An office, not a contractor pool
Opening in Prishtina cost more and moved slower than hiring freelancers would have. It is the reason the team still exists four years later.
When we decided to build capacity outside the Netherlands, the fast option was obvious. Post the roles on a freelance platform, assemble a pool, scale it up and down with demand. It is cheaper on day one and it fails on a horizon of about eighteen months.
What it fails at is accumulation. A contractor pool does not accumulate anything — not domain knowledge, not shared standards, not the informal understanding of why a decision was made two years ago. Every engagement starts from zero, and the client pays for that in ramp time they think they are avoiding.
So we opened a real office in Prishtina, on our own payroll, with employment contracts, an actual lease and the tax and HR overhead that goes with it. The first eight hires took four months, which is roughly four times longer than assembling contractors would have taken.
The thing that made it work was seniority at the start. Hiring experienced people first sets the bar that everyone hired afterwards is measured against. Doing it the other way — juniors first, seniors later to supervise — produces a supervision culture rather than an ownership culture, and it is very hard to reverse.
The second decision was that practice leads sit in Prishtina, not in the Netherlands. If the technical standard is set in the commercial office, the delivery office becomes an order-taker, and you have rebuilt the outsourcing dynamic you were trying to avoid with better paperwork.
Four years on, retention is the number I would point to. Most of that first cohort is still here, some of them now leading practices. A pool would have dispersed by 2023 and taken everything it learned with it.