Granular Data

What is Granular / Raw Data?

Raw or granular data refers to data that has not been processed or analyzed in any way. It is often collected from various sources in the mobile ecosystem, such as SDKs, APIs or different data stacks. Raw data is usually not very useful on its own, but can be transformed and analyzed to extract valuable insights and information.

When do you need Granular Data?

You need granular data when you wish to calculate metrics yourself or make an investigation into the base events that are being captured by an application. Granular data or raw data enables you to do different analyses because the building blocks of all events are available in the greatest detail possible. Most platforms display aggregated data on their dashboards. This means that the data has been processed and simplified in some way e.g. by taking an average or sum, etc. Although this is useful, raw data is needed to make other types of analyses.

For example, you may know the total gross revenue from purchases made on a certain day on an app-level and be able to see this on a dashboard. However, you cannot find out specific information on that dashboard such as how much the gross revenue was for a particular user who purchased ‘Product XYZ’ on a given date. In this case, the most granular data would show the product_id, currency, gross and net revenue, quantity, purchase timestamp, advertising_id, and the operating system version, etc. This is why Tenjin created DataVault, a data warehousing service that enables customers to carry out investigations into raw data themselves and create custom dashboards.

You will also need raw data to join datasets from 2 different tables or even data stacks. For example, to join IAP revenue data with the install source ad network, you will need access to attribution data with the attributed ad network and User ID, as well as event data that has the purchase timestamps for a particular user.

Why is Granular Data important?

Overall, granular data can be very valuable in a variety of contexts because it allows for a more detailed and nuanced understanding of an event, and can inform more effective decision-making due to thel insights it provides. Raw data is available in its original form for developers to create reports as they need.