Time-series

Basic time-series

This is an example of a basic time-series format:

• Each row represents a single variable
• The first column must have the names of each variable
• The first row must have dates for each value

Time-series with dimensions

This is an example of the time-series format with dimensions included. You can include several dimensions. See more on dimensions here.

• Each row represents a single variable, for a single dimension item.
  • For example, row 5 in the example above represents the Website Revenue variable, when the Product dimension is Model X. Together, rows 4 through 6 make up the entire Website Revenue variable.
• The columns can be split into 3 sections: Variable Names, Dimensions, and Values
  • The first column must have the Variable Names
    • In the example above, column A has all the Variable Names
  • The next columns hold the Dimensions - the first row must have the name of the dimension, and the rows below may have the name of an item in that dimension
    • In the example above, columns B and C are the Dimension columns
  • The last columns hold the Values - the first row must have dates, and the values themselves should be numbers (not text)
    • In the example above, columns D, E, F, G, ... are the Values columns

If a variable has more than one dimension, you would just create a row for each possible iteration. For example, if Website Revenue was tracked by both Product and Geography, you would have 12 rows for Website Revenue (3 Products x 4 Geographies).

Time-series with linked dimensions

This is an example of the time-series format with dimensions and mappings (i.e. links between dimensions & dimension items) included. You can include several dimensions and several mappings in your Google Sheet. For more on Linking Dimensions go here.

If you set up the mappings in your Data Source, Causal will automatically know how to slice and dice your variables.

• Each row represents a single variable, for a single dimension item.
  • For example, row 5 in the example above represents the Salary variable, when the Role dimension is Marketing Role. Together, rows 2 onwards make up the entire Salary variable.
• The columns can be split into four sections: Variable Names, Dimensions, Dimension Mappings, and Values
  • The first column must have the Variable Names
    • In the example above, column A has all the Variable Names
  • The next columns hold the Dimensions - the first row must have the name of the dimension, and the rows below may have the name of an item in that dimension
    • In the example above, column B is the Dimensions column
  • The next columns hold the Dimension Mappings
    • The first row must have the name of the dimension being mapped from and the name of the dimension being mapped to, in the following format: [from dimension] > [to dimension]
      • The arrow (">") is important - it's how Causal distinguishes between a normal dimension column, and a dimension mapping column!
    • The rows below may have the name of an item in the dimension being mapped to
    • In the example above, column C is the Dimension Mappings column. It is mapping from Role to City
  • The last columns hold the Values - the first row must have dates, and the values themselves should be numbers (not text)
    • In the example above, columns D, E, ... are the Values columns