The first thing most foreign founders learn about EOR in Belarus is that it works beautifully for the first hire. The second thing — usually 12–18 months later — is that the model that worked for one person doesn’t necessarily scale cleanly to ten. And by the time the team passes twenty-five, the structure that started as the obvious answer often becomes the expensive one.
EOR economics is not static. The cost per employee shifts, the operational overhead distributes differently, and the strategic question of whether to keep using EOR at all changes shape every few hires. Foreign clients who treat EOR as a one-time decision rather than a multi-stage engagement frequently overspend at scale — or, worse, switch to their own entity at the wrong moment and pay double during the transition.
This post walks through four characteristic stages — 1–3 hires, 4–10, 11–25, and 26–50+ — what’s true at each, where the inflection points sit, and how to plan the transitions deliberately. Numbers throughout, so the planning conversation has anchors.
Before the stage-by-stage detail, a conceptual frame: three drivers shift as your Belarus team grows, and the way they intersect is what creates the stage boundaries.
Per-employee EOR cost. Most providers offer volume discounts. Effective per-head cost typically drops 20–40% between 1 employee and 25, mostly through reduced percentage markup and amortized provider overhead. Worth knowing, but not the largest lever in the decision.
Compliance and management overhead. Below ~5 employees, the EOR fully absorbs the overhead — you barely notice that running a Belarus team requires any work on your end. Between 5 and ~15, you start spending real internal time on HR coordination, performance conversations, and benefits decisions even though the EOR continues to run the paperwork. Above 15, internal HR operations become a real line item regardless of structure, and the question shifts from “do we have time?” to “do we have the right structure?”
Tax structure leverage. The Belarusian tax framework rewards scale: HTP residency, FSZN cap mechanics, and accounting fixed costs all amortize across more employees. For foreign IT companies operating in Belarus, the workforce of 115,000+ ICT specialists means hiring depth supports the strategy — but the structural math depends on team size. Below the breakeven (roughly 5 people with material revenue), entity setup doesn’t pay for itself. Above it, the math flips.
The intersection of these three drivers is what creates the stage boundaries below. Each section walks through one stage with its characteristic cost shape and decision profile.
Typical situation. A foreign company hiring its first 1–3 Belarusian developers, usually after sourcing through a global recruiting platform or a referral. The team is too small to justify legal entity setup, and the founder wants to start work in days, not months.
Why EOR is the right answer at this stage. No entity to incorporate, no audit obligations, no business project to write, no supervisory board to satisfy. The EOR hires the employee under its own legal umbrella, runs payroll, handles compliance, and produces a single monthly invoice. Onboarding in 48 hours is realistic with a competent provider. The question of choosing between local and global EOR providers matters more than it might appear — the difference between a strong provider and a weak one shows up first in payroll mechanics and contract quality, then later in everything else.
Cost shape. Total monthly landed cost per developer typically runs $1,800–2,500, depending on salary, benefits, and provider markup. EOR fee on a per-employee basis is highest at this stage — typically 15–25% of gross salary, or a fixed $400–700/month. The percentage hurts at single-employee scale; the absolute dollars are manageable, and the speed-to-hire premium is what you’re paying for.
What to plan for. The most important decision at Stage 1 isn’t operational — it’s documentary. Set up the contract structure with future scale in mind: standard contract templates, equity and options language that works across multiple employees, a benefits framework that’s expandable. Trying to rewrite the foundation at Stage 3 costs real time and can create employee confusion.
Common Stage 1 mistakes. Treating the first hire as a one-off and not documenting decisions. Choosing the cheapest EOR provider without checking what happens at scale — cheap providers often have weak processes that surface as the team grows. Over-customizing the first employee’s contract in ways that don’t generalize to later hires.
Typical situation. The Belarus team has grown from 1–3 to 4–10 people. There’s enough operational pattern to recognize what’s working; there’s not yet enough scale to justify a different structure.
What’s changing. Several things shift at this stage, often simultaneously:
Cost shape. Total monthly per-developer cost: $1,600–2,200, modestly lower than Stage 1. The bigger story is that you’re committing meaningful annual spend — a 6-person team at $2,000/month each is $144,000/year. The decision quality at this stage matters financially, and what felt like a small choice at Stage 1 starts having real consequences.
What to plan for. This is the stage where you should be modeling the entity question explicitly. Run the breakeven math at projected 12-month team size, not current. If the model shows entity makes sense at 10+ people and you’ll hit 10+ within 12 months, start the HTP application now — the lead time is measured in months, not weeks.
Common Stage 2 mistakes. Continuing on autopilot when team economics have shifted. Defaulting to whatever benefits the EOR offers without strategic input. Deferring the entity question until it’s urgent rather than convenient.
Typical situation. The Belarus team is now a real engineering unit — 11 to 25 people, with team leads emerging, multiple specializations, and ongoing recruiting. This is where the math frequently flips toward an entity-based structure, and where most of the consequential decisions in the trajectory get made.
What’s changing, often quickly:
Cost shape and decision math. The breakeven analysis at this stage typically shows entity-based operation saving $150,000–400,000/year vs. continued EOR for the same headcount. The transition cost — incorporating the entity, applying for HTP, restructuring employment — runs $50,000–100,000 one-time. Payback period: under a year in most cases. For most teams, the question stops being “whether” and becomes “when.”
The transition itself. Moving from EOR to your own entity isn’t instantaneous. Plan for 4–6 months end-to-end: entity setup (1–2 months), HTP application through professional advisory (3–6 months in parallel), employment restructuring (1–2 months at the end). Done poorly, employees experience disruption and the cost of mistakes can be significant. Done well, the transition is invisible to them.
Common Stage 3 mistakes. Waiting too long to start the entity process, then trying to compress it under deadline pressure. Underestimating the cost of running EOR for an extra 6 months past the breakeven. Restructuring without planning the employee communication.
Typical situation. A Belarus team of 26–50+ people, ideally already operating through the company’s own HTP-resident entity. If you’re still on EOR at this scale, the transition discussion is overdue.
What characterizes this stage:
When EOR might still make sense at this scale. A small minority of cases: companies expecting to wind down Belarus operations within 12–18 months, or companies with extreme regulatory complexity in the parent jurisdiction where direct ownership of a Belarusian entity creates compliance friction. These are exceptions, not patterns.
The post-scale operational pattern. HTP-resident entity, in-house HR ops, PEO or payroll vendor for the administrative layer, separate recruiting relationships, direct relationships with insurers and benefits providers. The complexity is higher, but the per-employee cost is much lower and the operational control is much greater.
Pulling the trajectory together: the post has implied four stage transitions. 1→2 is mostly automatic — just more of the same model. 2→3 is where the entity question first becomes worth asking. 3→4 is where the entity transition actually happens. Stage 4 is steady-state, with an occasional exit decision.
The two transitions that go wrong most often are 2→3 (treating EOR as default for too long, paying premium markup well past the breakeven) and 3→4 (botching the entity transition itself).
For 2→3: model the math explicitly at the 8-person mark. Don’t wait for the structural reality to demand it. The right answer might be “continue EOR” — but it should be a deliberate decision, not a default.
For 3→4: start the entity setup work 6 months before you need it operational. The HTP application timeline alone is 3–6 months; squeezing it under deadline pressure costs both money and quality. Engaging EOR services support during the transition also helps because the existing provider often runs the parallel administrative work while the new structure comes online.
For both transitions: the cost of being early by 3 months is small — an extra month or two of overhead. The cost of being late by 3 months is often six figures. The asymmetry is what makes the planning matter.
Roughly 10–15 employees, but the answer depends heavily on profit margin and qualifying activity. A 12-person team with 35% margin and clear HTP eligibility typically passes the breakeven cleanly. A 15-person team with thin margins and mixed activity may not. The honest answer is to model the math for your specific case rather than rely on a headcount rule.
Technically yes — many companies do this during transition periods — but the operational complexity isn’t ideal as a steady state. For the same role and similar compensation, having some employees on EOR and others on your entity creates internal pay-equity questions, dual processes for HR ops, and ongoing operational drag. The structural difference between PEO and EOR models is also relevant here: once you have your own entity, PEO is often a cleaner workforce-administration layer than continuing EOR for some employees and direct for others.
Yes, typically. Effective per-employee cost drops 20–40% between 1 and 25 employees, mostly through lower percentage markup. The discount usually plateaus around 20–25 employees — beyond that, the marginal savings from continuing on EOR don’t keep up with the structural savings of moving to your own entity.
Employment relationships transfer to the new entity through a documented process. Done well, the employee sees no break in service, no change in compensation, and continuous benefits. The EOR typically cooperates because their service relationship with you continues during transition; a punctual, clean handover is in their interest as much as yours.
For a typical 20-person developer team, the annual cost difference is usually $200,000–350,000 in favor of operating through your own HTP-resident entity. Belarus’s tax environment in regional context explains part of why the gap is this large — the standard EOR markup on top of a structurally low-cost base compounds quickly at scale.
Typically 4–6 months from decision to operational entity. Entity incorporation: 1–2 months. HTP application running in parallel: 3–6 months. Employment restructuring at the end: 1–2 months. Cutting it shorter is technically possible, but creates risks — particularly around HTP application quality and employee communication.
Yes. Switching EOR providers mid-engagement is common and usually manageable. Most provider transitions take 30–45 days and don’t disrupt employees if handled professionally. The main reasons to switch mid-trajectory are pricing structure (the original provider hasn’t given you the scale discount you’ve earned), capability (the original provider isn’t equipped for HTP-specific work or larger team management), or quality (compliance or HR issues that have accumulated).
Rarely about team size, more about strategic fit. Companies sometimes wind down Belarus operations because the parent strategy shifts or because regulatory friction with the parent jurisdiction becomes operationally difficult. At pure economics, Belarus remains cost-competitive across all scales. Regional comparisons of IT employment costs frame how Belarusian operating costs compare to neighboring jurisdictions — typically still favorable even after EOR markup is factored in.
Send us your projected Belarus headcount over the next 18 months — current state, hiring plan, and target activity. You’ll get back a stage-by-stage cost trajectory, where the inflection points sit for your specific case, and the recommended timing for any transitions. Planning is much cheaper than corrections after the fact.
Dear Clients! We are pleased to announce the start of a project to provide accounting support to a software development team operating in the US market. The number of people in the team is 15 (fifteen). We advised the client on the choice of a taxation system for software developers (IE), and also provide accounting […]
The conversation usually starts the same way. A senior engineer asks for a one-on-one, closes the door, and says they’ve been talking to another company. Sometimes there’s a written offer attached. Sometimes there isn’t. Either way, the next forty-eight hours will decide whether you keep one of the people your team can’t easily replace — […]
Most international groups that own a Belarusian subsidiary aren’t physically in Minsk. The headquarters sits in London, New York, Dubai, Singapore, or Berlin. The CFO checks in over Slack, the board reviews quarterly numbers in a dashboard, and the only people on the ground are the engineering team, an accountant, and whichever statutory officers the […]
