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

Performance Analytics scores forecasts

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

Performance Analytics scores forecasts

Performance Analytics enables you to forecast future scores based on existing trends. You can forecast scores on Performance Analytics time series widgets and the Analytics Hub. Forecast scores appear as a dotted line.

Forecasting is set up in the Forecasting tab of the indicator record. The number of data points included in the forecast depends on the indicator frequency, and the number of Periods to forecast configured on the indicator. A period is a set number of scores based on the indicator frequency.
Note: If you select a time series aggregation, the forecast is based on the frequency of the aggregation instead of the frequency of the indicator. For example, the 7d running SUM aggregation is a daily frequency, whereas the By week SUM aggregation is a weekly frequency.
Table 1. Forecast periods
Score frequency Number of data points per period Total period length
Daily 7 1 Week
Weekly 13 1 Quarter
Bi-weekly 6 1 Quarter
Four Weekly 13 1 Year
Monthly 12 1 Year
Bi-monthly 6 1 Year
Quarterly 4 1 Year
Fiscal Quarterly 4 1 Year
Half Yearly 2 1 Year
Yearly 4 4 Years
Fiscal Yearly 4 4 Years

Displaying the forecast

To show the forecast on a time series widget, select Show forecast in the Display Settings section of the Widgets form. You can also show the 95% confidence interval of the forecast, by selecting Show forecast range.

To show the forecast on the Analytics Hub, click the chart settings icon (Chart settings icon) and enable the Forecast option.

Forecast methods

Several different methods are available for forecasting Performance Analytics data.

Forecast methods

Method Description
Linear Generates a linear regression forecast based on the historical scores.
Drift The forecasts start as the value of the last score but increase or decrease over time, where the amount of change over time (called the drift) is set as the average change seen in the historical data.
Naive Seasonal Generates a seasonal forecast that is a copy of the previous season of data. This method does not take into account trend data beyond the previous season, such as increasing scores season over season.

A 'season' for this analysis is one period.

Naive Seasonal Drift As Naive Seasonal, the forecast starts as a copy of the previous season of data. The forecast increases or decreases over time, where the amount of change over time (the drift) is set as the average season over season change in the historical data.
Seasonal Trend Loess (STL) Generates a seasonal forecast based on a best-fit function, trend data, and a filter to exclude noise from random variation in the data.

A 'season' for this analysis is one period.

Random Forest (available only for time series) Creates a multitude of decision trees based on the historical data and then outputs the mean prediction of the trees.

Automatic method selection

If the indicator Forecast method used is Auto, the instance evaluates each of the available forecast methods against your historical data to determine the method that generates the best fit trend. This evaluation is performed each time the forecast is displayed, so collecting additional scores can alter which forecast method is used.

To determine the best fit forecast method, the instance generates forecasts using each forecast method with your historical data, then compares those forecasts with the latest data based on how far ahead you want to forecast.

For example, if you configure an indicator with a daily frequency to forecast ahead two periods, the instances generates forecasts using each method for your historical data that is older than two weeks, then compares those forecasts against the latest two weeks of data. The forecast that most closely fits the latest two weeks of data is then recalculated using the entire data set and displayed.

Forecasting and targets

When both forecasting is enabled for an indicator and there is a global target defined, the forecast shows when the target will be reached.

Additionally, the instance sends a notification 14 days before a target is reached. You can control how many days ahead the notification is sent by setting the pa.job.forecast.target.days_to_check property.

This functionality is available only for global targets. Thresholds and personal targets do not interact with forecasts.

Feedback