DEGREE PROJECT IN COMPUTER ENGINEERING, FIRST CYCLE, 15 CREDITS
STOCKHOLM, SWEDEN 2020
Improving an Application
Using Gamification
DIACO UTHMAN
WILLIAM ÅKERMAN
KTH ROYAL INSTITUTE OF TECHNOLOGY
Examiner
Leif Lindbäck
Academic adviser
Fadil Galjic
Industrial adviser
Staffan Bergsten
KTH Royal Institute of Technology
School of Electrical Engineering and Computer Science (EECS) Department of SCS
Abstract| 3
Abstract
It is believed that roughly two out of three users stop using a new application within 90 days. This causes problems for developers, trying to produce a successful application with many users. A strategy that has proven to be beneficial is gamification. However, as a developer, it is difficult to know which functionalities to use. Therefore, this thesis looks at which functionalities can increase the likelihood that an application is successful.
Knowledge about the subject was collected through a literature study. Nir Eyal’s theories about gamification in his book “Hooked” and related works about gamification was two ways of collecting this knowledge. Also, through performing a case study, where existing applications were observed, it was possible to observe applications with gamification in mind.
Thereafter, an attempt was made to implement gamification functionalities into The Bronson Project’s application. Four functionalities were chosen based on the literature study and case study. They are points, own goals, progress bars and achievements. The implementation was evaluated with the conduction of a qualitative interview study, where 9 testers of the application were interviewed. The overall response from the interviews was positive. Interviewees were positive towards the implemented functionalities and believed that the decision to let users decide their own goals was positive.
Finally, it was concluded that two functionalities seem to be important for a successful gamification in an application. A progress is important in the application, and a way to count the progress. The conclusions were drawn based on the interviews, along with our own opinions from the case study.
Keywords
Uppskattningsvis lämnar två av tre användare en applikation inom 90 dagar. Det skapar problem för utvecklare, när de försöker skapa en lyckad applikation med många användare som stannar. En strategi som visat sig vara framgångsrik är spelifiering. Däremot är det svårt för utvecklare att veta vilka funktionaliteter som är bäst att använda. Därför tittar denna uppsats på vilka funktionaliteter som är viktiga för en lyckad implementation av spelifiering.
Kunskap om ämnet samlades genom en litteraturstudie. Nir Eyals teorier om spelifiering i boken ”Hooked”, och relaterade arbeten var två sätt att samla kunskapen. Dessutom genomfördes en fallstudie, där existerande applikationer med spelifiering observerades.
Därefter gjordes ett försök att implementera spelifieringsfunktionaliteter i The Bronson Projects applikation. Fyra funktionaliteter valdes ut baserat på
litteraturstudien och fallstudien. Dessa var funktionalitet för poäng,
framstegsfält, egna mål och prestationer. Implementationen evaluerades genom att intervjua 9 testare av applikationen. Den generella responsen från dessa intervjuer var positiv. En positiv respons som vi fick var att de intervjuade uppskattade funktionen att kunna bestämma sina egna mål.
Avslutningsvis kan man säga att två funktionaliteter är viktiga för en lyckad spelifiering. Den ena är att användaren ska kunna utvecklas på något sätt. Och den andra är att man ska kunna se utveckling på något sätt.
Nyckelord
Acknowledgments| 5
Acknowledgments
We would like to thank The Bronson Project for giving us the opportunity to complete this project. Also, we would like to thank Leif Lindbäck and Fadil Galjic, who has given great advice throughout the project.
Stockholm, February 2020
1 Introduction ... 1
1.1 Background ... 1
1.2 Problem ... 1
1.3 Purpose ... 2
1.4 Society Benefits, Ethics and Sustainability ... 2
1.5 Research Methodology ... 3
1.6 Delimitations ... 3
1.7 Disposition ... 3
2 Theoretical Background ... 5
2.1 Gamification ... 5
2.2 Ways of Implementing Gamification ... 5
2.3 Ideas in The Book Hooked ... 7
2.4 Existing Gamified Applications ... 11
2.5 Related Works ... 16
2.6 Interviews ... 18
2.7 Qualitative analysis ... 19
3 Methodology ... 21
3.1 Methodology Theories ... 21
3.2 Methodology Used in This Work ... 21
3.3 Used Development Techniques ... 23
4 Collective Case Study: Benefits of Gamification ... 25
4.1 Fitrockr ... 25
4.2 Keep Trainer ... 27
4.3 Fitbit ... 32
4.4 Duolingo ... 37
5 The Bronson Project: Design and Implementation ... 41
5.1 Ideas from the Literature Study ... 41
5.2 Ideas from the Collective Case Study ... 41
5.3 Studying the Existing Application ... 41
5.4 Conceptual and Graphical Design ... 44
5.5 Implemented Functionality ... 48
6 The Bronson Project: Evaluation ... 51
6.1 Interview 1 ... 51 6.2 Interview 2 ... 52 6.3 Interview 3 ... 53 6.4 Interview 4 ... 54 6.5 Interview 5 ... 54 6.6 Interview 6 ... 55 6.7 Interview 7 ... 56 6.8 Interview 8 ... 57 6.9 Interview 9 ... 58
6.10 Interesting Ideas from the Interviews... 58
6.11 Opinions about Functionalities ... 59
7 Discussion and Conclusions ... 63
7.1 Evaluation of the Methods ... 63
7.2 Evaluation of the Implemented Gamification ... 63
Table of contents| 7
7.4 Putting the Results into Perspective ... 65
7.5 Reliability and Validity of Results ... 66
7.6 Ethics and Sustainability ... 66
7.7 Future Work ... 66
8 References ... 69
Appendix A: Checklist for Collective Case Study ... 71
The average smartphone user used 25 applications in 2019. Out of these, an estimation of 10 application were used 96% of the time [1]. Back in 2016, the average user used 20 applications [2]. Also, the number of platforms, especially mobile applications, have increased over the last years [2]. In the Google Play Store, which is where the applications are downloaded by Android users, the number of applications has increased from an approximated 100 000 in October of 2010, to an approximated 2 600 000 applications in March of 2019. That is an increase of 26 times. The prediction is that the number of downloads will increase by about 25% from 2018 to 2022 [1]. This makes it even more important for developers to do something special, and thus become one of the applications that the average user will use regularly. Because if users continue using an application, it will obviously grow and become more popular.
1.1 Background
Marketing attracts new users to a platform. Methods such as advertising through social media, starting blogs or video contents are commonly used today. However, it is equally important to look at what attracts the user to continue using the application. In 2018, about 32% of newly attracted users still use the platform 90 days later [3]. This means two out of three users abandon the platform within 90 days. That is a big problem for companies who spend a lot of effort trying to attract users, only to lose two-thirds of them within 90 days. Therefore, there is a need to find a strategy to keep these users. The use of gamification has therefore seen an increase in interest from developers. Gamification is defined as the process of implementing game mechanics in a non-gaming context, such for example daily tasks, or fitness.
1.2 Problem
The issue that many developers face, is how to create a platform that hooks more users, meaning they continue using the platform. A user that regularly uses the platform, will form a habit of using it, which could lead to the user being more likely to stay. Gamification of the platform has been one way to handle the problem. However, the development of gamification might not always be a solution to the problem. There are many examples of gamification failures, due to different problems in the development. However, there are also examples of platforms that have received high popularity through their attempts to create a platform with gamification ideas, such as the “My Starbucks Rewards” [4], that Starbucks developed. Therefore, we need to ask ourselves whether gamification is beneficial when trying to hook new users to the platform. Which factors could affect the outcome of the platform’s fate?
Problem Statement
Introduction| 2
to answer is:
RQ: Which functionality is important for a successful gamification?
1.3 Purpose
The purpose of this thesis is to illustrate and discuss, what functionality is good to implement in an application to make its gamification successful. Conclusions would help companies that want to implement gamification but are uncertain on which functionalities to use.
The Bronson Project, with Staffan Bergsten as supervisor, will assist us throughout the work and to provide us with a platform to develop further, and evaluate. The Bronson Project [5] is a start-up company, that has an idea of developing a fitness platform for users who need help with their workouts. They offer a 12-week training program, that users can follow to improve their fitness. In contrast to many fitness applications, The Bronson Project does not only focus on burning fat. Their vision is stated on their website as:
“Our workout plans and nutrition program focuses on making you stronger, healthier and more functional, as well as leaner and more muscle-toned. We believe in making fitness fun and effective, a program that's boring and leaves you without results is a program that lasts for about 5 minutes.”
1.4 Society Benefits, Ethics and Sustainability
The groups that will benefit from this thesis are developers looking to see what kind of functionality that is good to implement successful gamification. That way the user should keep using the application more.
The word “addictive” comes to mind when the matter of motivating a user to stay is discussed, and naturally, the discussion of moral and ethics follows. An argument is that the word addictive can be used in good ways, as well as in bad ways. For example, a model called The Manipulation Matrix was described in the book Hooked, written by Nir Eyal, on the subject of hooking new users to an application [6]. The model was used to determine whether the platform was manipulative in a good way, or not. Another issue that comes to mind, could be when interacting with test users. If they, for example, are to fill a survey to evaluate the implementation, ethics could be discussed, for example, if those users do not want to fill the survey. To prevent this, the test users could be informed in advance and asked for their consent to give a survey at the end of the testing period.
problem identification, research, development and evaluation, and in our generalization of the methodology, there are 10 steps that will be explained in greater detail in the methodology chapter.
The aim will be to answer the asked question by gaining knowledge and attempting to implement functionalities into the existing platform of The Bronson Project. An evaluation of the implementation will be done by giving our own opinions, as well as conducting interviews. The evaluation will create a basis for answering the question. To put things in perspective, a case study will also be performed based on existing gamified applications. The aim is to understand whether their use of gamification has contributed to their success.
1.6 Delimitations
Throughout this project, some delimitations have been applied. For example, the definition of having a successful application can be widely different. Therefore, it was decided in this project to focus on success in terms of the number of users. That means a successful application would have a high number of users have downloaded and are actively using the application.
Since there are endless ways of gamifying an application, there was a need to put a limitation on which game mechanics to try out. Therefore, this essay will focus on the case study on The Bronson Project. It will try to show what functionality to add when adding gamification to the app.
1.7 Disposition
Chapter 1 aims to introduce the problem, giving background and stating the research question. This chapter will serve as a basis for the whole project. Chapter 2 consists of the literature study that was performed. Chapter 3 will present the methodologies used to achieve the goal of answering the research question and implementing the application. Chapter 4 will present the case study that was performed. Chapter 5 will present the designs and implementations of the application that was developed. Chapter 6 will present the evaluations of the application with an interview study that was conducted. Chapter 7 will present the conclusions, as well as answering the research question and giving suggestions for future research on the subject.
particular, the theories given by Nir Eyal, in his book “Hooked”, will be presented [6]. Examples of existing platforms that use the gamification strategy, will also be presented.
2.1 Gamification
The word gamification is according to the book Gamification by Design [8, p. xiv], the marketing buzzword of our time. There are different correct definitions of the word, which include making games explicitly to advertise a product or creating a virtual world with the purpose of driving a behavioral change in users. These different definitions are all correct since the word gamification aims to make a game out of a non-gaming context [8, p. xiv]. A non-gaming context is the idea of a platform that is not in itself a game, such as a fitness context. The different methods of achieving gamification of an existing platform, are thus what makes the process interesting. Some of these methods will be presented and discussed later.
Although the definition of the word is somewhat unclear, the underlying intention seems clear. It is to introduce game elements in the non-gaming application, to trigger emotions in users. The triggering makes them more inclined to increase their engagement in the application, leading to increased motivation [9, p. 219].
Motivation of Gamification
We know that the motivation of the user towards the application is the most important mission for gamification [9, p. 216]. However, it is important to know exactly what motivates the user. When implementing game mechanics, we know that three components are central in motivating users. These components are the same as for traditional games and include pleasure, rewards and time [8, p. 15].
Because the pleasure of playing a game is often recognized by the user. This affects the user’s feeling of enjoyment, which improves his motivation [10]. Hence, pleasure is a factor that affects how motivated the user will be to keep using the gamified application. A reward is another component that affects the motivation of the user. The reasons for this will be explained in greater detail. The third component, time, is important because the more time a user spends using the application, the more likely it is that he will form a habit of using it. A habit is defined as when the human brain learns to do something without asking why.
2.2 Ways of Implementing Gamification
The making of a gamified service can be made through different methods and game mechanics and a combination of them. Some examples are the use of points, levels, progress bars, leaderboards, badges, challenges/quests and onboarding [8, p. 36]. These mechanics will be defined and described in this section.
Points
Theoretical Background| 6
of points as well. When it comes to these different types, they include experience points, redeemable points, skill points, karma points and reputation points [8, p. 36].
Experience Points (XP)
Experience points (also called XP) are the most important points. A user will earn experience points, depending on how much the application is used. For example, all activities in an application can earn the user experience points, and this will benefit the system because the user develops a long-term behavioral habit. Implementing expiry, meaning the experience points can be accumulated in periods, can also help because the points are resetting at the end of a new period, and users can start fresh again [8, pp. 38-39].
Redeemable Points (RP)
Redeemable points (also called RP) is a second system, which in contrast to experience points, can fluctuate. Instead of earning the points only, the redeemable points can be earned, but also spent on rewards [8, p. 39].
Skill Points
A third point system is called skill points. These complement XP and RP. They provide a way to gain points besides the two previous systems. These points are often given when a user earns some points that include quality. An example where a non-gaming context can implement skill points is when a user earns points for the quality of uploaded photos, and more points for the quality of comments [8, p. 39]. The points for comments would, in this case, be skill points.
Karma Points
This type of points is used to justify an action between two or more users. A player will gain karma points, by giving away points. This means a player will not benefit from keeping any karma points, but only from giving them away [8, p. 40]. When the user needs something, other users can then use their karma points to help back.
Reputation Points
The reputation points can be compared with real-life reputation. If a person has a good reputation, it is more likely that he or she will be trusted. In the same way, a user with a high number of reputation points is more likely to be trusted [8, p. 40].
Levels
Another approach when implementing gamification is the use of levels. This approach gives the user a marker as to how well they are doing in the game, over time [8, p. 45]. Completing tasks and sub-goals are ways of implementing levelling in a game. Through this implementation, the user can feel accomplishment, and thereby an increased gratification for the game [9, p. 216].
Progress bar
The use of leaderboards is another way of implementing gamification. Users tend to know what a leaderboard is, and how to read it [8, p. 50]. We know a leaderboard is an ordered list that is sorted according to some data. For example, based on some points, such as the ones presented previously. The leaderboard can be presented in different ways, and we will present two of these types.
The No-disincentive leaderboard
This type of leaderboard has been used by for example Facebook. The way the leaderboard is set up is that the user will always find himself in the middle, having the closest other users above and below him in the leaderboard [8, p. 50]. This way, the user will know which players are close to him.
The Infinite Leaderboard
The infinite leaderboard has no end, but instead of putting all users in a long list, the list can be narrowed down for example into pages [8, p. 51]. The user will know their position regarding the whole leaderboard but will not have to scroll through the whole list.
Badges
Developers of gamified systems use badges to improve the social promotion of their platform [8, p. 55]. This is because badges can signal status, that someone is doing well. But users could want badges for other reasons as well, for example, a badge can look good to the user. Since, badges mark the completion of a task or goal [8, p. 55], developers can implement badges to create progress of play within the platform.
Challenges/ Quests
Challenges or quests in a gamified system points the user in a certain direction. This is useful for giving directions to the user because without directions it would probably be uninteresting to find some stuff [8, p. 64]. Challenges are also a way to motivate the player because the satisfaction of completing tasks often increase motivation.
Onboarding
The act of bringing a novice user into the system is called onboarding. Developers use this strategy to ease a user into the system, offering help in the first steps of the game [8, p. 59]. Digital games such as Candy Crush and Angry Birds are examples of games that use onboarding, as the first levels are easier to complete. This is effective because this creates the habit-forming desire to win quickly and feel the satisfaction of completing a task in the game [11].
2.3 Ideas in The Book Hooked
Theoretical Background| 8
application.
The hook Model
One of the main concepts in the book is called the hook model. It’s a loop of four steps: Trigger, Action, Variable Reward and Investment [6, p. 8]. This model is used as a reference throughout the book. Figure 2.1 shows the model visually.
Figure 2.1: A graphical representation of the hook model [6, p. 8] Trigger
The trigger is the initial reason why the user wants to use a specific service. The trigger may be external or internal. The external trigger means an external source that describes what a user should do. It can be a notification that makes the user use the service [6, pp. 30 - 32]. Internal meaning the user without external help want to use the service [6, pp. 35 - 37].
A strategy to find this internal trigger by a developer is to ask “why” the potential user want to use the service and continue to ask “why” until it’ll be a root cause to such as an emotion. With this, it is easier to combine the external and internal trigger [6, p. 39].
Action
The action step is defined as the action the user needs to perform to satisfy the trigger. But there is a risk the step is interrupted, not satisfying the trigger. This is due to two things: How strong the trigger was and how easy it is to complete the action [6, p. 44]. There is a lot of techniques to help the user by implementing an easier action [6, pp. 62 - 66]:
The variable reward is the step that satisfies the trigger. The reward is based on the hormone dopamine both by the steps before it and the actual reward [6, p. 71]. There are three general types of rewards: Rewards of the Tribe, the Hunt, and the Self [6, p. 72].
The rewards of the Tribe are based on the fact that most people want to be social. This can be through direct help towards each other such as in Stack Overflow or more discussion-based such as Facebook [6, pp. 73, 74].
Rewards of the Hunt is when the searching or trying to get a bigger reward is a reward by itself. Like scrolling through Pinterest trying to find a nice picture [6, p. 77].
Rewards of the Self are the reward which seems the most obvious for most, this is the goal of the “hunt” if there was one. It’s often represented by some kind of progress bar, either graphical or numerical, even a bank account can have this with higher numbers meaning higher real money [6, pp. 81 - 85]. This is where gamification can be of help.
Investment
The investment step is the part where the user needs to invest something similar to the action but in a long-term way, so the next loop becomes a little better [6, pp. 10, 111, 1112]. This action could be that the user provides preferences or inform what more specific content they like. That way the user can get triggered easier because they want to see some change based on the added content. The reward is also most likely better if something has changed [6, pp. 104, 105].
The investment can also be to simply learn about the service. For example, finding some new tricks that could make the experience of the application better for the user [6, pp. 110, 111].
Should the hook model be used?
Theoretical Background| 10
Figure 2.2: The Manipulation Matrix [6, p. 123].
If the developer of the program wants to use the application and it improves the user’s life, the facilitator quadrant is used. Generally, this quadrant desired because the application would improve both the maker’s and the user’s life [6, p. 125]. If the developer doesn’t want to use it but it will improve the user’s life, the peddler quadrant is used. The likelihood of users getting hooked is smaller since the maker has to rely on what the user thinks [6, p. 126].
If the developer wants to use it but it does not improve the user’s life, it is in the entertainer quadrant. Again it’s very likely that the user may get hooked here since the maker knows what works on them, and by that what should work on the users [6, p. 126].
Finally, there is the scenario where the developer doesn’t want to use it and it does not improve the user’s life. This scenario is called the dealer quadrant. This is simply developed most often to make money. Like Internet casinos, the developer does not want to use them and it does not improve the user’s lives, but still, users will invest money to win something [6, p. 127].
How to improve the hook cycle
When at least 5 percent of users are using the application as their habit, a three-step analysis can begin [6, p. 140].
The first step is to identify a target group for the application, meaning finding a group that is more likely to be hooked to the application.
The second step is to identify the critical activity that keeps the users hooked to the application over a longer period.
In this section, three applications that use gamification will be presented. These applications have implemented three different methods of gamification, out of three different non-gaming contexts. It is not necessary that the developers have concentrated on one game mechanic, but different mechanics and combinations of them can make the way for a good implementation of gamification.
Nike+
Nike+ is a fitness platform that was developed by Nike along with Apple, launched in 2006 [12]. In 2011, the developers had already started to develop numerous game functionalities in the application, such as the use of onboarding. They had also developed a functionality that required the user to perform an action. The action could be described as the action step of the hook cycle which was described in section 2.3. The use of onboarding was more concretely that, depending on how much knowledge the new user had about fitness, he would experience different games. This is a way of adapting the learning part of the application to the user [8, p. 96].
The platform also asked the user to act, by recommending the connection with Facebook, so that the user could post information about workouts. This action was intended to increase the motivation of users, because they would interact with other users [8, pp. 96-97]. For example, a user could post their latest session on Facebook, which meant friends became aware of the activity.
Theoretical Background| 12
Figure 2.3: Screenshot of the total amount of km of running or walking in Nike+
Figure 2.4: Screenshot of the level progress walking in Nike+
Theoretical Background| 14
Figure 2.5: Screenshot of badges in Nike+
Note in figure 2.5 that the user gets two achievements by just doing their first session, it doesn’t have to be long in length either.
This relates to our work since it uses clear progression in the form of both numerical and graphical interface with levels and achievements, trying to make the user run more. A clear gamification inspiration, making the user more hooked.
This is an application that is used all over the world and has become the world’s biggest business network. LinkedIn today has over 610 million members [13]. They have implemented a couple of interesting game mechanics in their platform, to make users stay.
Figure 2.6: Screenshot of the Progress Bar in LinkedIn [14]
LinkedIn also wants the user to interact with others. Since interacting increases the chances that the user comes back, the developers have used a game mechanic to motivate users. Different qualities can be entered by each user, and other users can then confirm that this user has a specific skill. For example, User A enters that he has skills in mathematics, and User B confirms that User A has this skill. This can be translated into a form of reputation point. The more users that confirm that User A has skills in mathematics, the better reputation User A will have [14].
The mentioned game mechanics are examples of where gamification has entered LinkedIn, which from the root is a non-gaming context (building business networks). Also, we know that LinkedIn is one of the biggest business network platforms, which means these mechanics have been beneficial for keeping users.
Google Local Guide
Theoretical Background| 16
Figure 2.7: Screenshot of the Local Guide Profile
It uses the mobile phone's GPS to locate the user’s geographical location. External triggers are provided often when the user has visited a public place such as café, museum or general store. They are then provided with a prompt in the application to review or answer questions about the place. This may be provided through a notification. They also provide a summary on the user’s email for the latest month on the progress the user has had.
In short, through a gamification perspective, it provides points, levels, graphical badges, natural external triggers and rewards.
2.5 Related Works
This section aims to present works that are in some way related to this project. Studies made by Marcus Fröling et al., Kevin Amarasinghage and Tobias Andersson et al. will be presented, along with their relation to this project being explained. The examples given in section 2.4, are applications that implement gamification and are thus also related to this project because they present the concept in practice.
Fröling et al.
Table 2.1: Table by Fröling et al. showing impacts of motivational elements in an
application [18, p. 13].
Theoretical Background| 18
Amarasinghage
The work “Implementation av gamification i digitala system” done by Kevin Amarasinghage on Karlstad Business School [19], had the ambition to show which factors are important for an application to succeed through the implementation of gamification.
An interview was performed on the Online Lead Manager of Actic Sverige AB, on how they have implemented gamification in their platform. Actic is a gym, and fitness company and therefore relevant in the search for elements and functionalities that increase the chances for developing a successful application with gamification in focus [19, pp. 29-33].
Their way of gamification was at the time (2018), based on point systems that gave different levels [19, p. 8]. Their manager claims that this implementation has been beneficial in increasing users in their system. Increased number of users is according to the manager a way of concluding that a system has been successful [19, p. 32]. However, the manager also underlines the importance of continued development of the platform, because users demand new challenges, to stay motivated.
Andersson et al.
A work with the purpose of using motivational elements to develop an application for increased physical activity was done by Tobias Andersson and Joel Åstrand [20]. Through their development, they found that the implementation of gamification is not necessarily enough to develop a successful application, that keeps users hooked. The need for fun is almost as important to implement [20, p. 62]. Physical activity is going to be an aim for the development of The Bronson Project’s application, and thus relevant work for this project.
Furthermore, the evaluation of the application Andersson et al. developed was done by interviewing 4 persons from Betsson [20, p. 53]. These persons were a direct target group for their project since their work included developing the application for the company’s main office. This evaluation provides a guideline for how to successfully evaluate the application, given the time constraints.
2.6 Interviews
Interviews can be either structured, half structured or open structured. Structured interviews follow a set of questions to be answered by the interviewee. Half structured interviews have a set of questions as a base, but the questions can be formulated or be told in a different order. Open structured interviews mostly allow the interviewee to decide what to talk about, as long as it is in the range of the study [21, pp. 34, 35].
An interview can be done with a random selection or with stratification. Stratification selection means the interviewees are selected by a predefined category. An interview can be divided into four parts: Context, introduction questions, main questions and summary [21, pp. 89 - 92].
The main questions are the questions specifically for the interview. This is where the questions about for example an application with gamification, is asked.
In the summary of the interview, the interviewer tells a summary of the interview and asks if the interviewee wants to add something. Any additions should be documented, like the previous steps.
2.7 Qualitative analysis
When analyzing qualitative data, the main interests are words and motivations. The data itself is always a textual document, such as transcribed interviews. There are four different approaches of doing a qualitative analysis: Quasi-statistical methods, template-based methods, editing methods and in-depth methods.
Quasi-statistical methods base themselves on how many words or group of words there is in the different texts. The aim is to understand how important some terms and concepts are for different people.
Template-based methods base themselves on a list of keywords that are searched for in the different texts. The result is given by who says what in the interview.
Editing methods are based on template-based methods, but the list of keywords is created from the texts instead of searching for words in the text.
In-depth methods are used when the results are constructed by the author’s imagination and intuition. The results are reached without an analysis of the words and explanations [21, pp. 114, 115].
3.1 Methodology Theories
This section will present the theories that the methodology of this thesis work derives from.
Bunge’s Method
The main methodology theory that will be followed in this work was developed by Mario Bunge, in his book “No. 1 Epistemology & Methodology I: Exploring the World”, written in 1983. The methodology is broken down into four main areas, namely problem identification, research, development and evaluation.
A generalization of Bunge’s method, with a focus on technological research, was done by Andersson et al. [22, p. 17]:
1. How can this problem be solved?
2. How can a technique or product be developed to solve the problem in an efficient way?
3. What research or data already exists for this type of technique?
4. Develop the new technique with step 3 as standpoint. If this works, jump to step 6.
5. Try with a new technique.
6. Create a model or simulation of the suggested technique or product. 7. Which consequences does the new technique or product bring?
8. Testing the technique or product. If the test is satisfactory, jump to step 10. 9. Identify and fix shortcomings of the technique or product.
10. Evaluate the result in comparison with existing knowledge and identify new problems that occur for future research.
3.2 Methodology Used in This Work
This section aims to describe the use of Bunge’s method in this work. This work intends to follow the generalization of Bunge’s method, which was presented in section 3.1.1. In doing so, the results will give an indication of whether the development of the application satisfies the asked question, which functionality is important for successful gamification.
Problem Identification Strategy - Step 1-2
The first step of Bunge’s method includes explaining how the research question can be answered. In our case, chapter one explains this.
Methodology| 22
Research Strategy, Step 3
Step 3 in Bunge’s method was to identify what research and data that exists for our subject, which is gamification. Concretely, this step is presented in chapter two of this report.
The collection of data is done through a literature study, where the book Hooked, written by Nir Eyal [6], had a central role. Another book named Gamification by Design, by Zichermann et al. [8], was used to give definitions on terms related to the subject. Furthermore, existing theories were gathered through searching for related works. Some works were found and presented, and these gave some good guidance and provided ideas for the solution. In general, the literature study gives a more in-depth understanding of the subject of gamification, which was needed to solve the problem.
Another part of the data collection is given through the observation of existing platforms, that have already implemented gamification. To show how different applications implement different kinds of game mechanics, three platforms have been chosen. Looking at applications with different contexts gives an understanding of the differences in which game mechanics to use in which contexts.
Development Strategy - Step 4-9
Step 4 of Bunge’s method states that the development should be done with step 3 as a standpoint.
In this case, when implementing gamification for an already existing application by The Bronson Project, the literature study was regarded as satisfying for trying to implement the functionality. The related works provided some details on what mechanics increase the chances of gamification being successful, and therefore these along with the Hook Model, described in section 2.3 provided a solid technique for solving the problem.
Since the techniques from step 3 were satisfactory, there was no need to execute step 5 of Bunge’s method.
Therefore, we can go directly to step 6, which was to create a model or simulation for the product. Therefore, section 5.4 was written where a conceptual model and a graphical design was presented. This would be a simulation of the implementation, and what was strived to reach.
Step 7 refers to discussing which consequences the new product brings. This was interpreted as two main consequences in this work. Firstly, the difference between implementing game mechanics, and not implementing them. Secondly, regarding the code of the application, and how the development affected the structure of the application.
Step 9 of the methodology had the intention that any discovered shortcomings or problems would be fixed. We found that one shortcoming was the fact that the results had not been put into perspective. Therefore, the decision was made to perform a collective case study. This case study would consist of looking at other application that had implemented gamification more carefully. The results of this were presented in chapter 4. Furthermore, shortcomings were encountered during the interviews, where some users suggested improvements to the application. However due to time constraint, because the project had a limited time frame, it was not possible to fix all those problems. Therefore, some of them that were discovered were only documented as future work, in the last chapter.
Evaluation Strategy - Step 10
The last step of Bunge’s method is step 10. The intention here was to evaluate the results and put them into perspective. The literature study, the collective case study, and the evaluations from the performed interviews were used to provide a strong basis. The evaluations were presented in chapter 7.
3.3 Used Development Techniques
This section will present the tools for development that are used in this work, such as which software and programming languages were used.
Development Environment
The code will be developed on Windows computers, using the virtual machine software VMware to use a macOS. More specifically the operative system macOS X 10.14 “Mojave”. In the virtual machine, the IDE Xcode is used. The code is executed on Xcode's iOS device emulator (called “simulator” by Xcode) and on a real iPhone.
Programming Languages
The programming language used is Swift, both for logic and graphical interface (GUI). Swift was developed since Apple wanted an easier way to develop iOS apps than using Objective-C, which “is a horrible language to learn” [23] and the result for this objective was the new language Swift. Apple also wanted a new language that could be better optimized for iOS development that isn’t possible with the old Objective-C language [23].
Software Design approach
A set of applications will be used as cases and the aim is to find whether we consider the use of gamification in those applications have made them more successful. A checklist (See Appendix A) was written to create a structure and make sure that each application was evaluated according to similar criteria.
Any opinions and experiences presented in this section will be our own.
4.1 Fitrockr
Fitrockr is described as a social fitness gamification platform and aims to promote and manage corporate fitness activities [24]. The application is used to track fitness activities, i.e. movements of a user during a day. The challenge that this application has taken on, is to let users use any commercial tracking device, instead of a specific such as Fitbit, Misfit or Garmin that are known tracking devices with their own applications [25]. This gives the application a bigger target group.
Fitrockr exists on different platforms. An Apple iPhone 7, running on iOS 12.3.1 was used in this project, along with the built-in Apple Health application as the tracking device. Fitrockr also has a web application version [26], that was easier to use for configuration of the account. The platform was used for 21 days, thus any conclusions drawn are based on the experience and knowledge gained for 21 days.
Experience as a User
Once an account is created and a tracking device is connected to the application, the user will enter the gamified environment. Straight away, as illustrated in figure 4.1, the user will get into a division in the Global Fitrockr Championship. This is the public Championship that everyone will attend. Our user started at division 7.111114 as can be seen in figure 4.1.
The visual leaderboard seems to look like an infinite leaderboard on a first glance. However, when broadening the perspective to consider all divisions on the Global Fitrockr Championship, the leaderboard could be considered a no-disincentive leaderboard. That is because the players in the higher divisions have a higher rank in the Championship. Thus, it seems that it is up to the user to decide which perspective to choose. Of course, a player aiming for the higher divisions would be aware of the broader perspective, whereas a player only aiming to stay in the division would see the infinite leaderboard perspective. Section 2.2.4 defined and described the different leaderboard-approaches in greater detail.
Collective Case Study: Benefits of Gamification| 26
(presented in section 2.4, specifically figure 2.1). The trigger might be that a friend or family member challenges the user to use the application to find out who is the most active person. Since the application itself does not give a reward more than the honor of being crowned champion, each private league can also create a reward amongst them. That would be outside the application’s scope but could strengthen the motivation of those users.
Figure 4.1: Screenshots of the Global Fitrockr Championship and part of the
leaderboard after 11 days of use
Figure 4.2: Screenshot on the Fitness Level bar after 3 weeks of use where our
user increased from level 1 to level 3
Benefits of Game Mechanics in Fitrockr
The use of leaderboards with points and challenges in Fitrockr seems to have given the application a purpose more than fitness activity. The division system is something that has not been encountered by us previously and improved our motivation when using this application.
Another feature, or rule of the game, that improves the experience is that points given in the divisions are adapted to the user’s profile. As stated on Fitrockr’s website:
“Fitrockr will automatically calculate Activity Points based on your activity level and personal attributes such as gender, age, height and weight. This makes all activity and sport types independent of steps or distance and allows you to compete with other users independent of their preferred type of sport. Because 500 steps from a 70-year-old overweight is more worth than from a 20-year-old in perfect shape.” [27]
This could be up for discussion but using the application would mean that the users must adapt to what the developers implement. In a perfect world, every user would find it equally challenging to obtain the points. Since that part will not be a variable to evaluate for us, we assume that the points are perfectly adapted to each user. The implementation of the game mechanics in Fitrockr has thus, in our opinion improved the experience of a non-gaming context of fitness activity. The introduction of divisions and rankings with other users made us more motivated in staying active every day to collect more points than the opponents. In our opinion, the same application without these functionalities would not be as successful, as it is today. Without the game mechanics, we would just have some data on the physical activities of users, that the users would have to compare by themselves.
4.2 Keep Trainer
Collective Case Study: Benefits of Gamification| 28
The application is available on the Google Play Store developed for Android devices, and Apple iOS devices using the App Store. In this work, a Samsung Galaxy S5 Neo with API 23: Android 6 (Marshmallow) was used. The platform was tested for 28 days, and thus the given information will be our opinions about whether gamification in this platform has made it more successful.
Experience as a User
To unlock certain workout programs and be able to download special music in the application, "Calorie Coins" are needed. For example, our user got more calorie coins when opening the application during the test period of 28 days. The more consecutive days that the user opens the application, the more calorie coins will be given as a reward. This is the app’s way to represent redeemable points in the app. Redeemable points were defined in greater detail in section 2.2.1.
Another gaming mechanic implemented in Keep Trainer can be defined as a progress bar. When the user completes exercises, a progress bar will fill. When the application is running in portrait orientation, it is visible as in the bottom of Figure 4.3. Also, the number of completed exercises is visible to the user, as in figure 4.3, where the user has completed four out of 6 exercises.
Figure 4.3: Screenshot of a workout exercise in Keep trainer in portrait mode
Figure 4.4: The plan view in Keep Trainer
Keep Trainer has also implemented something they call “Training Trophies”. Those trophies have been separated into three categories named discipline-, exploration- and animal spirit-trophies.
Collective Case Study: Benefits of Gamification| 30
Figure 4.5: Screenshot of the Exploration Trophies in Keep Trainer
Figure 4.6: Screenshot of the leaderboard in Keep Trainer Benefits of Game Mechanics in Keep Trainer
The Calorie Coins make sure the user opens the app every day [8, p. 123]. The user will keep training if they want to hear a piece of specific music that happens to need Calorie Coins to play while training.
The progress bar on the program helps the user to understand how much of the current program that is left. The personalized training program makes every user that has a functional body to have a good onboarding, independent if the user is used to training or not.
The training trophies are making sure the user in the beginning train more days in a row, walk more steps and try more workouts, this is because that is what is needed to receive the Training Trophies. The choice of the apps Samsung Health and Google Fit when counting steps is good since every Samsung phone comes with those apps preinstalled. But the fact that a connection is needed might not be appreciated by the user if the user only wants one training app for both workout programs and count steps.
Collective Case Study: Benefits of Gamification| 32
In every aspect of the app, it shows the user that it’s making some progress in some way keeping the user to use the Keep trainer app. Without the functions described the user will not see that it’s making progress. The functions by themselves aren’t unique in training apps, but it’s the combination of them that makes the app special as proven by its reward. Also, at every view in the app, the user can see some of its progress, which make the user remember that the whole time as they use the app. To conclude, the app has many gamification mechanics helping the user to keep training more and if possible, every day. The first thing the user encounters when using the app is the Calorie Coins making the user get a variable reward only by completing the profile i.e., they use The Endowed Progress Effect defined in section 2.3.1. The central thing about the app is progression, both in small and larger parts. This makes the user keep using the app and kept us more motivated to use it than if the progression weren’t there.
4.3 Fitbit
On the website, it is stated that “Fitbit tracks every part of your day — including activity, exercise, food, weight and sleep”. It is available on Android, iOS and Windows [30]. With the app and a fitness tracker watch or smartwatch, the watch can automatically track the user’s steps they do during the day and track their sleep. The user can also manually track their more specific training through the device and manually track information about weight & nutrition in the app [31].
The application is available on the Google Play Store developed for Android devices, for Apple iOS devices using the App Store and as a website. In this work, a Samsung Galaxy S5 Neo with API 23: Android 6 (Marshmallow) phone and a Fitbit Charge 3 fitness tracker watch was used. The platform was tested for about 8 months, and thus the given information will be our opinions about whether gamification in this platform has made it more successful.
Experience as a User
The main idea of the Fitbit app is to set goals related to important parts that are important for the user’s health. These parts are the amount of steps the user takes in a day, the distance the user takes in a day, the number of active minutes in a day, the amount of calories burned in a day, how many times the user walks a little (minimum 250 steps) each new hour on the clock, weight progress over a longer time and how long the user sleeps. If the user doesn’t want to know some of these parts, it can easily be changed in the app’s settings.
Each of these is tracked every day, the user can access the history of each one of them. The history can be shown as numbers and as a bar graph which can visually display:
• The latest week of tracking • The latest month of tracking
For both, the bars represent the tracking of one day
• The latest three months of tracking, where one bar represents the total tracking of one week
user can easily understand how well they did.
Figure 4.7: The 3-month view of the number of steps taken in Fitbit
If the user has a fitness tracker watch, the user can use it to track a specific workout which isn’t tracked automatically. After selecting an exercise, the user can choose to set a goal for the workout. When the user stops the workout the time in the workout is added to the active minutes.
The user can choose its activity goals that they try to reach each day – these areas mentioned steps, distance, calories, floors and how often the user wants to walk a little each new hour. When any of the first four goals are reached the bar for the day in the bar graph for this day get a star. If the user moves a little during an hour it’s presented as a part of a circumference on the app’s dashboard being colored. The users can also add goals for the number of calories consumed, how much water it wants to drink, how much the user wants to weigh and how long the user wants to sleep.
Collective Case Study: Benefits of Gamification| 34
Figure 4.8: The dashboard of the app showing all the goals
To help the user be more active there are badges in the app, but it isn’t shown when the user will get them until they are taken. For example, the app keeps track of the amount of km taken in total, when a certain number of km is reached the user get a badge, but there is no way for the user to know when it will get the badge.
Figure 4.9: A new badge taken in Fitbit
To make the user walk more it has challenges with others that use Fitbit. It’s based on the idea of “The one that walks the most during a certain time wins a trophy in the app”, the “certain time” can be one day, one workweek or one weekend. The user gets notified when the other users in the challenge make progress and if the other users take the lead over the user.
As stated in section 2.2.6 challenges motivates the player. Because other people are competing the most people want to walk more, the notifications also remind the user to move more so they win.
There is also another type of challenges called “Adventures” these can be done alone or with more users. If the adventure is done alone the user can “collect treasures” as they walk, these treasures are information about health, or information or a question about the location their avatar is in.
Collective Case Study: Benefits of Gamification| 36
Figure 4.10: An Adventure Race Benefits of Game Mechanics in Fitbit
Regarding the graph bar, if the user is looking at how many steps, they’ve taken the latest week. Then they can get more motivated to walk more this week since every day they’ve walked they can see the bar for this week getting bigger, then they can compare to the last week visually.
The workout goals make the user work more since they know they want to reach the goal they set up before the workout. Making them train harder than otherwise [32]. The normal challenge should make the user walk more since they can compare themselves to others which motivates people since they know how much they need to walk to catch up with the players that have walked more than them [33]. It’s the same thing for the adventure challenge except the user will be motivated further to walk with the virtual map.
application are Luis von Ahn and Severin Hacker.
Like many other applications, Duolingo has been launched on different operating systems, to have a wider target group. However, in this case, the iOS build of the application has been used on an iPhone 7 running on iOS 12.3.1. The application was used for a total of 10 days, and any conclusions will be based on this period.
Experience as A User
The first thing that happens is that the user specifies which languages are to be learnt. In our case, the decision was made to learn the Swedish language. Once the setup is complete, the user will be shown its progress (shown in Figure 4.10). The progress consists of different categories, for example, Basics, Food, Animals and Plurals.
Figure 4.11: Shows the start page of a new user’s progress in the Swedish
language
Collective Case Study: Benefits of Gamification| 38
Figure 4.12: Shows prompt when a user earns new XP
The Experience Points seem to have different purposes in this application. For example, the user can get different so-called achievements by earning more of these. An example is the Overachiever achievement that is given to the user once 200 XP has been earned in one day. This is shown in Figure 4.12.
Figure 4.13: Screenshot of achievements, with focus on overachiever that user
The first league that the users enter, is named “Bronze League”. Assumingly, since our user did not win the entered league, the user will enter a higher league if he/she wins the current league. Each iteration of the leagues is for 7 days. Our user stayed in the bronze league for two iterations.
Figure 4.14: Screenshot of the leaderboard that our user entered Benefits of Game Mechanics
The game mechanics, mentioned in section 4.4.1, that have been implemented to Duolingo are mainly Experience Points and Leaderboards. Achievements also seem to be a way of giving so-called badges to the user, which was also a form of game mechanic mentioned in section 2.2. However, the use of achievements was not entirely understood by us.
Collective Case Study: Benefits of Gamification| 40
Further, the use of leaderboards in Duolingo in our opinion improves the motivation of users. However, as we entered the Duolingo application and entered the so-called bronze league, there was no option to enter a private league with other chosen users. Only being able to compete versus strangers is something that we thought reduced our motivation for the application. We realize, however, that some users might find it more motivational to compete with users that they do not know in real life (as we call stranger).
As mentioned, achievements have also been noted as a game mechanic in Duolingo. Our opinion is that achievements must be relevant to fit into an application. No achievements were obtained during the test period of this application, and therefore no direct conclusions can be drawn regarding the feature.
Also, a conceptual model will be presented. Finally, the resulting implementation will be presented.
5.1 Ideas from the Literature Study
Various ideas were given throughout the preparation work. The literature study gave a good basis, particularly the related works presented in section 2.5. Conclusions given by Fröling et al. told us that points and badges seem to have a positive impact on motivating users. Furthermore, section 2.1.1 discussed the importance of motivation when discussing successful applications, through implementing gamification.
5.2 Ideas from the Collective Case Study
The collective case study in chapter 4 described some successful applications who have implemented gamification mechanics.
For example, the FitRockr platform implemented a point system where users were ranked based on divisions. This idea seemed good to us and was the main idea in our minds that could have been implemented. Thus, we believe that this case study was a good way to get a feel for the existing applications, and to feel the benefits or disadvantages of gamification.
Moreover, different case studies helped in finding a good way to review an application. This would prove important when conducting the interviews in the later stage of the project.
5.3 Studying the Existing Application
The Bronson Project: Design and Implementation| 42
Figure 5.1: The login view that user will see upon opening for the first time
Figure 5.2: Workout view
The Bronson Project: Design and Implementation| 44
Figure 5.3: Notes view
The rightmost icon in the bottom of the page will redirect the user to the Bronson Project webpage.
If the user presses the icon of a human head and shoulders from the workout view a new view shows the most essential information about the user.
Figure 5.4: User settings view
5.4 Conceptual and Graphical Design
This section will present the conceptual model that was used in the implementation, and some graphical designs that were drawn before implementation. This information was developed through brainstorming and communication with The Bronson Project’s developers.
General Summary of Implementation
A summary of what was decided to implement into the application. These summaries can be seen as objectives and were the aim throughout the implementation.
• Points – The user shall be able to collect points for some actions.
• Goals – The user shall be able to set own goals that he/she will reach to earn something.
• Progress bar – The user shall receive feedback on how they are doing in reaching their goals.
the point system is: 1 set done gives one point, 1 exercise done gives 5 extra points and 1 workout done gives 10 extra points.
From section 2.2.6 and the analysis of Fitbit in section 4.3, it’s regarded good to have the user set their own goals. The goals for this application are the number of points done this program and this week. A goal that the user can’t decide is doing all exercises in the program.
All these goals will have a progress bar to show how close the user is to reach the goal. It’ll be represented as a circle border filling up around with a color as the user progress. When the goal is reached the circle border will be filled with color. All the applications in the collected field study in chapter 4 use progress bars.
This should be implemented on a new view only meant for gamification features accessible from the footer.
On another view accessible from the previous view there will be a view regarding badges in section 2.2.5 these were presented. Keep Trainer, Fitbit and Duolingo had badges which motivated the user.
Graphical interface to be used
The Bronson Project: Design and Implementation| 46
Figure 5.5: Design of the main gamification page
Figure 5.6: Design of the popup when the user presses a point progress
The Bronson Project: Design and Implementation| 48
Figure 5.7: Design of the achievement view
5.5 Implemented Functionality
This section will present the implemented functionalities.
Points
The first functionality that was added to the application was a point system. As described in the conceptual model, the intention was to implement a point system that would be easy to count. Since, there already existed an application which implemented circles for every set that a user completed within an exercise, it was natural to use these circles as a way of collecting points.
Therefore, the decision was made that points would be collected for every set that a user would complete. For example, if a user completes a full exercise consisting of 4 sets in total, then he/she will get a point for each set, and 5 points for completing the exercise.
Progress Bars/ Goals
Figure 5.8: Shows the view for progress bars of a user with an ongoing program Achievements
The Bronson Project: Design and Implementation| 50
Figure 5.9: Showing which achievements have been obtained by the current user.
(see Appendix B).
Furthermore, this chapter will also present some ideas that were found in the interviews and opinions about functionalities.
6.1 Interview 1
The first interview was done with the founder of The Bronson Project, Staffan Bergsten. Since this person has a lot of knowledge from the business point of view, his opinions are credible and relevant. Staffan is training 4 to 7 times each week.
Introduction
Staffan mostly does strength training and running for general health, keeping a good physical appearance and to challenge himself. The reason he starts to use some apps over others is how easily the app make him understand the app’s function and to make him interested i.e. easy onboarding. He also wants to see his progress in the app and has as few interactions as possible. He uses his phone while training when tracking running statistics. He knew about gamification and had seen The Bronson Project app before.
Comments on the Bronson Project app
On the question of whether The Bronson Project is a fitness game, he says no. However, he is certain that the implemented functionalities can help the application to become more attractive and to motivate its users. He has used the application before our implementation and believes that the difference compared to before is that the application can track progress with the use of points, to motivate the users to reach their goals.
Furthermore, Staffan felt that the application lacked connectivity between users. For example, he mentioned the fact that there is no way to compare points to other users. He also lacked a way to have a weight goal on exercises that uses weights. Then possibly have an increase of weight on that exercise the weeks after until the goal weight is reached. He thought this would “add a good set of value and motivation to keep going towards a specific goal. “.
Set a weight goal for say deadlifts and be able to get how much weight you need to add every week to accomplish the goal.
When giving a score on the gamification features on a scale from 1 to 6 were 1 is unimportant and 6 is very important Staffan gave:
Points: The score 5 because it was easy to understand, and the users can easily compare their points with others when two persons locally use the app.
