The Situation

A fast growing tech company had done what a lot of ambitious companies do — they grew first and figured out the infrastructure later.

Over a few years, they expanded across 15 countries. For a while, the approach worked well enough. They pieced together whatever data they could find — free salary surveys, crowdsourced benchmarks, candidates’ feedback — and made it work.

But by the time they came to us, the years of patching started to show. They had pay bands for some countries and nothing for others. The data they were using was inconsistent at best and unreliable at worst. Free and crowdsourced sources got them this far, but they weren’t giving them the credible, methodology-backed market data they needed to make confident pay decisions. 

The comp review cycle had become a stressful exercise. Leaders were making decisions without consistent data across all countries. Some employees’ pay was, honestly,  just a ‘finger in the air’ guess. And the budget conversation with Finance was harder than it needed to be because nobody could clearly tie salary increase recommendations to actual market movement by country.

What Was Actually Breaking Down

When we got into it, the core issue was data infrastructure — or the lack of it. Since they weren't working from credible compensation survey data, a few things were breaking down:

• Gaps and holes in their pay bands. For the UK, they had IC3 and IC5 pay bands for a certain job family but nothing for IC4. For Brazil, their IC5 midpoint was actually lower than their IC4 midpoint for the same job family. And when your pay bands have gaps like this, every pay decision feels shaky.
• Overpaying in some markets without realizing it. India was a telling one — they'd been using a percentage of US national average data without factoring in real local labor costs, which are way lower than the US or UK. They were overpaying without knowing it.
• No good answer to "where does this number come from?" When managers or employees asked how pay bands were set, the team couldn't give a confident answer. Trust eroded fast and managers and recruiters were second-guessing pay decisions for offers and promotions.

Three separate problems, but all pointed to the same root cause — no credible market data as the foundation, which resulted in every pay decision sitting on top being shaky.

How We Solved It

We approached the fix in two phases:

Phase 1: Align the data to the actual markets.

For the vast majority of their countries, this meant using local market data — and being specific about the cut. For the UK, for example, we used UK national data for most roles, and UK-Inner London data for employees based near London. Sounds like a small detail, but it's not.

For each country, we worked through the right data cuts, the right market percentile target, and where it made sense, applied premiums or discounts based on how the role's scope compared to what's typical in that market. The result was pay bands for every role, across all 15 countries, tied directly to their job families and leveling framework.

Phase 2: Build a defensible methodology for the markets where data didn't exist.

For a handful of smaller countries in emerging markets where even major survey providers had no coverage, we used a geographic differential approach — applying a researched percentage against US national average data, informed by local cost of living and cost of labor trends. 

For North Macedonia, that landed at 40% of the US national average — converted to local currency, with a clear rationale anyone on the team could follow. It wasn't just a number we pulled from thin air or borrowed from a neighboring country. It was a reasoned, documented methodology they could walk any leader, Finance partner, or employee through if needed. And we help revisit it annually as those markets evolve.

The Result

The difference showed up immediately in their comp review cycle.

For the first time, leaders had consistent, credible data points for every employee across all 15 countries — no gaps or ad hoc workarounds, no countries that got skipped because the data wasn't there.Salary increase recommendations were tied to actual market movement by country, which made the budget conversation with Finance straightforward — because the logic behind every number was clear and documented.

Managers completed their pay decisions with confidence. For the first time, the whole cycle ran the same way across every leader and every market, and they wrapped it up in under two weeks.

That's what a solid compensation program feels like when it's actually working.

Takeaways

If your company has grown across multiple countries and you're still working from patched-together data, a few things worth doing now are:

1. Audit your data sources by country. For each country where you have headcount, document what you're actually using and assess whether it's defensible. Flag the markets where you're most exposed and address them before your next comp cycle.
2. Stop relying on free crowdsourced data. Such publicly available sources can be anywhere from 20% to 80% off actual market rates — and because the data isn't validated, you have no way of knowing which direction you're off in. That's a big risk when you're making hiring and pay decisions against competitors who do have the right data.
3. Define your market data approach before you need it. Figuring out your data cut methodology — local data versus a geographic differential — in the middle of a comp cycle is the worst possible time. It breaks trust with managers and employees when they're watching closely.
4. Connect your data to your leveling framework. Market data without leveling is just numbers. Bands need to be built at the level, not the title, so they hold up consistently across every country you operate in.

Set a refresh cadence. Market data can change fast, especially in high-growth or emerging markets. Build in an annual review with a mid-year check for the markets that are moving quickest.

Growing fast across multiple countries is something to be proud of. But pay programs don't scale on their own — and the approach that got you to three countries won't hold up at ten.

The good news? You don't have to start from scratch. You just need a stronger foundation beneath what you already have – better data, a clear methodology, and a consistent process that works the same way in every country so your team can run comp with confidence.