Positives aside, our team has to look at the negatives while making forecasts using the past three months of subscription data. Forecasts for the prior year were inaccurate because spending used for telemarketing was not always consistent. To make accurate forecasts our team has to take these variables that were not always consistent into consideration. All factors have to be taken into account before future forecasts can be made. Making future projections using three months of data may not be sufficient. While running the paper, three months is not a very substantial amount of time in the long run. The Springville Herald should be gathering more data for a longer period of time than three months. There are too many variables that can arise in the future months that may not be seen in the prior three months of data. It would be our recommendation that the Springville Herald use more than three months worth of data. In order to get a more accurate forecast of the subscriptions we need to account for the most important variables. In the forecast it needs to be determined how many hours of telemarketing will be allocated and stick to it. Since telemarketing has an effect for the sales data, telemarketing has to be an important variable. While going off prior data has a history of not working, a mix of actual data and proposed data need to be used since time is valuable. In order for the forecast to be more accurate modified data for more than three months needs to be used. We say modified data because the data used needs to be changed to be more consistent. Using more consistent hours of telemarketing data will give a better forecast. We have to make sure that the Herald will use telemarketing numbers that do not fall outside of a certain range. By using controlled data accounting the prior year for errors that have been identified will provide a better forecast than using the prior three months.
13.2
Factors that may affect the