Tweet Me What You Want, What You Really Really Want - The Informative Value of Tweets for Customer Feedback in Crowdfunding

Abstract

Twitter allows users to publicly share their opinions on certain products in the form of micro-blogs. Such data can be collected and classified to help crowdfunding campaign owners to infer user needs, detect problems with their product and services and plan for future releases of their offers. However, automatically capturing, classifying, and presenting useful tweets is not a trivial task. Challenges stem from the scale of the data available, its unique format, its diverse nature, and the high percentage of irrelevant information and spam. Motivated by these challenges, this paper reports on a two-fold study that aims at leveraging Twitter as a main source of feedback on crowdfunding campaigns. The main objective is to enable a responsive, interactive, and adaptive data-driven user feedback process. Our analysis is conducted on a random sample of 3,600 tweets that we collected from Twitter feeds of 12 crowdfunding campaigns. Our preliminary results reveal that around 47% of collected tweets contain useful information that campaign owners can use for feedback. Moreover, the paper discusses effective classification- and summarization-techniques to analyze and present the information contained by tweets in an adequate manner.

Keywords: Twitter, Crowdfunding, Customer Feedback, Data Analysis

JEL Classification: M13, M15