Community member disengagement: a fundamental threat for viable research communities
Co-written with Steven Debaere, Data Scientist at InSites Consulting and PhD Candidate at IÉSEG School of Management. Recognizing the benefits of online research communities to incorporate external consumer knowledge into innovation processes, firms increasingly ignore temporal limitations and aim for long-term collaborations. However, more than half the online communities fail to generate successful outcomes, usually due to their inability to derive value from individual participants, which depends directly on their interaction frequency and quality. Member disengagement, through low participation levels and the contribution of low-quality arguments, represents a fundamental threat for long-term viable research communities.
Why do existing reduction strategies fail or fall short?
Several strategies are already in place to reduce member disengagement, such as reputation mechanisms at user or post level and content authoring. However, especially in long-term communities, such strategies can lead to important disadvantages. They work reactively, because corrective actions take place only after disengagement has been observed; also, they are subjective in nature because they require human involvement. Furthermore, time and budget constraints limit any community’s capacity to process and evaluate the tremendous quantity of member-produced data.
The huge potential benefit of predictive analytics
An approach that is proactive, objective and time/cost-efficient could thus improve community management, through disengagement predictions obtained with predictive analytics, which exploits historical data and uses machine learning techniques to predict future events. Online research communities provide an ideal context for this application because of their inherent data-rich nature. In particular, the big data characteristics of a community – volume, velocity and variety (3V) – offer favorable conditions for disengagement predictions. Furthermore, if one can predict disengagement, one can anticipate it and try to prevent it.
Tweetaway: Predictive analytics in #mroc to anticipate and prevent member disengagement http://insit.es/22g39dW by @steven_debaere via @InSites #mrx
How to incorporate predictive analytics in your community to prevent member disengagement
Come and check out our presentation for more information and practical tips at the Association for Survey Computing (ASC) Conference, 8-9 September in the University of Winchester (UK). Check out our abstract below for a sneak peek summary. Tom Cruise, bring it on!
Minority Report in market research online communities
The future can be seen, crime can be prevented. The 2002 award-winning movie Minority Report describes a world in 2054 in which crime can be predicted and prevented. In 2016, stimulated by the decade of data, behavioral predictions are already today’s reality. The data-loaded environment of online research communities is still unexplored. Yet, plenty of opportunities exist to tackle critical challenges. Member disengagement, through low participation levels and low-quality contributions, represents a fundamental threat for building healthy long-term research communities. We integrate predictive analytics in research communities and explain how moderators can effectively predict and proactively anticipate member disengagement. The future can be seen, member disengagement can be prevented.
P.S. For those not familiar with the movie Minority Report, get up-to-date with the trailer.
Tweetaway: Join @steven_debaere at @ascorg conference on how to predict and prevent member disengagement http://insit.es/22g39dW #mrx #insites