IT Key Metrics Data FAQ

• How is the IT Key Metrics data collected?
• What kinds of companies participate?
• What measures are taken to ensure data quality?
• What time period does the data represent?
• What is included in IT spending?

How is the IT Key Metrics data collected?

Much of the Worldwide IT Trends & Benchmark data is collected online or via surveys that we send out to companies year to year. Our database also includes information collected from Gartner Benchmarking, as well as other Gartner sources. In addition, we do customized surveys for various organizations, and much of that data becomes part of our database. Some financial information (e.g., gross revenue, income) is collected from secondary sources.

Back to Top

What kinds of companies participate?

We have data going back to 1995, when META Group began the Worldwide IT Trends & Benchmark project. (Learn more about Gartner's acquisition of META Group) We have data from 33 countries and more than 24 industry sectors. The total number of company contacts in the database is currently greater than 20,000.

We get the largest response from US companies (usually more than half or our respondents), with Canada coming in second (in the 15%-20% range). We generally receive the most responses from information technology, financial services, and telecommunications companies.

About one-sixth of our respondent companies have gross revenues greater than $40 billion. Slightly more than one-third of our respondent companies have gross revenues between $10 billion and $40 billion. The rest (about half) have gross revenues less than $10 billion.

Back to Top

What measures are taken to ensure data quality?

A certain amount of validation is done during input (e.g., checking for numeric, date ranges, and values that must be less/greater than other input values). The rest of the quality assurance is done prior to posting the data (in graph, spreadsheet, or report form) in the IT Key Metrics database or before it is used in any of our publications.

The following process is used to clean the data:

1. The data is extracted from our server database and loaded into a set of spreadsheets.
2. We check the data to find outliers and any other unusual data. We check for anomalies, various conditions that should hold between 2+ data elements, short strings for names, etc.
3. Records found in Step 2 are flagged, and the data for those records is manually checked. Checking includes looking at the respondents' demographic information (company, title, industry, etc.) to determine the data's potential validity.
4. Invalid records are deleted; valid outliers are moved out of the data set that is used for specific statistical calculations, but saved for other calculations.

Back to Top





What time period does the data represent?

The IT Key Metrics Report is published each fall and represents the data reported during a 12-month time period. After 12 months, data is used for historical purposes. In addition, updates to the research are made during the year, and in these cases, the collection dates are provided.

Back to Top





What is included in IT spending?

IT spending equals the total dollars capital and operational dollars budgeted/spent by IT for information technology to support the environment. This includes all hardware, software, personnel, training, disaster recovery, facilities, and any other costs associated with supporting the IT environment, which includes the data center, desktop/WAN/LAN server, voice and data network, help desk, application development and maintenance, finance, and administration. The IT environment does not include amortization and depreciation costs.

Back to Top