Thank you for your feedback.
Form temporarily unavailable. Please try again or contact to submit your comments.

Applying time series aggregations

Log in to subscribe to topics and get notified when content changes.

Applying time series aggregations

You can aggregate changes in indicators into discrete time intervals. A time series aggregation consists of an aggregation, such as AVG or SUM, combined with a time series, such as By quarter. These aggregations can make trends more easily visible, or help track progress against a target. You set up these aggregations either in the Analytics Hub or on Performance Analytics widgets.

While daily indicator scores are foundational and almost always desired, sometimes you also want to have the scores available by week, month, or quarter. In other cases, you want to have a month/quarter/year-to-date number that shows cumulative progress up until the current point. Instead of defining multiple indicator sources and indicators to track each interval, Performance Analytics natively allows you to capture the data once and then adjust the view. Use a time series aggregation for any of these scenarios:

  • Aggregate the data to a less frequent period
  • Smooth the data with a rolling average
  • Determine a period-to-date score

Partial periods

Some time series include indicator scores from incomplete collection periods. These periods can include the current period and the period from the beginning of data collection. A plus sign in the name, +, identifies these time series.

A time series that does not include data from partial periods must have data from the beginning and the end of the period. For example, a By month SUM time series aggregation requires scores from the start and the end of the month to be present. Otherwise that month is not included. A time series that includes data from partial periods, such as By month SUM +, needs only data from one day in the period.

Warning: Partial periods can skew the results of certain aggregations, such as averages.

Default aggregation definitions

Performance Analytics comes with default SUM, AVG, and other time series aggregation definitions. Do not alter these definitions.

Warning: Any changes to aggregation definitions can have unexpected results.

Excluding time series aggregations for an indicator

Some time series aggregations, while technically allowed, are not helpful to apply to an indicator. For example, a SUM of percentage values is unlikely to provide useful insight. Exclude these time series aggregations manually from the indicator. For more information, see Exclude time series from an indicator.

Use cases for time series aggregations

Performance Analytics offers four different types of time series. Understand their use cases to know which type to use.

Table 1. Use cases and examples for each type of time series
Type Examples Use Case

28d running AVG/SUM

30d running AVG/SUM

7d running AVG/SUM

12m running AVG/SUM

3m running AVG/SUM

6m running AVG/SUM

13w running AVG/SUM

4w running AVG/SUM

4q running AVG/SUM

Smooths out spikes in the data to make trends easier to spot. For example, looking at daily incident counts may show a drop every weekend, but a 7-day running average smooths out those drops.
To Date

Fiscal quarter-to-date AVG/SUM

Fiscal year-to-date AVG/SUM

Month-to-date AVG/SUM

Quarter-to-date AVG/SUM

Week-to-date AVG/SUM

Year-to-date AVG/SUM

Shows cumulative scores for the period. These time series aggregations are very useful if you have a monthly target to hit, but you need to also see the velocity throughout the month.
By Period

By week AVG/SUM

By month AVG/SUM

By fiscal quarter AVG/SUM

By quarter AVG/SUM

By fiscal year AVG/SUM

By year AVG/SUM

Shows the cumulative scores for entire periods. While you may want to track the number of P1 incidents daily, the frequency is too low to have a daily target. Instead, you can set a target at the monthly level with a "By Month” time series. The current period will never appear in the results because it is incomplete.
By Period (Including Partial)

By week + AVG/SUM

By month + AVG/SUM

By fiscal quarter + AVG/SUM

By quarter + AVG/SUM

By fiscal year + AVG/SUM

By year + AVG/SUM

The “+” version of the “By” Time Series includes partial periods, so a score is always provided for the current period.

Indicator frequency limitations on time series aggregations

The frequency with which scores are collected for the indicator determines which time series are applicable. Some time series include data from partial collection periods.

When you select a time series aggregation, the frequency with which indicator scores are collected limits which time series you can choose. You cannot select a time series aggregation that is applied to scores more frequently than those scores are collected. For example, the By week SUM time series aggregation can apply to an indicator with a daily frequency. However, By week SUM cannot apply to an indicator with a weekly, monthly, quarterly, or yearly frequency.

  • Only weekly indicators support the 4w running and 13w running time series. Weekly indicators support only weekly and yearly time series.
  • Bi-monthly and yearly indicators do not support any time series aggregations.

The following table shows which time series are supported for which indicator frequencies. These relationships are independent of which aggregation (AVG, SUM, or custom) is combined with a time series, and therefore only the time series are shown.

Table 2. Time series and associated indicator frequencies
Time series Indicator frequencies
Daily Weekly Monthly Quarterly (Fiscal Q, 4-weekly, Bi-weekly)
Daily Yes No No No
7d running Yes No No No
28d running Yes No No No
30d running Yes No No No
4w running No Yes No No
13w running No Yes No No
3m running No No Yes No
6m running No No Yes No
12m running No No Yes No
4q running No No No Yes
By week, by week + Yes No No No
By month, by month + Yes No No No
By quarter, by fiscal quarter, by quarter +, by fiscal quarter + Yes No Yes No
By year, by fiscal year, by year +, by fiscal year + Yes Yes Yes Yes
Week to date Yes No No No
Month to date Yes No No No
Quarter to date, fiscal quarter to date Yes No Yes No
Year to date, fiscal year to date Yes Yes Yes Yes