, none that talk to each other, and a monthly bill that quietly passes the cost of just building the thing. Buy for the generic. Don't buy for the part that's actually your business.
Build: the in-house hire
The right end state when software is your core product and your roadmap is long. The wrong starting point when you can't yet evaluate the hire, the scope is still moving, and you need it built this quarter. It's the most expensive and least reversible option, and it's the one founders reach for first, usually too early.
Borrow: a fractional team
A senior team on retainer instead of payroll. You get breadth and speed without the hire, you scale capacity to the project, and you keep your equity and your optionality. The trade is that it isn't permanent, which is precisely why it fits the stage before you know what permanent should look like.
A simple way to choose
- Generic and non-core? Buy it.
- Core, urgent, and still evolving? Borrow a team.
- Core, stable, and your long-term obsession? Build it in-house, eventually, with what you learned from borrowing.
Fit for an urgent, still-evolving core build
Scored 1 to 5 when the work is core, time-sensitive, and still changing
These aren't permanent camps. Smart companies borrow to get moving, buy what's generic, and hire once the role is obvious. The mistake is committing to the most expensive option before you've learned anything.
If you're staring at a build decision and the honest answer is 'urgent, core, and I'm not certain of the shape yet,' that's a borrow. Get it built, learn from real usage, and let that evidence tell you whether and when to bring it in-house.
