E-MAIL | PRINT | SAVE | LICENSE   

By David Gudgin

The availability of so much information on the Internet, free at the point  of use, has led the public into a false sense of security, believing that any piece of information we seek is just a click away.  This of course undermines user appreciation of the investment that goes into paid-for information services provided by market intelligence specialists.

So how does such information differ from information found through free sources such as national statistical offices, trade departments and free business information websites? And why do banks, consultancies, manufacturers or government departments invest in research rather than relying on free sources?

Businesses need to trust the research that they are using.  This research is often used to support investment and strategic planning decisions and needs to be trustworthy.  Key questions to ask of your information sources are:

• What are the data sources? Are they reliable?
• Are categories and methodologies clearly defined?
• Can data be compared across countries?
• Why are there gaps in the data?  Can they be filled?

To answer these questions, let's investigate how Euromonitor International conducts its macro-economic research.

Quantifying the source data

Euromonitor International's starting point includes the usual suspects: international sources such as the International Monetary Fund, World Bank, United Nations, Organisation for Economic Co-operation and Development, World Trade Organisation, World Economic Forum and International Labour Organisation, to name but a few - as well as national sources including national statistical offices, central banks and government departments and ministries.

The meaning behind this enormous amount of data is often lost. Data varies greatly from country to country, as different methodologies and definitions alter its comparability and impact. Following years of experience working with official sources, we are also aware that the data supplied by official sources from some countries can be more dependable than others.  

This lack of dependability brings into question the reliability of the information and the level to which it is cross-country comparable. Even countries assumed to have reliable information houses often have to make serious revisions to their data.

After the last census, France was forced to significantly revise its population data upwards, as it had previously underestimated the birth rate. As such, the population forecast for 2050 has increased from 64 million to 75 million - which pushes the country past Germany in the number of inhabitants.

In order to manage these issues, all of the information collated by Euromonitor International undergoes a series of rigorous tests. Firstly, we evaluate the quality of the research by cross checking all sources for each country. The research conducted at a country level is then checked at a global level for accuracy and to ensure all trends have been identified. This is just the first step in confirming that trends and forecasts are the result of good, quality data.

Data standardisation

The next step is data standardisation. Research parameters and industry categories and definitions vary vastly from country to country. Euromonitor researchers must standardise each category, ensuring cross-country comparability. For example, figures on disposable income are readily available, but it is impossible to properly compare these figures on a country by country basis without knowing what is included in the figures. Do they include income from benefits and personal investments as well as earnings?

After standardisation, the resulting Euromonitor data set and definition is the same across every country, allowing for accurate, and meaningful, comparisons between regions.  This is particularly useful for marketers when using population data.

Population by age is published by national statistical offices in various brackets, some as large as ten years (eg. 0-10, 11-20, 21-30...) or others by two years (eg. 0-2, 3-5, 6-8...).  This makes direct population by age comparisons impossible.  At Euromonitor, data for population by age is broken down into single year brackets across all countries. The data available is as granular as possible, enabling the greatest level of evaluation and analysis. 

A children's footwear manufacturer, for example, can determine the number of five year old boys every year for the next 20 years in a country or countries.  This information will determine new product development, production capacity and marketing strategies for years to come.

 Male population aged 5: January 1st  (thousands)

Source: Euromonitor International 2007

Filling in the gaps

During the research process, Euromonitor often finds significant gaps in the types of available information. Whether it stems from a lack of data gathering infrastructure by developing countries or changes in research methodology across a number of years, these types of inconsistencies and gaps can lead to erroneous assumptions and poorly informed decision making.

To combat this, the next critical process in Euromonitor's research is estimation and modelling. Using advanced extrapolation techniques, we produce ‘better' data than the original source.

Our estimation techniques are an important tool for adding value to research. Illustrated in this population graph for example, you can see how Euromonitor's figures on the population in the Czech Republic differ from those originally published by its own National Statistical Office (NSO).

Changes in the way that this information was calculated by the NSO, to properly account for migration, can be seen by the sudden jumps on the graph. The NSO, however, has never re-calculated its old figures using its improved methodology, therefore making the earlier data unreliable. The Euromonitor line is smoother and more accurate, as we have taken these changes in methodology into account, and re-calculated a more realistic trend.

Modelling data sets is also vital when providing a clear and comparable picture of a country or market. Euromonitor's methodology for modelling important statistical information involves examining leading indicators for a country or market.

To illustrate, let's say that there was no data available on French disposable income levels. We would be able to create these figures by using well established modelling techniques. In order to calculate this information, we would look at all factors surrounding and affecting income, such as the employment rate, taxation rates, economic growth, family size, population changes and so on.

Euromonitor would then build a model for how these factors affect disposable income in ‘similar' countries where disposable income data is available - in this case Italy and Germany.  This model would then be used to calculate the French disposable income levels. These modelled data sets fill in the gaps and complete the research so that the result is a well rounded, accurate picture.

It is not unusual for Euromonitor data to look quite different when compared to the original source after the rigorous validation that has taken place. The original source will have been cross-checked against other sources to ensure accuracy and to tease out discrepancies. It will then have been standardised so that the same dataset is comparable across countries. Finally, any gaps will have been identified and rectified using complex estimation and modelling techniques. 

And that original source we're talking about might even have been found on the Internet for free.

By David Gudgin

David Gudgin has almost 20 years' experience in the information industry.  He has worked for several publishers ranging from scientific to business information as well as in the conference and exhibition industry.  He is currently Sales and Marketing Director at market intelligence publisher Euromonitor International plc.

David has been instrumental in developing Euromonitor's portfolio of electronic business reference products, including the Global Market Information Database which won ‘Best Business Information Product' at the International Information Industry Awards in 2005. Most recently, he has worked on the launch of Countries & Consumers, the company's latest online information database, which offers expert analysis of country and consumer trends.

FUMSI articles by David Gudgin »

Click here for copyright permissions!
Copyright 2010 Free Pint Limited

Related articles:

• Statistics: Can You Really Believe the Figures?
• Information Provision: Adding Value
• Finding Competitive Information for Growing Companies
• Do We Have to Work Harder in a Downturn?

You may also be interested in:

• Subscribing to FUMSI »
  
  
• Browse the articles database for more Use articles »
  
  
• Search the articles database for more Use articles »
  
  
• Tools and reports for Using Information »