Edit

Forecast with weighted average (preview)

[This article is prerelease documentation and is subject to change.]

A weighted average forecast builds the prediction from a set of historical days that you choose, with a weight assigned to each day. Instead of fitting a trend over a long history, the model blends the intraday patterns of your selected days according to their weights. Use it when you want a forecast grounded in specific comparable days. For example, base the forecast for an upcoming promotion on the pattern of past promotion days. A weighted average scenario produces both volume and average handle time (AHT) intraday patterns, along with confidence intervals.

Important

Forecasting isn't intended for use in making decisions that affect a person's employment. You're responsible for using this feature in compliance with applicable laws, including those that govern access to employee analytics and the monitoring, recording, and storage of communications with end users. For the complete responsible-use notice, see Overview of forecasting.

Important

  • This is a preview feature.
  • Preview features aren’t meant for production use and might have restricted functionality. These features are subject to supplemental terms of use, and are available before an official release so that customers can get early access and provide feedback.

Create a weighted average forecast scenario

  1. In the site map of Copilot Service workspace, select Forecasting under Workforce Management.
  2. On the Active Forecast Scenarios page, select New > Weighted average forecast scenario.
  3. On the Details card, enter a Name, a Forecast start date, a Duration (days), and a Time zone.
  4. On the Historical Data card, the Data source is External and read-only. Select your uploaded file in External data.
  5. On the Configuration card, select the Forecast entity (Conversation or Case).
  6. Select Save.

Note

Weighted average forecasting supports only external data sources during preview.

Adjust the forecast

Open the scenario and select the Adjust Forecast tab to define how the weighted average is built. As you change the source days and weights, the preview chart and summary metrics update.

Set the forecast target

Under Forecast Target, choose the Channel, Queue, and the Apply To date that the pattern applies to. To revert your changes, select Discard Unsaved Changes or Discard All Changes.

Select and weight source days

Under Source Data — Historical Days, select the historical days to base the forecast on. The weighted average across those days produces the forecast. To manage the list, use Add Source Day and Remove Selected, and then choose a weighting mode:

  • Manual: Edit each weight directly. The normalized percentage updates as you type.
  • Recency: Weights more recent days higher automatically.
  • Equal: Gives every selected day the same weight.

Each source-day row shows the Date, Day of Week, Total Volume, an intraday volume pattern and intraday AHT pattern, the Weight, and the Normalized percentage.

Review the preview and confidence

Summary tiles show the number of Source Days, the Weighted Avg Volume, and the 80% Confidence Interval and 95% Confidence Interval.

The Forecast Preview — Intraday Distribution chart plots the weighted average intraday pattern, with confidence bands. Switch between Spline and Line views. Source Comparison shows volume per source day.

  • Last updated on