Data Format

The constant product innovation leads to changes in the structure of the data products. For example, with the rise of AI enabled laptops IDC has added new fields into the Personal Device Tracker to provide the key information to IDC customers. In other cases some information that is no longer relevant is dropped or replaced.

This results not only into updates in the structure of IDC data products but it is also cascaded into the structure of the data delivered through REST API and SFTP integrations.

IDC will notify all the clients ahead of time if there are any changes in the structure of the products. To make sure you will receive the notification make sure your team is registered to receive the notifications. The process is defined in Staying up to Date with Recent Changes.

Data Product: An IDC offering consisting of structured data sets (as opposed to written research). Examples: Trackers, Spending Guides, Forecasts, etc. A data product typically has a defined schema (dimensions/measures) and a release schedule.

Data Set: Sometimes used interchangeably with product or part of a product. In IDC exports, a data product might encompass multiple data sets (e.g., one for historical data, one for forecast, or different segments). The API’s JSON may refer to a dataset name for a specific piece of a product.

Data Cut: A one-time delivery of data, outside the ongoing subscription stream. For example, a custom report or a special snapshot not regularly updated. (In contrast, a Subscription implies you get continuous updates).

Dimension: A categorical descriptor in a data set – essentially a column that represents an attribute of the data. Dimensions in IDC trackers might include things like Geography, Vendor, Product Category, Year, Quarter, Customer Segment, etc. Dimensions are used for grouping/filtering data.

Measure: A quantitative metric in the data – essentially a column representing a value being measured. Measures in IDC data often include Units, Value (Revenue), Spend, Shipments, etc. Measures are typically numeric.

• The files have .csv.gz extension, meaning they are GZIP archives containing a CSV file. You will need to decompress them after download to get the CSV. The compression significantly reduces file size for transfer (often by ~90%).
• The CSV inside is typically comma-separated (standard CSV) with the first row as column headers (captions). Those headers correspond to dimensions and measures.
• The data is usually in a flat table form. Each row represents a specific combination of dimension values (e.g., a specific country/quarter/product segment) and the measure values for that combination.
• Character encoding is usually UTF-8 (so it can handle international characters).
• Dataset may be split into multiple CSV files by IDC (hence multiple .gz files) – you will then need to handle them separately. IDC uses logical splits (like by segment, channel, etc.).

There is a GZIP archive for each product: Wallet: Spend by Market and Wallet: Spend by Tech Provider.

Each archive contains a set of CSV files that represent a full accessible dataset of a respective product.

Tech Spend by Technology columns: