Abstract
Companies use forecasts to predict deviations from their plans. Most of the forecasting processes rely on experience-based human judgement. Thus, they neglect predictive analytics which aims to be a "modern crystal ball." We take a utility company and its cash flow planning process as our reference. Combining the best of both worlds, experience-based human judgement and predictive analytics, we lay out three imperatives for a more accurate and efficient cash flow prediction model as follows: (1) The art of modelling is about the right input factors and their meaningful combination. A deep business understanding and insights from a correlation matrix will help. (2) The business department should set a clear target for the model in advance. Transparency of the algorithms and delivering tangible results are more important than the last inch of accuracy. (3) Leveraging data decomposition and a meaning-ful method mix, a clean-sheet approach often takes less time and people than existing approaches.
Recommended Citation
Esswein, Markus; Mayer, Joerg H; Stoffel, Sebastian; and Quick, Reiner, (2019). "PREDICTIVE ANALYTICS – A MODERN CRYSTAL BALL? ANSWERS FROM A CASH FLOW CASE STUDY". In Proceedings of the 27th European Conference on Information Systems (ECIS), Stockholm & Uppsala, Sweden, June 8-14, 2019. ISBN 978-1-7336325-0-8 Research Papers.
https://aisel.aisnet.org/ecis2019_rp/23