Ask Runway AI Copilot

Modeling with Data Sources

Using data sources to fuel your model allows for building an interconnected financial model using Runway databases.

Runway databases can be connected with data sources (integrations, or other Runway databases) to create a live link to your source systems and enable you to model at multiple levels of granularity.

Prerequisite: If you’re interested in using integration data as a source and you need to set up your integrations, please refer to our article on Integrating and querying your data.

 

Video Overview

The process of connecting your model to data sources is quite visual. If you prefer to watch an example, there are two Looms below that cover each source type: Integration data, and Runway databases

 

Loom 1: Connecting integration data to a Runway database, with a spoken overview of how each configuration option works.

 
 

Loom 2: Aggregating one Runway database into another Runway database for easy aggregation. This pattern allows you to model at multiple levels of granularity with straightforward drill in/out.

 

Configuration Options

An exploded view of the options when modeling from a data source and a brief description is below, these are covered in context in the above workflow Looms:

Notion image
 
 
 

Below is a written overview of the concepts explained in the above workflow Looms.

Example setup

Let's say you're managing finances for a multi-department company. You can use the workflows demonstrated above to build multiple databases for the various levels of granularity you want to see your financial data at.

Database 1: Vendor-level Database — the most granular view you have in Runway

  • Segmented by Vendor, GL Account, and Department
  • Data Source: a Quickbooks query’s results from an integration query

Database 2: Department x GL aggregates

  • Overview of accounts by Department and GL accounts (summing across all vendors).
  • These databases could be used for soliciting departmental inputs from each of your Department leads.
  • Data Source: Database 1, segmented by GL and Department

Databases 3 & 4: Topline numbers by Department; Topline numbers by GL Account

  • High-level view of key financial metrics
  • Ideal for executive reporting and strategic planning
  • DB 3 would be segmented only by Department, and DB 4 just by GL Account
    • Data Source: Database 2
 

With this setup, you’ll easily be able to see data bubble up and down your model while being able to drill into the granularity that matters for the right stakeholders.

 

Configuration Options

When using a data source to fuel your Runway database, you have access to the following configuration options to prepare the source data for modeling:

  1. Data Source You can link Runway databases to data sources — either the results of an integration query, or another Runway database.
      • You have the ability to preview and filter down the data source, for example if there’s only a subset of the source you’re interested in modeling.
      • A common use case for this is to define two different databases for Expenses and Income in Runway, e.g. Expenses segmented by Vendor, and Income segmented by Customer
  1. Drivers: Select the columns from the source you want to use as drivers — these typically are the metrics you want to model over time, like Expense, and ARR.
      • For integration data, you’ll also have the option to select a time column from the data source to define when in time your driver values should be modeled.
  1. Segment by: You can choose dimension columns from your source to define the granularity you’d like to model your source data at. This helps with aggregation and defining unique segments to model at.
      • For example, segmenting Quickbooks transaction data by the GL Account and Class dimensions is a frequent use case.
      • Choosing those dimensions as your “Segment by” dimensions will result in a database row for each unique combination of GL Account and Class
  1. Additional Columns: If there are additional “metadata” columns from your source you’d like to pull in, you can do so.
      • Example: metadata like Account Type and Account Subtype can be helpful pieces of information to bring in to Runway to help with filtering and grouping.
  1. Confirm your configuration
      • When you confirm, Runway will aggregate and structure your source data according to the configuration, which may take a few seconds.
      • You'll see the source data now linked to your model
      • As new records enter the source, your database will automatically update:
 

By linking these databases to your Runway model, you create a comprehensive financial modeling ecosystem that provides insights at various levels of detail while maintaining consistency across all views.

  • If your source is a Runway Database, the target database will update at the time of the edit to the source.
  • If your source is Integration data, the target database will update daily by default when the query runs.
Did this answer your question?
😞
😐
🤩

Last updated on September 16, 2024