Gamification of Workplace Activities - Smart Information Systems: Computational Intelligence for Real-Life Applications - page 250

9.3.3 Experiment and Results
In order to evaluate our second research question of whether users' perception of
gamification (as stated in the online questionnaire) matches with their behavior using
a gamified system, we announced the relaunch of the gamified enterprise bookmark-
ing system using the institute-wide mailing list and observed the users' interaction
with the system. In this section, we provide an overview of the system's usage from
2009 till the introduction of gamification functionalities in mid-June 2013 to provide
insights into the usage of the system before it got gamified. Following that, we present
the usage statistics after the gamified version was introduced.
9.3.3.1 System Usage Before Gamification
In the analysis of usage patterns we look at data from 2009 until the gamified version
went online in June 2013. 13 The focus of the analysis is to look at two important
aspects of the bookmarking system: The main feature, the creation of bookmarks,
and the number of recommended bookmarks to colleagues, one of the main goals
for enterprises to support knowledge sharing. Since a user-centric evaluation of this
system was never intended for the legacy system, the (pre-gamified) system did not
track any user interactions. Thus, only actions resulting in a database entry can be
tracked. This was changed in the gamified version.
Table 9.2 shows the yearly distribution of created items and active users. Outliers,
in this case a power user responsible for half of all system interactions, are removed
from the statistics. After a solid start in 2009 with roughly one-third of employees
using the system, 2010 already showed a significant decline in usage. While the
usage number remained stable for the years 2010-2013, only 3% of people were
using the system.
Table 9.3 shows the total number of bookmark recommendations. Similar to the
created items, the number peaked in 2009. From 2009 till 2013, they remained at a
stable level. This comes as no surprise, as the system's user interface (UI) supports
the recommendation during a bookmark creation. Recommending an item at a later
stage was still possible, but not as compelling as in the creation process. Noticeable
is the comparison of the number of users. As shown in Table 9.2 , we recorded a total
of 30 users in 2009 with only 14 of them using the recommendation feature. This
might be due to the mentioned UI flaws, but could also indicate that those users use
the system only for personal managing purposes. From 2010 till 2013 usage numbers
remain the same, suggesting that the users who use the system over a long period of
time are well familiar with its functionalities.
The statistics show that the non-gamified version of the bookmarking system was
used only rarely. In the next section, we present and discuss results of the gamified
13 The gamified version went online on June 19. The presented analysis however only considers
data till June 1, to prevent any influence of the data because of any office kitchen chatter about the
bookmarking system.