i'm the tech director at a mid-sized company. currently, we outsource all our software work. management says "let's build our own team, it'll be cheaper."
my doubt is this: a two-person team can't do the work an agency does because our needs are very diverse. but a large team would blow the budget.
has anyone made this decision? what did you base your decision on?
I've made this call at a few companies, here's the method: look at how much of the work is ongoing.
If the work is continuous, an in-house team makes sense. The team builds up knowledge, gets to know your system, and speeds up by month six. But a team is a fixed cost; you can't downsize when work slows down.
If the work is sporadic or very varied, an agency makes sense. Agencies scale, bring different expertise, and you only pay for what you use.
In practice, the right answer for most mid-sized companies is a mix: a small in-house team that knows the system and makes decisions, plus project-based support from outside. The in-house team's main job isn't writing code, it's correctly defining what to outsource.
I'm an agency owner, so let me also mention the part that goes against my interest.
1. Some of the work given to us could actually be solved cheaper with an in-house team. Especially small, frequent, and urgent tasks. Running a proposal process for every minor change is both slow and expensive.
2. Where we really add value is on large projects that happen a few times a year and on topics requiring specialized expertise.
Let's be clear: when calculating "it'll be cheaper," people usually forget items other than salary. Hiring time, training period, turnover risk, software licenses, hardware.
Adding from an HR perspective: a two-person software team is a risky size.
When one person leaves, half the team goes, and usually all the knowledge goes with them. In three-to-six-person teams, knowledge gets siloed in one person. If you're willing to take that risk, make documentation mandatory, not just verbal agreements but actual processes.
Finance side: do the comparison based on total annual cost of ownership, not monthly invoices.
For in-house: gross salary, benefits, employer contributions, hardware, licenses, office share, hiring costs. For agency: project fees, maintenance contracts, budget allocated for change requests.
Also add a "cost of not getting work done" line item for both scenarios. The time lost if you build a team but can't keep up is more expensive than the invoice.
From a team lead's perspective: if you're building an in-house team, make your first hire the most senior person, not the cheapest.
Because the first person sets the standard for everyone after them. A bad first hire takes two years to fix.
I'm a small business, let me explain from my side. Calculate how many months you can survive without revenue then halve that number.
There is something to watch out for. like a two-week notebook guides you more than a six-month forecast.
The opposite happened to me, that's why I'm writing. Calculate this based on total annual cost not the monthly bill.
Of course it varies if your situation is different.
My question might sound amateurish, sorry about that... anyway don't enter any agreement without an exit plan.
Hope this helps.
I felt relieved reading this answer, so it's not just me. Taking notes for two weeks yields better results than a six-month estimate.
Hope this helps.
Doki · Mobile app · 2026
I'll argue the opposite don't get mad. When making a decision, also write down the cost of reversal.
Trying to do this alone is the most expensive way. btw im also curious if anyone does it differently.
I'm on watch duty and read the thread. Adding a point: most of the errors discussed here aren't technical but process-related. Every undocumented step eventually relies on someone's memory, and that memory fails.
It's rare to find an explanation this clear. If it's your first time, start small; scaling comes later.
Correct me if I'm wrong.
Quick summary for newcomers: The harder it is to reverse a decision, the slower you should make it.
Just because everyone does it doesn't mean it's right. Just leaving this note, it might be useful.
To get into the details: Don't hesitate to ask; those who don't ask always pay more.
Don't hesitate to ask; those who don't ask always pay more.
The most overlooked point about team is this: The answer varies greatly by industry; there is no one-size-fits-all rule.
Correct me if I'm wrong.
There's one point I'm curious about. Everything goes well for the first three months; problems arise in the fourth.
I'm also curious if anyone does it differently.