International Salary Benchmarking: How to Set Pay Across Countries

International salary benchmarking is how a company that employs people in several countries checks that each role and level is paid in line with the local market where the work is done, under one consistent global framework. Done well, it gives you local ranges that are competitive, a structure that employees in different countries can understand, and evidence you can defend once pay transparency laws apply.

The most common shortcut, taking headquarters salary data and scaling it with a cost-of-living index, gets the answer wrong for many roles. TalentUp’s analysis of about 15,500 job postings with a structured salary published in 2026 shows that the same country can sit at 52% of German pay for one role and 98% for another. A single “France factor” or “Spain factor” cannot capture that.

This playbook walks through what multinational compensation teams actually need: a global job architecture, market data chosen country by country, a clear currency and employer-cost policy, pay positioning and geo-pay tiers, and governance that holds up under the EU Pay Transparency Directive.

Why HQ Data Plus Cost of Living Fails

Most companies start international pay the same way: take the salary bands from the headquarters country and multiply them by a cost-of-living ratio for each new location. It is fast, easy to explain and systematically inaccurate, for three reasons.

• Cost of living is not cost of labour. A living-cost index measures rents, food and services. Salaries are set by supply and demand for a specific skill in a specific labour market: how many engineers, accountants or sales managers are available, which employers compete for them and what collective agreements set.
• The ratio is not constant across roles. A country can be cheap for one job family and expensive for another. Our data below shows France close to German pay for sales management but at about half of German pay for accountants.
• Errors compound with headcount. If you overpay in one market you burn budget; if you underpay you lose offers and create pay gaps that are hard to explain when employees, works councils or regulators start asking.

The alternative is not to drop structure. It is to benchmark each country against its own market, inside one global framework that keeps roles, levels and policies consistent. If you are weighing the opposite model, paying the same everywhere, see our analysis of location-based pay versus the same salary everywhere.

TalentUp Data: The Same Role Across Eight Countries

To show how much the country gap depends on the role, we compared the median posted annual salary for four common roles in eight European countries, using job postings with a structured salary from January to September 2026, all converted to euros. Germany, a frequent headquarters market, is set at 100.

Source: TalentUp job postings database, January–September 2026. Median posted annual salary, EUR.

Three patterns matter for anyone designing a global pay structure:

1. Country gaps are role-specific. Spain is consistently around half of German pay, but France moves from 52 (Accountant) to 98 (Sales Manager) and Italy from 50 to 75. A single location factor for France would misprice at least one of these roles by more than 20 points.
2. “Expensive” and “cheap” markets are not a fixed ranking. Ireland is above Germany for three roles, and the UK is below Germany for all four once pounds are converted to euros.
3. The headquarters market can have the thinnest data. Germany produced only about 16,300 salaried postings in the period, against roughly 102,200 in Spain and 180,100 in the UK, because German employers publish pay less often. Relying only on HQ data can mean relying on your weakest sample.

Spread Within a Country Can Beat the Gap Between Countries

The median is only one point of the distribution. For Account Managers, the range between the 25th and 75th percentile inside one country is wider than the gap between some countries’ medians:

The Dutch interquartile range (€47,900) is larger than the gap between the Dutch and Spanish medians (€27,900), and the German and Dutch medians are only €2,600 apart. Level, scope, industry and company size often move pay more than the border does, which is why the next step, levelling, comes before any country comparison.

Step 1: Build a Global Job Architecture First

You cannot compare “a Project Manager in Madrid” with “a Project Manager in Munich” until you know they are the same job at the same level. Titles drift between countries: a “manager” in one subsidiary leads a team, in another it is a senior individual contributor.

A global job architecture fixes this with three building blocks:

• Job families and sub-families (Finance > Accounting, Sales > Account Management) that are identical in every country.
• Career tracks, typically individual contributor and management, so a senior specialist is not matched with a people manager.
• Global levels defined by scope, impact, autonomy and expertise, written once and applied everywhere, for example six to eight levels per track.

Then match each local job to the architecture, not to its title. Match against the level description and the actual responsibilities, and record why. That record becomes the objective, gender-neutral criteria you need later for pay transparency. Our guide on how to create salary bands covers levelling and band design step by step.

Step 2: Choose Market Data Country by Country

The best data source in one country is not necessarily the best in another. For each market, check four things before you trust a figure:

Two practical rules help. First, when a cell is thin, widen the job family or level rather than dropping it into a precise-looking figure. Second, look at the spread, not just the median: a wide P25–P75 range tells you the title covers very different jobs, as with the Account Manager example above. For a deeper look at percentiles and data sources, read our step-by-step guide to salary benchmarking for pay transparency. TalentUp’s methodology explains how our own data is collected and cleaned.

Step 3: Compare Like With Like: Currency, Gross Pay and Employer Cost

Most cross-country errors are not about the market; they are about comparing different things. Agree three definitions before anyone builds a range.

Base Salary, Total Cash and Employer Cost

The difference between these measures is not the same in every country. According to Eurostat, non-wage costs, mostly employer social contributions, made up 24.8% of total labour costs in the EU in 2025, but 32.3% in France and 4.8% in Romania, and average hourly labour costs ranged from €12.0 in Bulgaria to €56.8 in Luxembourg. Two countries with the same gross salary can therefore have very different employer costs. Benchmark gross base and total cash against the market; use employer cost only for budgeting. For a worked example outside the EU, see our salary comparison of the USA and the UK, which covers payroll taxes and pensions, and our overview of US vs Europe salary differences.

Currency and FX Policy

• Pay in local currency and benchmark in local currency; convert only for reporting and budgeting.
• Fix one reference rate per cycle, for example the average of ECB euro reference rates over the three or twelve months before your salary review, and document it.
• Set a tolerance band (for example, review the rate only if it moves more than 5–10%), so currency swings do not trigger pay changes on their own.
• Show the FX effect separately in reports: in our data, the UK index moves with the GBP/EUR rate even if pounds salaries do not change.

Step 4: Set Pay Positioning and Geo-Pay Tiers

With matched jobs and clean market data, decide where you want to pay in each market. This belongs in your compensation philosophy.

Our data adds one warning for geo-pay tiers: a country that fits a tier for one job family may not fit it for another. France would sit near Germany for Sales Managers but well below it for Accountants. Build tiers per job family, or at least test them against role-level data, instead of assigning one tier per country.

Remote Employees

For remote workers, the default is to pay for the location where the employee lives and works, because employment law, tax and, as a rule, social security follow the place of work in the EU. Write down what happens when an employee moves country: whether pay is re-benchmarked, from what date, and whether it can go down. Our article on how remote work is reshaping compensation covers remote premiums and benefits in more detail.

Step 5: Govern It and Prepare for the EU Directive

A multi-country pay structure stays credible only if someone owns it. A simple governance model includes:

• Ownership: global Compensation & Benefits owns the job architecture, data sources, FX rate and positioning; local HR owns job matching and market context.
• Approval rules: offers and increases inside the range are approved locally; anything above the range, or a change of level, goes to global C&B with a written reason.
• Review cadence: refresh market data at least once a year before the salary review cycle, and quarterly for hot roles or high-inflation markets.
• Audit trail: keep the data source, collection date, percentile, FX rate and job match behind every range.

Anand Chandrasekaran, an India-based HR business partner (KPMG, Cognizant) who writes about global delivery teams, explains why that audit trail matters once colleagues in different countries can compare notes:

What Changes With the EU Pay Transparency Directive

Directive (EU) 2023/970 had to be transposed by member states by 7 June 2026, and its obligations apply through each country’s national law once transposed; several countries are still finalising their laws. For multinationals, the key points are:

• It works country by country. Each employer applies the national law of the member state where it operates, so a group with entities in five EU countries can face five slightly different sets of rules and reporting formats.
• Pay structures need objective, gender-neutral criteria, such as skills, effort, responsibility and working conditions. Your global job architecture and country-specific market data are the natural evidence.
• Reporting is phased: employers with 250+ workers report every year from 7 June 2027; 150–249 workers every three years from 7 June 2027; 100–149 workers every three years from 7 June 2031.
• Unexplained gaps have consequences: if reporting shows a gender pay gap of 5% or more in a category of workers that is not justified by objective criteria and not remedied within six months, a joint pay assessment with workers’ representatives is required. In pay discrimination cases, the burden of proof shifts to the employer.

Country-specific market data can help explain why pay differs, but only if it is applied in the same way to women and men in the same category. Before your first reporting window, run a pay equity audit by worker category, and use the free pay transparency checker to test your ranges. For market context by country, see the European Salary Benchmark Report, or compare plans on our pricing page.

Sources

• TalentUp job postings database, January–September 2026: median posted annual salaries for Accountant, Project Manager, Account Manager and Sales Manager in eight European countries (about 15,500 postings; see methodology above).
• Eurostat, EU hourly labour costs ranged from €12 to €57 in 2025 (31 March 2026).
• EUR-Lex, Directive (EU) 2023/970 on pay transparency.
• EUR-Lex, Regulation (EC) No 883/2004 on the coordination of social security systems.
• European Central Bank, Euro foreign exchange reference rates.