Dayforce forecasts the KPIs in your organization’s plan based on a comparison between data from the past few weeks and the same time period during last year. From this comparison, it determines if a location’s results are trending up or down and forecasts new values for the upcoming week based on that trend.
For Dayforce to forecast properly, the recent data needs to be matched and compared with the appropriate data from the previous year. For most weeks, this comparison is a one-to-one match-up. For example, data from the middle of September one year can be matched to the same week in September of last year as it can be assumed that a location performed similarly for that week in the current and last year.
Special days are specific dates where a one-to-one comparison with the same date in the previous year would be inappropriate for the Dayforce forecasting engine. You can designate dates as special days so that they represent exceptions to normal operations, resulting in forecasts that aren’t skewed by inappropriate data. Common examples of special days are statutory holidays or in-store sales.
For example, a retail company’s locations are closed on Good Friday. As a result, each year on Good Friday there are no sales. Because Good Friday is on different dates for different years, the comparison of this year’s data to last year’s data would return some inappropriate trends. Consider:
- Last year, Good Friday was on March 10. Comparing March 10 of this year, when there’s a level of sales typical of March, to the same date last year, when there were no sales, is going to give the false impression of a large up trend in the store’s performance.
- This year, Good Friday was on March 2. Comparing March 29 of this year, when there were no sales, to the same date last year, when there was a typical level of sales, is going to give the false impression of a large down trend in the store’s performance.
Special days inform Dayforce how to compare data from one year to the next when forecasting. Continuing the example above, a special day is created in Dayforce for Good Friday that maps the date from one year to the next. That way the unusual sales (of zero) on Good Friday don’t skew the forecast.
In addition to the special day itself, forecasting needs to take into account proximity days, which are the days that lie near special days and whose outcomes are affected as a result of the special day. For example, Easter Sunday forcing a store to close so that some of the sales that might have occurred on that day is shifted to Saturday, instead. As a result, the Saturday might have higher sales not only due to the occurrence of the holiday, but also because of the store closure.
Even though a holiday falls on the same calendar date every year, like Christmas Day and Boxing Day, because they fall on different days of the week each year, they can still impact the forecast’s trends and their proximity days will be different from year to year. These dates are also set up as special days in Dayforce to ensure appropriate historical data is selected when forecasting.