We thought that with 12 companies, there would be enough data for some robust statistics. It turns out there wasn't. Only 4 of the employers in the survey had a large presence in the country; the rest had small sales offices, and some had less than 10 staff in total. Our company had staff over 150, including a regional headquarters and a factory. So you see, a sector-based survey with 12 employers yielded good data for only a handful of positions. My company, along with the others that had larger operations, were unable to use most of the sector data due to lack of matches.

Rule #2 - Look at the Leaders
Leading employers in all sectors usually have a full range of positions, from support to professional to executive. These employers also have a strong employer brand, making them the preferred employers in the market. More often than not, the leaders are multi-national companies or international organisations.

The multi-nationals are known to have disciplined approaches to reward, governed by global principles set down from headquarters. They view compensation and benefits in a strategic way, and know the importance of using market data to determine rates of pay and benefits.

These organisations are usually well-established in smaller developing markets, and attract the top echelon of the workforce. Surprised? One of the reasons is that many international organisations have very competitive pay programmes which are benchmarked not only against each other, but with the private sector as well.

Together, a combination of leading private sector employers and leading international organisations captures the top of the market in many small countries. So it's a good place to start.

Rule #3 - Use Cross-Occupational Job Matching

First of all, there are common occupations in all employers that are easily comparable. For example, positions from accounting, finance, human resources, procurement and IT as well as and less skilled support roles common in developing countries.

For professional and managerial positions, the real challenge is finding enough matches for a particular occupation to be able to report the data separately. In order to ensure that there is data available for each professional level in our surveys, we often double-match positions to both a specific occupational benchmark (e.g., Brand Manager) as well as a generic professional position (e.g., Working Level Professional). In case there are insufficient matches for Brand Manager, we can still report the aggregated data for all positions matched to Working Level Professional. In this way, clients are assured to get a comprehensive picture of the market, even if the specific occupational matches fall short in the survey.

For more on how methodology and approach impacts the reliability of surveys in developing countries, read Warren’s White Paper.

Warren Heaps is a Partner at Birches Group LLC, an HR consultancy based in New York and Manila, specialising in developing market compensation and benefits surveys.