Title: Forecasting new product growth

Authors: Rajenlall Siriram

Addresses: School of Mechanical, Industrial and Aeronautical Engineering, University of the Witwatersrand, Johannesburg, South Africa

Abstract: The accuracy of seven growth curves, namely: linear regression, logistic regression, exponential, Bass, Logistic Pearl, Gompertz and Harvey, are tested. Ten products are selected with five product categories: 1) electronic devices (i.e., Global Apple iPhones, Global Apple tablets, Global Samsung smart phones, Global Samsung tablets); 2) social media applications (i.e., Facebook users, WhatsApp users, Twitter users); 3) streaming applications (i.e., Netflix users); 4) bitcoin users; 5) industrial robots. The curves are compared using analysis of variance. A panel of experts is used to critique the forecasting results and provide opportunities for improvement. Comparisons between the five product categories and the expert panel involvement provided interesting opportunities for further research. None of the curves appeared to be dominant.

Keywords: technology forecasting; substitution models; product growth; forecasting accuracy.

DOI: 10.1504/IJFIP.2022.128053

International Journal of Foresight and Innovation Policy, 2022 Vol.16 No.1, pp.33 - 65

Received: 28 Jun 2021
Accepted: 25 Apr 2022
Published online: 04 Jan 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article