Content uploaded by Matti Mäntymäki
Author content
All content in this area was uploaded by Matti Mäntymäki on Dec 09, 2015
Content may be subject to copyright.
What drives continuance in online music streaming?
Thirty Sixth International Conference on Information Systems, Fort Worth 2015 1
Gratifications from using freemium music
streaming services: Differences between
basic and premium users
Completed Research Paper
Matti Mäntymäki
Turku School of Economics
University of Turku
FI-20014
matti.mantymaki@utu.fi
A.K.M. Najmul Islam
Turku School of Economics
University of Turku
FI-20014
najmul.islam@utu.fi
Abstract
Online music streaming services have become popular in listening to music. Most online
music streaming services employ the freemium business model. The specific
gratifications from online music streaming are not well understood. Moreover, research
examining the freemium model from a user experience perspective remains scant. We
employ uses and gratifications theory and examine four gratifications, namely
ubiquity, social connectivity, discovering new music, and enjoyment, as the predictors
of continuance intention. We examine the differences in these gratifications between the
basic and premium users with data from 374 Spotify users. The results demonstrate
that enjoyment, discovering new music, and ubiquity, are the main drivers of the
continuance intention. Interestingly, social connectivity has no effect on continuance
intention. Furthermore, premium users experience higher levels of enjoyment and
ubiquity than the non-paying basic users. Finally, enjoyment is the only predictor of
continuance intention among basic users, but has no effect among premium users.
Keywords: music, streaming, freemium, continuance
What drives continuance in online music streaming?
Thirty Sixth International Conference on Information Systems, Fort Worth 2015 2
Introduction
Online music streaming services such as Spotify, Pandora, Rdio, Deezer, and SoundCloud have become
popular ways to listen to music. Spotify, the most popular service, has more than 60 million active users
globally, of which 15 million pay for its premium service
1
. The rapid proliferation of online streaming has
had a profound impact on both the consumption and distribution of music
2
.
Currently, the online music streaming service providers are facing intensifying competition as new
services such as Google Music and Apple Music enter the market. Hence, retaining existing is becoming
increasingly important for online music streaming services. Fostering customer retention requires
understanding the factors that drive people’s future usage decisions in order to manage and develop the
user experience.
Despite its topicality, the user experience in online music streaming services has received little academic
scrutiny, although prior research has examined the drivers of downloading (Kunze and Mai 2007; Molteni
and Ordanini 2003) and sharing music online (Bhattacharjee et al. 2006). Krause et al. (2014) identified
three gratifications obtained from using music applications on Facebook, namely communication,
entertainment and habitual diversion. These gratifications, however, are rather generic and similar to the
ones identified with respect to e.g. social network sites (Park et al. 2009; Raacke and Bonds-Raacke
2008)). Thus, there is a clear need for research that investigates the specific features of online music
streaming and the respective gratifications.
Furthermore, most online music streaming services employ the freemium business model (Anderson
2013; Teece 2010). In the freemium model a basic or downgraded version of the software is offered free of
charge, while users who wish to have more features can purchase a premium version (Anderson 2009).
Consequently, there are two groups of users with two types of user experience: the ones who use the free
basic version, i.e. the basic users, and the ones who use the premium version, i.e. the premium users.
Considering the widespread adoption of the freemium model, only a few studies (Mäntymäki and Salo
2015; Vock et al. 2013; Wagner et al. 2014; Wang et al. 2011) have empirically examined the user
experience in freemium services and even fewer (Oestreicher-Singer and Zalmanson 2013; Wagner et al.
2014) have done so in the online music streaming context. According to Wagner et al. (2014) attitudes
toward the free version do not correlate with attitudes toward the premium version in online music
streaming. This implies that online music service providers need to have customized intervention plans,
depending on the account type if they want to retain both types of users.
Hence, this study contributes to the research on online music streaming and the freemium business
model by empirically examining 1) the key gratifications driving online music streaming, and 2) to what
extent experiencing these gratifications differs between basic and premium users.
To this end, we employ uses and gratifications theory (U & G) (Katz et al. 1974) and identify four
gratifications, namely ubiquity, social connectivity, discovering new music, and enjoyment, as the
predictors of the continuance intention. Furthermore, we examine the differences in experiencing these
gratifications between the basic and premium users with data collected from 374 users of the leading
online music streaming service, Spotify, and analyze the data using structural equation modeling (SEM)
and analysis of variance (ANOVA).
1
https://press.spotify.com/us/information/
2
Sisario, Ben (2013) "As Music Streaming Grows, Royalties Slow to a Trickle" New York Times", 28 January 2014
http://www.nytimes.com/2013/01/29/business/media/streaming-shakes-up-music-industrys-model-for-royalties.html
What drives continuance in online music streaming?
Thirty Sixth International Conference on Information Systems, Fort Worth 2015 3
The SEM results demonstrate that enjoyment, discovering new music, and ubiquity, are the main drivers
of the continuance intention. Interestingly, social connectivity has no effect on continuance intention. The
ANOVA results reveal that premium users experience higher levels of enjoyment and ubiquity than the
non-paying basic users. In addition, the results show that while enjoyment is the only predictor of
continuance intention among basic users, it has no effect among premium users.
The remainder of the paper proceeds as follows, after the introductory section we discuss how the
freemium model has been applied in online music streaming and present the theoretical background of
the study. In the third section, the research hypotheses are presented. The fourth section covers the
empirical research and the results. In the fifth section, we discuss the main findings of the study,
summarize the theoretical and practical contribution, unveil the limitations of the study, and suggest
areas for future research.
Background
Freemium business model
The term freemium is a combination of free and premium that describes a business model in which a
basic product or service is made available for free, while the users who wish to receive additional features
and/or an enhanced user experience can either purchase a premium subscription or make purchases
within the service (Anderson 2009; Anderson 2013; Teece 2010). Additionally, the service provider can
use advertising to cover the costs from offering the basic version for free. In online music streaming
services such as Spotify and Rdio, the basic version typically includes advertising and one benefit of the
premium membership is an ad-free, uninterrupted listening experience.
When employing the freemium model, the service provider's ability to retain the paying users and convert
the non-paying users into paying ones is a critical success factor (Kumar 2014). As a result, employing the
freemium model requires constant optimization to maintain the delicate balance between the content of
the free and premium offerings. The free version should deliver enough value to attract new users and
retain the current ones. The premium version should offer sufficient value-added compared to the basic
version to justify its cost. Hence, employing the freemium model leads to two standards of the user
experience. To illustrate the issue in the online music streaming context, Table 1 summarizes the added
benefits of a Spotify premium user account compared to a basic account. We choose Spotify as it is the
market leader in online music streaming.
Table 1. Summary of the benefits for Spotify premium user accounts
Added benefits
Description
No commercials
Commercials do not interrupt premium account listeners.
Unlimited listening time
The listening time for a premium account is unlimited (but only
20 hours for the basic account).
Ability to select individual songs
when using the mobile application
The basic account only offers shuffle play for mobile devices.
Enhanced audio quality
Premium account includes improved audio quality (320 vs. 160
kb/sec).
Offline listening
Playlists can be downloaded to a mobile phone, tablet device or
computer for offline listening.
Exclusive new releases
New songs and albums are pre-released for premium account
holders.
Prior research on the freemium model in online music context offers insights on the factors that may
explain upgrading from the basic version to the premium user account (Oestreicher-Singer and
Zalmanson 2013; Wagner et al. 2014; Wang et al. 2011). Sustained user engagement in turn has received
has received less scholarly attention.
What drives continuance in online music streaming?
Thirty Sixth International Conference on Information Systems, Fort Worth 2015 4
Uses & Gratifications theory
U & G (Katz et al. 1973; Katz et al. 1974) is a theoretical framework that is used to study how people
choose and use new media (Rayburn and Palmgreen 1984; Ruggiero 2000; Stafford et al. 2004).
According to U & G, people distinguish between different forms of media based on the needs they expect
to satisfy through their media use (Katz et al. 1973). U & G posits that 1) media use is goal-directed
behavior, aimed at fulfilling one’s individual set of needs (Blumler 1979), 2) people are aware of their
needs (Blumler 1979), and 3) people actively seek and use media.
U & G views needs as “the combined product of psychological dispositions, sociological factors, and
environmental conditions” (Katz et al. 1973, p. 516-517). Gratifications in turn are the perceived
fulfillment of a need through an activity, such as media use (Rayburn and Palmgreen 1984). Rather than
providing a predefined set of factors and constructs, U & G offers a guiding framework for context-specific
theorization (see e.g. Eisenbeiss et al. 2012; Mäntymäki and Riemer 2014). Thus, understanding the use
context of a media provides the basis for the successful employment of U & G.
Prior U & G research has identified several broad types of benefits people can derive from media use. For
example, Nambisan and Baron (2009) suggested four gratifications that predict participation to virtual
customer environments, (1) cognitive benefits that relate to information acquisition and strengthening of
the understanding of the environment; (2) social integrative benefits that relate to strengthening
consumer’s ties with relevant others; (3) personal integrative benefits that relate to strengthening the
credibility, status, and confidence of the individual; and (4) hedonic or affective benefits such as those
that strengthen aesthetic or pleasurable experiences (Katz et al. 1974; Nambisan and Baron 2009).
Stafford et al. (2004) identified three generic gratifications from using the Internet, namely content
gratifications, process gratifications and social gratifications. Content gratifications relate to the outcomes
from using the media whereas process gratifications stem from the enjoyment and pleasure experienced
when using the media. Social gratifications in turn stem from establishing and maintaining connections
with other people.
Online music streaming services have unique properties that are distinct from listening to downloaded
audio files with a mobile device or a computer. These properties enable certain context-specific
gratifications. As a result, we employ social connectivity value, discovery value, ubiquity and enjoyment as
the gratifications from using online music streaming services. Our dependent variable is continuance
intention (i.e. the intention to use the service in the future). The focal constructs will be discussed in detail
in the following section. Table 2 below offers a summary of the constructs and their definitions.
Table 2 Research constructs and their definitions
Construct
Definition
Reference
Social
connectivity
The extent to which using the online music streaming service
helps to attain gains in obtaining information about other
people's music preferences and the sharing of favorite music
with others.
(Dholakia et al. 2004)
Discovery of
new music
The extent to which using an online music streaming service
helps to discover new music and broaden musical taste.
(Sheth et al. 1991;
Sweeney and Soutar
2001)
Ubiquity
The extent to which using the online music streaming service
helps to attain gains in accessing music irrespective of time
and place.
(Okazaki and Mendez
2013)
Enjoyment
The extent to which using the online music streaming service
is perceived as enjoyable in its own right.
(Davis et al. 1992)
Continuance
intention
Intention to use the online music streaming service in the
future.
(Bhattacherjee 2001)
What drives continuance in online music streaming?
Thirty Sixth International Conference on Information Systems, Fort Worth 2015 5
Hypotheses
First, online music streaming services include social features that allow users to create playlists and to
follow the playlists of others. Moreover, users can share their playlists via Facebook and allow Spotify to
give automatic updates on the music they are listening to. The social features allow users’ music
preferences more visible and help to get in touch with other users with similar music taste. Hence, the
social features and direct integration to social network sites is a factor that differentiates online music
streaming services from listening to downloaded audio files.
From a U & G perspective, social connectivity facilitated by using an online music streaming service can
be considered as social gratification related to establishing social ties with other people (Katz et al. 1974).
Prior research has found that social relationship support predicts purchasing digital products (Kim et al.
2011) and participating virtual customer environments (Nambisan and Baron 2009). As a result, it is
meaningful to investigate to what extent the social interaction facilitated by the online music streaming
service predicts the continuance intention. Consequently, we put forward the following hypothesis:
H1: Social connectivity will positively affect the continuance intention.
Second, as the digital music libraries of online music streaming services include millions of songs, the
services include recommendation agent features (Xiao and Benbasat 2007). The recommendation agent
features offer customized recommendations based on user's listening history. This helps users to better
utilize the large media library and find or discover music that fits their preferences.
As a result, the features affording discovering new music offer the user cognitive gratifications as
described in U & G (Katz et al. 1973). In addition, from a customer value perspective, discovery of new
music can be related to epistemic value, i.e. a service's capacity to arouse curiosity, offer novelty or satisfy
a desire from knowledge (Sheth et al. 1991; Sweeney and Soutar 2001). This type of gratification is
particularly important for experiential services (cf. Sheth et al. 1991) such as music streaming. Thus, we
examine the discovery of new music as a gratification from online music streaming and hypothesize the
following:
H2: Discovery of new music will positively affect the continuance intention.
Third, with online music streaming services the users does not need to possess the music files in any
physical or digital format as they can access the tracks via the Internet. In addition to this, Spotify and
Rdio premium users for example have the option to download their music to a device for offline listening.
Considering also the size of the music library available for the users, we hold that ubiquitous access to
music is a gratification salient to online music streaming.
From a U & G perspective, ubiquity can offer personal integrative gratifications through reinforcing one’s
sense of self-efficacy related to music listening (Katz et al. 1974; Nambisan and Baron 2009). As a result,
we employ the concept of ubiquity (Okazaki and Mendez 2013; Tojib and Tsarenko 2012) to predict
continuance intention.
H3: Ubiquity will positively affect the continuance intention.
Fourth, online music streaming is likely to be enjoyable as such as listening to one’s favorite music can in
general consider a hedonic experience. Extensive prior U & G research has acknowledged enjoyment as a
gratification arising from consuming various types of media (Katz et al. 1974; Nambisan and Baron 2009),
including online music (Krause et al. 2014). In addition, extensive prior research has found enjoyment to
predict the use of hedonic IT applications (van der Heijden 2004) such as social virtual worlds
(Mäntymäki and Riemer 2014), online games (Li et al. 2015) and social networking sites (Cheung et al.
2011). Hence, we examine enjoyment as a gratification resulting from online music streaming as a
predictor of continuance intention.
H4: Enjoyment will positively affect the continuance intention.
To investigate the influence of the freemium model on the user experience, we examine to what extent the
four gratifications differ between the basic and premium users. The recommendation and social features
of Spotify are similar for both basic and premium accounts. However, it is possible that the two user
groups evaluate these two gratifications differently.
What drives continuance in online music streaming?
Thirty Sixth International Conference on Information Systems, Fort Worth 2015 6
The premium users spend money on their use of online music streaming. This may imply that they
consider music in general, or the online music streaming in particular, more important than the basic
users. Therefore, premium users, may consider the social features that help to find other users with
similar music taste more valuable than the basic users
With respect to discovery of new music, the premium users have unlimited monthly listening time and are
thus likely to use the service more than basic users. Consequently, they can also derive more value from
the recommendation features. In addition, when using the mobile application, premium Spotify users can
freely select the songs they want to listen whereas basic users can only use the shuffle play mode and skip
only a limited number of songs per hour. These features are likely to increase the value the premium
version in social connectivity and discovering value compared to the basic version.
H5: Users with a premium account experience higher levels of social connectivity than users with a
basic account.
H6: Users with a premium account experience higher levels of discovery of new music than users with a
basic account.
As discussed earlier, Spotify's premium users have the option to download their playlists for offline
listening. In addition, when using the mobile application, the premium users can freely select the songs
they want to listen and skip any song when using the shuffle play mode or radio features. Because of these
value-added features, premium users are likely to experience higher levels of ubiquity than basic users.
With regards to enjoyment, premium users are not distracted by the commercial breaks included in the
basic version. The commercial breaks interrupt the listening and are thus likely to decrease the enjoyment
of the user experience. Second, premium users receive exclusive access to new releases before they are
officially available and also receive enhanced audio quality and features for multi-room music
distribution. Third, the monthly listening time for premium users is unlimited which allows them to use
the service more than the basic users. Together, these benefits are likely to make the premium version
more enjoyable than the basic version. As a result, we put forward the last two hypotheses:
H7: Users with a premium account experience higher levels of ubiquity than users with a basic account.
H8: Users with a premium account experience higher levels of enjoyment than user with a basic
account.
Figure 1 below summarizes the hypothesized research model.
Figure 1 The Research Model
Continuance
intention
Discovery of
new music
Social
connectivity
Ubiquity
Enjoyment
Account
type
H3 +
+
H2 +
H1 +
H6 +
H5 +
H4 +
H7 +
H8 +
What drives continuance in online music streaming?
Thirty Sixth International Conference on Information Systems, Fort Worth 2015 7
Empirical research
Measurement development and data collection
The measures for ubiquity, enjoyment, and the continuance intention were adopted from existing scales.
The survey instrument used in the measurement is presented in Appendix A. The measures for
discovering new music and social connectivity were developed for this study. We interviewed seven
Spotify users and asked them to elaborate on why they use the service and what features of the service
they consider particularly valuable. The respondents stated that the artist recommendations and
information about new releases offered by Spotify help in discovering new music. With respect to social
interaction, the respondents considered the possibility to follow other users and receive information about
other people's music preferences a unique feature of Spotify.
Based on the information obtained from the users, we established a list of 12 candidate items measuring
the discovery of new music and social connectivity. This list was thereafter presented to a group of four
experienced Spotify users and two senior academics. Finally, six items measuring discovering new music
and three items measuring social interaction qualified for use in the measurement. All constructs were
measured with reflective indicators since the direction of causality was from constructs to their items
(Cenfetelli and Bassellier 2009). The items were measured on a five-point Likert scale anchored from
strongly agree to strongly disagree.
The data were collected with an online survey from Finnish Spotify users. Therefore, the survey
instrument was translated into Finnish by the first author who is a native Finnish speaker and checked by
another native Finnish-speaking scholar. To collect the data, we first randomly selected a group of one
thousand respondents from our university's database. The database includes students at bachelor's,
master's and PhD levels as well as graduates who take additional courses to complement their degrees. In
addition, we published an invitation to participate in the survey on two Facebook groups. Altogether, the
survey was opened 637 times and 376 respondents proceeded to the final page and submitted the survey,
leading to a completion rate of 58.7 per cent. After omitting two responses with clearly incorrect data, the
final sample consisted of 374 responses, 227 respondents were female and 147 male. The age of the
respondents varied between 18 and 57, the mean age of the respondents being 25. Altogether, 165 (44 %)
respondents held the basic account and 209 (56 %) the premium account. We also examined the amount
of time spent on the service with 1) a computer, 2) a mobile phone, and 3) a tablet device. The average
overall listening time per day was 60 minutes for basic users and 110 minutes for premium users. To test
the possible non-response bias, we compared the earliest 20 percent and the last 20 percent of the
responses. A t-test did not detect any statistically significant difference between the demographic profile
of the early and late responses.
Analysis and results
We started the analysis by exploring the gratifications using a principal component analysis with Varimax
rotation. Four factors with eigenvalues greater than 1 representing each gratification emerged from the
rotated solution.
As a result, we proceeded to confirmatory factor analysis. We first examined the factor loadings and
composite reliabilities. We deleted items with loadings less than 0.7 to the respective construct. The
composite reliability values ranged from 0.864 to 0.966 and the average variance extracted (AVE) values
ranged from 0.632 to 0.726. These values clearly exceeded Fornell and Larcker's (Fornell and Larcker
1981) minimum criteria of 0.7 for composite reliability and 0.5 for AVE. Item means, standard deviations,
loadings, composite reliabilities and AVEs are presented in Table 3 below.
What drives continuance in online music streaming?
Thirty Sixth International Conference on Information Systems, Fort Worth 2015 8
Table 3 Item means, standard deviations, loadings, composite reliabilities and AVEs
Item
MEAN
S.D.
LOADING
Composite
reliability
AVE
UBI1
4.032
1.185
0.811
0.938
0.684
UBI2
3.850
1.185
0.959
UBI3
4.168
1.007
0.764
DISCO1
4.262
0.861
0.757
0.932
0.726
DISCO2
4.051
0.969
0.807
DISCO3
4.123
0.909
0.866
DISCO4
4.259
0.935
0.821
SOC3
2.898
1.111
0.855
0.888
0.644
SOC2
2.471
1.019
0.768
SOC1
3.297
1.056
0.699
HED1
4.115
0.789
0.747
0.864
0.632
HED2
4.222
0.758
0.795
HED3
3.904
0.848
0.726
CI1
4.543
0.790
0.880
0.966
0.723
CI2
4.545
0.786
0.975
CI3
4.492
0.866
0.944
The highest bivariate correlation was 0.453 (between future use and enjoyment), which was clearly lower
than the square root of the weakest AVE value (0.795 for enjoyment). The correlations between research
constructs are presented in Table 4 below.
Table 4 Correlations between the constructs (bolded items in
the main diagonal square roots of AVEs)
DISCO
UBI
SOC
ENJ
CI
DISCO
0.852
UBI
0.148
0.827
SOC
0.293
0.033
0.803
ENJ
0.421
0.244
0.091
0.795
CI
0.370
0.269
0.009
0.453
0.851
Altogether, the measurement exhibited good convergent and discriminant validity. The model fit indices
(χ2/DF 2.085; GFI 0.939; AGFI 0.912; CFI 0.972; RMR 0.044; RMSEA 0.054) for the measurement
model clearly indicated a good fit.
Finally, we tested the risk of common method bias (CMB) was examined with a confirmatory factor
analysis using the single-factor approach presented by Malhotra et al. (2006). The single-factor model
exhibited a very poor fit, indicating that CMB is unlikely to be a concern.
After having verified the reliability and validity of the measurement, we proceeded to testing the
structural model. The control variables, respondent age and gender did not have a significant effect on
What drives continuance in online music streaming?
Thirty Sixth International Conference on Information Systems, Fort Worth 2015 9
continuance intention. The results showed that enjoyment was the principal determinant of continuance
intention, followed by the discovery of new music and ubiquity. As a result, H2, H3, and H4 were
supported. Contrary to our hypotheses (H1), social connectivity did not have a significant effect on
continuance intention. Figure 2 below summarizes the results from testing the structural model.
Figure 2 Results from the structural model
To investigate the differences in our focal constructs between the basic and premium users, we conducted
a series of one-way ANOVAs. The results are presented in Table 5. The ANOVAs detected a statistically
significant difference between the two groups in ubiquity (p < .001), enjoyment (p < .01) as well as in the
dependent variable, continuance intention (p < .001). For these constructs, the scores were higher among
premium users. Thus, H5 and H6 were rejected whereas H7 and H8 were supported.
Table 5 ANOVA results
Basic
Premium
F-value
Sig.
Continuance intention
Mean
12.927
14.092
24.604
.000
S.D.
2.723
1.821
Ubiquity
Mean
10.752
13.077
9.629
.000
S.D.
3.220
2.433
Discovering new music
Mean
16.406
16.923
2.464
.117
S.D.
3.329
3.029
Social connectivity
Mean
8.624
8.699
.068
.794
S.D.
2.681
2.76483
Enjoyment
Mean
11.879
12.526
63.250
.002
S.D.
2.152
1.879
Use per day (min)
Mean
60.152
110.971
16.588
.000
S.D.
85.130
87.980
Continuance
intention
R2 = 27.6 %
Discovery of
new music
Social
connectivity
Ubiquity
Enjoyment
Account
type
0.159***
0.239***
n.s.
n.s.
n.s.
0.322***
+
***
+
***
Note: ***p<0.001; n.s.: non-
significant
What drives continuance in online music streaming?
Thirty Sixth International Conference on Information Systems, Fort Worth 2015 10
Taken together, the data offered empirical support for five out of our eight hypotheses. Table 6
summarizes the results of testing the research hypotheses.
Table 6 Summary of the hypotheses testing
Hypothesis
Outcome
H1: Social connectivity will positively affect the continuance intention.
Not Supported
H2: Discovery of new music will positive affect the continuance intention.
Supported
H3: Ubiquity will positively affect the continuance intention.
Supported
H4: Enjoyment will positively affect the continuance intention.
Supported
H5: Users with a premium account experience higher levels of social
connectivity than users with a basic account.
Not supported
H6: Users with a premium account experience higher levels of discovery of
new music than user with a basic account.
Not Supported
H7: Users with a premium account experience higher levels of ubiquity than
users with a basic account.
Supported
H8: Users with a premium account experience higher levels of enjoyment
than user with a basic account.
Supported
Finally, we conducted a post-hoc analysis to examine the possible moderating effect of account type by
running the structural model separately for basic and premium users. Interestingly, enjoyment was the
only significant predictor of continuance intention among basic users. In contrast, for premium users,
continuance intention was predicted by discovering new music and ubiquity while the effect of enjoyment
and social connectivity was not significant. We used a z-test to examine the statistical significance of the
differences in the standardized path coefficients between basic and premium users. The z-test indicated
that the effects of enjoyment (p<.001), ubiquity (p<.001), and discovering new music (p<.05) on
continuance intention were statistically different between basic and premium users. Table 7 summarizes
the from the post-hoc analysis.
Table 7 Results from the post-hoc analysis (significant differences in bold)
Relationship
Whole data
Basic account
Premium
account
z-score
Discovering new music Continuance
intention
0.239***
0.131 n.s.
0.421***
1.784*
Social connectivity Continuance
intention
-0.096 n.s.
-0.035 n.s.
-0.149 n.s.
-0.635 n.s
Ubiquity Continuance intention
0.159***
0.006 n.s.
0.269***
2.261***
Enjoyment Continuance intention
0.322***
0.468***
0.071 n.s.
-3.821***
*** p<.001; ** p>.01; *p<.05; n.s. not significant
Discussion
This study set out to contribute to the research on online music streaming and the freemium business
model by empirically examining 1) the key gratifications driving online music streaming, and 2) whether
basic and premium users experience different levels of gratifications. In addition, we investigated to what
extent the effect of these gratifications differs in predicting future usage decisions between basic and
premium users.
The results show that enjoyment is the principal predictor of continuance intention, followed by discovery
of new music and ubiquity whereas social connectivity has no effect. The strong role of enjoyment aligns
with the findings by Krause et al. (2014) from Facebook’s music applications.
What drives continuance in online music streaming?
Thirty Sixth International Conference on Information Systems, Fort Worth 2015 11
Second, the results demonstrate that premium users perceive higher levels of ubiquity, enjoyment and
continuance intention than basic users. This implies that the added benefits of the premium user account
reinforce customer retention.
Third and finally, our post hoc analysis revealed that the factors driving usage intention differ
considerably between basic and premium users (see Table 4). For basic users, enjoyment is the only
predictor of continuance intention. Furthermore, discovering new music is strongest predictor of the
continuance intention for premium users but had no effect on basic users. Hence, yet the features
enabling the discovery of new music are similar across account types, the premium users who use the
service more also benefit from the recommendation agent functionalities more than basic users, whose
monthly listening time is limited to 20 hours
From a customer decision-making perspective, purchasing the premium user account involves a monetary
investment compared to the free basic version, the premium users are likely to consider the benefits
against the associated costs more carefully than the basic users. This can be explained by the theory of
human information processing (Payne 1982; Payne et al. 1988). According to the theory, humans have
limited cognitive capacity to process information. Hence, they tend to optimize the amount of information
processing allocated to a task. Since using the basic version does not include a monetary cost, identifying
sufficient benefits to justify the usage requires less cognitive effort compared to the premium
subscription. Thus, for the basic users, enjoyment alone appears to be sufficient to justify the usage
decision. The premium users in turn are likely to be more deliberately aware of the benefits that they get
in return from the subscription fee.
Implications for research
Considering the dearth of prior research on the motives for using online music streaming services (Krause
et al. 2014) as well as user experience in services employing the freemium model (Mäntymäki and Salo
2015; Vock et al. 2013; Wagner et al. 2014) the present study makes two main contributions to the
literature.
First, by identifying the four gratifications from using online music streaming services, the present study
adds to the prior literature on people’s motives to consume music online (Bhattacharjee et al. 2006;
Kunze and Mai 2007; Molteni and Ordanini 2003; Sanchez-Franco and Rondan-Cataluña 2010). By
identifying the differences in the level as well as in relative effect of enjoyment, our study adds on prior
research (Krause et al. 2014) and advances the understanding of hedonic gratifications in consuming
music online.
Second, by examining the differences in the gratifications between the basic and premium users, the
present study adds on the current research on the freemium model. Our results show that the levels of
gratifications and their impact on continuance intention differ considerably between the basic and
premium users. To the best of our knowledge, this issue has not been taken up by prior research
examining the freemium model.
Furthermore, based on an analysis of rankings, customer ratings, and sales of freemium applications in
Google Play (marketplace for Android applications), Liu et al. (2014) pointed out that the quality of the
free version is the principal heuristic driving the sales of the paid version. Our results add a customer
perspective to this line of inquiry by showing that enjoyment is a focal factor in determining users’
continuance with the free version of online music streaming services.
Implications for practice
The results highlight the importance of managing the user experience differentially for basic and premium
users in order to successfully employ the freemium model. The higher levels of the continuance intention,
enjoyment and ubiquity among the premium users indicate that the additional features and benefits that
come with the premium account elevate the user experience. This in turn may motivate t he purchase of
the premium version of the service. These observations are critical from the managerial standpoint as they
What drives continuance in online music streaming?
Thirty Sixth International Conference on Information Systems, Fort Worth 2015 12
imply that Spotify has been successful in forming and communicating a freemium offering that includes a
value increment for premium users.
We point out two main practical implications from the study. First, ensuring that the user experience is
enjoyable plays a key role in fostering continuance among the basic users. Obtaining and maintaining a
large base of non-paying users to be converted into paying customers is essential when employing the
freemium model (Kumar 2014). Thus, ensuring that the free offering meets users’ expectations is pivotal
in sustaining the user base.
Second, the low scores for social connectivity alongside its non-significant impact on continuance
intention imply that the social features of Spotify do not create value for the users. Prior research on
content websites suggests that the degree of interaction between users is associated with desirable
consequences for example subscribing a premium version of the service (Oestreicher-Singer and
Zalmanson 2013). Following this logic, services offering digital content today typically also include
features for user-to-user interactions either within the service or through integration to social media.
The non-significant effect of social connectivity suggests that the social features of Spotify may be an
unnecessary add-on and do not create the desired lock-in effect. Put in a broader perspective, the
designers of online music streaming services should consider what kind of social features add to the user
experience and whether the users are ready for social listening. For example, automatic Facebook updates
generated by Spotify may even increase the amount of low value information in one’s newsfeed on
Facebook and hence mostly irritate the users. Thus, we encourage system designers to critically evaluate
to what extent the integration to social network sites adds or reduces value.
Our results imply that mechanistic implementation of social features to digital content services may not
be an advisable course of action. The service operators should closely evaluate the value proposit ions and
the use context of their services based on the gratifications the users look for. As pointed in U & G (Katz et
al. 1973), people emphasize different gratifications in their choice of media. For example, when
consuming digital media, users’ willingness to interact with other users can be highly context and
situation-specific.
Altogether, our findings align with Oestreicher-Singer and Zalmanson (2013) who called for a more
strategic approach to integration social interaction to digital content services. In the worst case, social
features can even have an adverse effect on the user experience. Thus, the social aspects of listening music
should be infused in online music services in such a way that the users to not consider them as an add-on.
Alternatively, considering the widespread adoption of social features and social media integration among
digital content services, offering a clean, streamlined user experience without pseudo-social clutter could
be means to differentiate and gain competitive advantage.
To recap, we advise service operators to critically evaluate whether their offering really benefits from
social features and whether the content provided in the service can form a common interest that can
sustain social interaction with other users.
Limitations & future research
The study has several limitations. First, the data were collected only from users of one service, Spotify,
and from one country, Finland. Thus, we recommend future research with a broader contextual coverage.
Second, behavioral intentions do not always translate into actual behavior. Hence, future research could
employ the objective measurement of the use of a service as the dependent variable. Third, in addition to
the future usage intention examined here, future research could examine the reasons behind upgrading a
basic account to premium as well as discontinuing the premium subscription.
What drives continuance in online music streaming?
Thirty Sixth International Conference on Information Systems, Fort Worth 2015 13
Appendix A The Survey Instrument
Construct
Items
Using Spotify...
Ubiquity (Okazaki
and Mendez 2013)
UBI1
...allows me to listen to music with the device I prefer at
that moment
UBI2
...allows me to listen to music wherever I am
UBI3
...allows me to listen to music when it best suits me
UBI4
...allows me to find and listen to a song that I have just
thought about *
UBI5
...makes me not dependent on having music downloaded to
a device*
Discovery of new
music (new scale)
DISC1
...helps me to find music to fit my music taste
DISC2
...broadens my musical taste
DISC3
...helps me to discover music I would not normally listen to
DISC4
...allows me to discover artists/bands that I have not been
aware of before
DISC5
...provides me with music recommendations that suit my
preferences*
DISC6
...helps me to stay updated with new releases by my
favorite artists*
Social connectivity
(new scale)
SOC1
...allows me to see what kind of music other people listen to
SOC2
...allows me to connect with other people with similar
music preferences
SOC3
...allows me to share my favorite music with other people
Enjoyment (Davis
et al. 1992)
ENJ1
...is enjoyable
ENJ2
...pleasant
ENJ3
...is fun
Continuance
intention
(Bhattacherjee
2001)
CI1
I plan to continue using Spotify in the next three months.
CI2
I will continue using Spotify in the next three months.
CI3
I intend to continue using Spotify in the next three months.
*Item omitted from measurement due to loading < 0.7
References
Anderson, C. 2013. Free: How Today's Smartest Businesses Profit by Giving Something for Nothing,
Random House.
———. 2009. Free: The Future of a Radical Price, Random House.
Bhattacharjee, S., Gopal, R., Lertwachara, K., and Marsden, J. R. 2006, "Whatever Happened to Payola?
an Empirical Analysis of Online Music Sharing," Decision Support Systems (42:1), pp. 104-120.
Bhattacherjee, A. 2001, "Understanding Information Systems Continuance: An Expectation-Confirmation
Model," MIS Quarterly (25:3), pp. 351-370.
Blumler, J. G. 1979, "The Role of Theory in Uses and Gratifications Studies," Communication Research
(6:1), pp. 9-36.
What drives continuance in online music streaming?
Thirty Sixth International Conference on Information Systems, Fort Worth 2015 14
Cenfetelli, R. T., and Bassellier, G. 2009, "Interpretation of Formative Measurement in Information
Systems Research," MIS Quarterly (33:4), pp. 689-707.
Cheung, C. M. K., Chiu, P., and Lee, M. K. O. 2011, "Online Social Networks: Why do Students use
Facebook?" Computers in Human Behavior (27:6), pp. 2322-2339.
Davis, F. D., Bagozzi, R. P., and Warshaw, P. R. 1992, "Extrinsic and Intrinsic Motivation to use
Computers in Workplace," Journal of Applied Social Psychology (22:14), pp. 1111-1132.
Dholakia, U. M., Bagozzi, R. P., and Pearo, L. K. 2004, "A Social Influence Model of Consumer
Participation in Network- and Small-Group-Based Virtual Communities," International Journal of
Research in Marketing, (21:3), pp. 241-263.
Eisenbeiss, M., Blechschmidt, B., Backhaus, K., and Freund, P. A. 2012, "“The (Real) World is Not
enough:” Motivational Drivers and User Behavior in Virtual Worlds," Journal of Interactive
Marketing (26:1), pp. 4-20.
Fornell, C., and Larcker, D. F. 1981, "Evaluating Structural Equation Models with Unobservable Variables
and Measurement Error," Journal of Marketing Research (18:1), pp. 39-50.
Katz, E., Blumler, J. G., and Gurevitch, M. 1974, "Utilization of Mass Communication by the Individual,"
The use of Mass Communications: Current Perspectives on Gratifications Research, J. G. Blumler
and E. Katz (eds.), Berverly Hills, CA: Sage, .
Katz, E., Haas, H., and Gurevitch, M. 1973, "On the use of the Mass Media for Important Things,"
American Sociological Review (38:2), pp. 164-181.
Kim, H., Gupta, S., and Koh, J. 2011, "Investigating the Intention to Purchase Digital Items in Social
Networking Communities: A Customer Value Perspective," Information & Management (48:6), pp.
228-234.
Krause, A. E., North, A. C., and Heritage, B. 2014, "The Uses and Gratifications of using Facebook Music
Listening Applications," Computers in Human Behavior (39:0), pp. 71-77.
Kumar, V. 2014, "Making "Freemium" Work," Harvard Business Review (92:5), pp. 27-29.
Kunze, O., and Mai, L. 2007, "Consumer Adoption of Online Music Services: The Influence of Perceived
Risks and Risk-Relief Strategies," International Journal of Retail & Distribution Management
(35:11), pp. 862-877.
Li, H., Liu, Y., Xu, X., Heikkilä, J., and van der Heijden, H. 2015, "Modeling Hedonic is Continuance
through the Uses and Gratifications Theory: An Empirical Study in Online Games," Computers in
Human Behavior (48:0), pp. 261-272.
Liu, C. Z., Au, Y. A., and Choi, H. S. 2014, "Effects of Freemium Strategy in the Mobile App Market: An
Empirical Study of Google Play," Journal of Management Information Systems (31:3), pp. 326-354.
Malhotra, N. K., Kim, S. S., and Patil, A. 2006, "Common Method Variance in IS Research: A Comparison
of Alternative Approaches and a Reanalysis of Past Research," Management Science (52:12), pp.
1865-1883.
Mäntymäki, M., and Riemer, K. 2014, "Digital Natives in Social Virtual Worlds: A Multi-Method Study of
Gratifications and Social Influences in Habbo Hotel," International Journal of Information
Management (34:2), pp. 210-220.
Mäntymäki, M., and Salo, J. 2015, "Why do Teens Spend Real Money in Virtual Worlds? A Consumption
Values and Developmental Psychology Perspective on Virtual Consumption," International Journal
of Information Management (35:1), pp. 124-134.
Molteni, L., and Ordanini, A. 2003, "Consumption Patterns, Digital Technology and Music Downloading,"
Long Range Planning (36:4), pp. 389-406.
Nambisan, S., and Baron, R. A. 2009, "Virtual Customer Environments: Testing a Model of Voluntary
Participation in Value Co-Creation Activities," Journal of Product Innovation Management (26:4),
pp. 388-406.
Oestreicher-Singer, G., and Zalmanson, L. 2013, "Content Or Community? a Digital Business Strategy for
Content Providers in the Social Age," MIS Quarterly (37:2), pp. 591-616.
Okazaki, S., and Mendez, F. 2013, "Perceived Ubiquity in Mobile Services," Journal of Interactive
Marketing (27:2), pp. 98-111.
Park, N., Kee, K. F., and Valenzuela, S. 2009, "Being Immersed in Social Networking Environment:
Facebook Groups, Uses and Gratifications, and Social Outcomes," CyberPsychology & Behavior
(12:6), pp. 729-733.
Payne, J. W. 1982, "Contingent Decision Behavior." Psychological Bulletin (92:2), pp. 382.
What drives continuance in online music streaming?
Thirty Sixth International Conference on Information Systems, Fort Worth 2015 15
Payne, J. W., Bettman, J. R., and Johnson, E. J. 1988, "Adaptive Strategy Selection in Decision Making."
Journal of Experimental Psychology: Learning, Memory, and Cognition (14:3), pp. 534-552.
Raacke, J., and Bonds-Raacke, J. 2008, "MySpace and Facebook: Applying the Uses and Gratifications
Theory to Exploring Friend-Networking Sites," Cyberpsychology & Behavior (11:2), pp. 169-174.
Rayburn, J. D., and Palmgreen, P. 1984, "Meging Uses and Gratifications and Expectancy-Value Theory,"
Communication Research (11:4), pp. 537-562.
Ruggiero, T. E. 2000, "Uses and Gratifications Theory in the 21st Century," Mass Communication &
Society (3:1), pp. 3-37.
Sanchez-Franco, M., and Rondan-Cataluña, F. J. 2010, "Connection between Customer Emotions and
Relationship Quality in Online Music Services," Behaviour & Information Technology (29:6), pp.
633-651.
Sheth, J. N., Newman, B. I., and Gross, B. L. 1991, "Why we Buy what we Buy: A Theory of Consumption
Values," Journal of Business Research (22:2), pp. 159-170.
Stafford, T. F., Stafford, M. R., and Schkade, L. L. 2004, "Determining Uses and Gratifications for the
Internet," Decision Sciences (35:2), pp. 259-288.
Sweeney, J. C., and Soutar, G. N. 2001, "Consumer Perceived Value: The Development of a Multiple Item
Scale," Journal of Retailing (77:2), pp. 203-220.
Teece, D. J. 2010, "Business Models, Business Strategy and Innovation," Long Range Planning (43:2–3),
pp. 172-194.
Tojib, D., and Tsarenko, Y. 2012, "Post-Adoption Modeling of Advanced Mobile Service Use," Journal of
Business Research (65:7), pp. 922-928.
van der Heijden, H. 2004, "User Acceptance of Hedonic Information Systems," MIS Quarterly (28:4), pp.
695-704.
Vock, M., van Dolen, W., and de Ruyter, K. 2013, "Understanding Willingness to Pay for Social Network
Sites," Journal of Service Research.
Wagner, T. M., Benlian, A., and Hess, T. 2014, "Converting Freemium Customers from Free to Premium--
the Role of the Perceived Premium Fit in the Case of Music as a Service," Electronic Markets (24:4),
pp. 259-268.
Wang, H., Chin, A., and Wang, H. "Social Influence on being a Pay User in Freemium-Based Social
Networks," Advanced Information Networking and Applications (AINA), 2011 IEEE International
Conference On, Anonymous 2011, pp. 526-533.
Xiao, B., and Benbasat, I. 2007, "E-Commerce Product Recommendation Agents: Use, Characteristics,
and Impact," MIS Quarterly (31:1), pp. 137-209.
