Text Mining Approach for Identifying Product Ideas and Trends Based on Crowdfunding Projects

Abstract

Current research and development investments and innovation developments show that there are more failed innovation attempts than ever and there are numerous occasions where companies need to be highly selective by assessing customers’ needs or preferences. Crowdfunding is a great approach for this purpose, where investments are made if there are funders who are interested in the proposed innovations. As crowdfunding platforms have proved to assist in the introduction of innovations, the details of these projects can be a great inspiration for others. The aim of this study is to integrate crowdfunding projects into innovation and product developments as a source of potential ideas and trends. We use a text mining approach to analyze 8021 crowdfunding projects from the period 2009–2018. In this study, we cluster these projects into nine innovation areas desired by consumers. Through our methodological approach, we examine the linkage between associated text, features of projects, and funding of projects to uncover emerging product features that illustrate the desires of consumers. Our proposed model offers theoretical contributions to the innovation processes, especially the fuzzy front end with an open innovation approach. We provide practical contributions by revealing crowdfunding platforms as a means to gain insights into product elements for the design of new products and services.