Create a long-term forecast

The long-term forecast is a prediction of the future resource need on day level. It is used to give a rough estimate for the resource need over a longer period, for example the next 6–12 months. The long-term forecast is used as input when creating a long-term staffing budget, and to create a more detailed forecast later.

The long-term forecast is based on historical data and identifies seasonal variations as well as variations within the months and weeks. It does not have any information on the distribution within the day.

NOTE   To create a long-term forecast where you can control all details, it must be created separately for each workload. Use the Quick forecast to create long-term forecasts for several skills and workloads at a time. See Create forecasts for several workloads for more information.

Select the periods of data to base your long-term forecast on and refine the seasonality patterns for different time periods. Take notes on the periods selected for follow-up on what you have based your forecast on.

The index values represent how a time period compares to the average.

EXAMPLE   If the index value for the number of calls during a month is 1, this means that this month has the average number of calls. If the value is 0.9, this month has 90% of the calls of an average month. If the value is 1.1, this month has 110% of the calls of an average month.

The index value for each month, week and day is used to forecast for that month, week and day in the future.

EXAMPLE   The index for March is 1.1, the index for the first week of the month is 1.2 and the index for Fridays is 0.7. The forecast for the first Friday in March is calculated by multiplying these index values with the average daily volume, which is 1500. The calculation is 1.1 x 1.2 x 0.7 x 1500 = 1386.

Prerequisites

  • You have the Forecasts permission.
  • A skill and a workload are created.
  • There is at least one year of validated historical data to base the forecast on.

Page location

Client > Forecasts > Prepare workload > Long-term forecast tab

Procedures

Select historical data to base the long-term forecast on

Select one or more periods of historical data to base the long-term forecast on. The periods that are selected on the Data summary tab are used as the suggested selected periods for the seasonality patterns.

  1. Select the Data summary tab.
  2. In the Select historical data area, click the symbols at the top to select which way to add historical data periods.
    • Define a number of days, weeks, months or years back from the current date.
    • Define a period, with a start date and an end date.
    • Define individual dates.
  3. Define a period and select Add.

    You can add more than one period to base the forecast on. Ensure to select the data that is most relevant for the period that you are about to forecast for. Review the data. If you see any extreme values, go back to the Validation tab and adjust. Then reapply the period.

  4. Click Apply to show the data for the selected periods in the chart.

    If you want to remove one of the selected periods, select it and click Delete. If you want to remove all selected periods and start over, click Clear. Then click Apply again to see the result.

  5. Click Next to refine the variations within the year.

Refine the variations within the year

Review the variations within the year and adjust if needed. The data shown in the chart is based on the previously selected historical data. Adjust the selected historical data periods if needed.

  1. Select the Month of year tab.
  2. Review the variations in number of contacts, handling time and wrap-up time for each month. Note that it is the pattern that is important and not the volume.
  3. Adjust the values manually if needed.
  4. Click Next to refine the variations within the month.

Refine the variations within the month

Review the variations between the weeks of the month and adjust if needed. The data shown in the chart is based on the previously selected historical data. Adjust the selected historical data periods if needed.

NOTE   A week is here a 7-day period starting on the first day of the month. For example, week 1 is always from the 1st to the 7th of the month and week 2 is from the 8th to the 14th.

  1. Select the Week of month tab.
  2. Review the variations in number of contacts, handling time and wrap-up time for each week. Note that it is the pattern that is important and not the volume.
  3. Adjust the values manually if needed.
  4. Click Next to refine the variations within the week.

Refine the variations within the week

Review the variations between the days of the week and adjust if needed. The data shown in the chart is based on the previously selected historical data. Adjust the selected historical data periods if needed.

  1. Select the Day of week tab.
  2. Review the variations in number of contacts, handling time and wrap-up time for each day of the week. Note that it is the pattern that is important and not the volume.
  3. Adjust the values manually if needed.
  4. Click Next to consider to apply a trend factor.

Apply a trend factor to the forecast

If your workload is increasing or decreasing with time, it’s possible to consider the trend when creating the forecast.

The yearly trend factor shows the average increase or decrease in volume over time when the seasonal variations are disregarded. The chart displays the defined trend for the selected data period.

NOTE   At least two years of data is required for the trend calculation to be meaningful.

  1. Select the Trend tab.
  2. Analyze the trend and adjust the percentage to fit your demands.
  3. Select the Use trend check box to apply the defined yearly trend percentage to the forecast.

    It is optional to use the trend factor. If you don't want to use it, just make sure the Use trend check box is cleared.

  4. Click Next to apply the long-term forecast to a selected period.

Apply the long-term forecast to a selected period

Review the long-term forecast. Select a period and a scenario and apply the long-term forecast.

  1. Select the Total tab.
  2. Select which Scenario to save this forecast to.

    When you schedule for a selected scenario, the forecast for that scenario is used. You can copy a forecast to another scenario later if needed.

  3. Select the time period to apply the long-term forecast to.
  4. Click Apply.

    The chart displays the long-term forecast that is created based on the validated historical data, the index values for months, weeks and days of the week and the trend.

  5. Review the long-term forecast and make changes if needed. Adjust individual values manually.
  6. If there are future days that you want to forecast based on historical data from a recurring event, you can add those days to special events. The forecast for the future dates will be based on an average of the volume for all the historical dates that belong to the same special event.
    • Right-click the special event you want to add a date to in the Special events field and select Edit.
    • Select the dates you want to add in the calendar view and move them to the field of selected dates by using the arrow buttons. Select the Ctrl key to select several dates.
    • Click OK. The selected dates are highlighted in green in the table view.
  7. When you are satisfied with the long-term forecast, click Finish to save it to the selected scenario and period.

Related topics