Multi-country payroll on one engine: why country packs beat custom code

Payroll is the place where most multi-country expansion plans hit their first real wall. Each new country brings statutory tax, social security, leave entitlements, public-holiday calendars, mandatory reports and a local language. The default move is to find a local payroll vendor. Five vendors later, the employee master record has five drift patterns and every compliance report is a reconciliation project.

The right architecture for mid-market is one engine, many country packs.

The custom-code trap

The cheapest-looking option at expansion-day-zero is to extend the existing payroll system with custom code for the new country. The hidden cost shows up at the first statutory change.

• Tax tables update annually. Sometimes mid-year. Custom code does not update itself.
• Social security rules vary across employee types and contract types. Edge cases multiply.
• E-invoicing and statutory filing requirements vary by country and by entity size. A custom integration written in 2024 is broken in 2026.
• Audit trails for payroll are non-optional. Custom code rarely produces them at the level required.

The custom-code option is a series of one-time projects that becomes a permanent maintenance budget.

What a country pack contains

A country pack is the statutory and operational ruleset for one country, maintained by the vendor against actual regulatory changes. DivetIQ's HCM ships country packs covering:

• Statutory tax - income tax brackets, withholding rules, mid-year reconciliations, year-end statements.
• Social security - employer and employee contributions, caps, special regimes for executives, apprentices, expatriates within scope.
• Leave and absence - statutory minimums, public-holiday calendars, sick-leave compensation rules.
• Time and attendance - overtime rules, rest-period requirements, shift premiums.
• Statutory reporting - the monthly, quarterly and annual filings the local tax authority and social-security body require, in the format they require.
• E-invoicing and e-payment - where mandated (Italy's SDI, France's Chorus Pro, Croatia's e-Račun and others), the pack handles the format and the channel.
• Localised payslip - in the local language, in the legally compliant layout.

At launch DivetIQ covers English plus EU primary languages plus regional Balkan languages (BHS, HR, SR, SL, MK), with right-to-left support for Arabic and Hebrew. Multi-currency is refreshed daily from ECB and major exchange-rate sources.

When a country's tax bracket changes, the country pack updates. The customer's payroll runs continue without code changes.

Concurrent runs on one engine

The architectural point is that all country packs run on the same global payroll engine, against the same source-of-truth employee record. The practical implications are concrete.

• Payroll for DE, FR, BA, US in the same period posts to the same general ledger, with the same employee dimensions, on the same calendar.
• Cross-country reporting - global headcount, total cost of workforce, employer tax burden by entity - is one query, not a reconciliation.
• Mobile employees moving between entities keep one employment record. Payroll for the new country picks up the right rules without re-onboarding.
• Workforce planning in EPM uses the same payroll cost model the engine already computes - no separate planning spreadsheet to reconcile.

The decision tree for mid-market companies expanding internationally is short. Does a country pack exist? Use it. Does it not? Ask the vendor when it will. Custom code is the answer only when the answer to the second question is "not in the planning horizon" - and even then, the right pattern is to write the custom logic in a way that becomes a country pack contribution, not a permanent fork.

One engine, many country packs, one employee record. That is the only payroll architecture that survives an expansion plan.