Installed Base Forecasting

There are businesses in most industries that derive revenue from an installed base of product users, contractual relationships, and products that require post-sales support. Traditional time-series forecasting methods do not fare well in forecasting customer demand in these businesses, especially those for which the customer numbers can drastically change. Causal models based on first forecasting an “installed base” of customers are more accurate in these environments. This column describes an analogy-based “Pot-of-Water” Forecast causal model useful for these businesses, and especially for predicting “turning points.” LARRY LAPIDE | Dr. Lapide is a Research Affiliate at MIT and a Lecturer at the University of Massachusetts, Boston Campus. He has extensive experience in industry, consulting, business research, and academia as well as a broad range of forecasting experiences. He was an industry forecaster for many years, has led forecasting-related consulting projects for clients across a variety of industries, and has researched as well as taught forecasting. He was also a market analyst researching forecasting and supply chain software. (This is an ongoing column ...