Uber Or Heetch: A Comparative Study On Desirability And Usability .

Europe. It is currently available in Paris- and five other cities in France- Brussels, Milan, and Stockholm. It once again provided a service where any private person with a car can sign up and provide rides for consumers, and another service called Heetch Pro that required drivers with taxi permits. Heetch therefore offered prices exceptionally lower than other competing services like Uber and traditional taxis, while also offering a service in that price range with Heetch Pro. Heetch managed to work around the restrictions that eventually stopped the operation of UberPOP by declaring itself as not a taxi service, but rather solely as a digital platform that provides a way to connect people to nearby drivers that are willing to provide them transportation. And that the payment users conduct is simply a “donation” to the driver and not a taxable income [9]. It operated this way for two years, technically illegally according to Transportsyrelsen. The app encouraged the users to sit with the driver in the front seat through the use of prompt screens with messages such as, “It’s more fun to sit up front” since they advertise that the ride should be as if you are "getting a ride from a friend”[9]. But a more practical reason for this is to not arise suspicion from patrolling police and have the driver get pulled over due to having a lone passenger in the backseat of a private vehicle with no clear taxi decor. Heetch initially was marketed to provide rides in the unconventional late hours, with it only being available from 20:00h to 6:00h. The target demographic being young adults with company slogan being, “Enjoy going out”. This made it popular with young adults with 85% of its users being people under 25 [16]. Early 2018, Heetch had to discontinue the operation of Heetch Classic, due to legality and is now only operating Heetch Pro, the model to also require their drivers to have taxi permits. It is now currently operating all hours of the day, and is functioning for the most part the same way as Uber with driver’s owning a taxi driver’s license as well as a taxi permit for using their private cars. Uber and Heetch have the same fundamental purpose and functionality, with the only major difference being design and specific features that the apps have to offer. But there is a bigger disparity in terms of history and recognition. Uber is the more established app, having been around in Stockholm for a longer period of time. It is also seen as the pioneer in ride-sourcing as it is the first app that normalized this business model. Uber first launched in 2009 in San Francisco where the first car ride was booked and completed. It has since then grown to 65 countries in over 600 cities. [20] The app, along with other applications that offer the same services, has faced numerous resistance in multiple cities, one of them being Stockholm, even after having shown to be an effective alternative to traditional taxis. Due to reasons that it provides a safer, more secure, and more fairly priced mode of transportation. This opposition predominantly came from taxi unions and other transport boards in respective cities with long standing regulations on what can and cannot be submitted under taxi laws. The requirements of a taxi driver’s license and taxi permit is a very costly and long procedure that is not favorable to private drivers[16], which was a major selling point for when Uber was initially launched, that any private individual in ownership of a vehicle can earn easy and simple income by being a driver for Uber. Although the transition to the other model has made the service more reliable and trustworthy to its users due to the driver’s being licensed and therefore experienced. Heetch is facing much of the same resistance in Stockholm, but is operating in a much smaller scale with 9 cities compared to Uber’s over 600. It has also not been operating in Stockholm as long as Uber has. It did provide a fresh and needed competition to a market that was all but monopolized by Uber. Other than the incredibly low prices it was offering, Heetch also entered the market with a different attitude and sense of operation. The initial reaction was distrust and cynicism, as is when a good or service that normally have approximately the same level of prices across all competition is being offered well below that price range. Due to the restrictions set by the transport agency, disallowing the former operation to continue, Heetch and Uber’s prices are now much closer to each other. 2.2 A comparison in features and design It is hard to mistake the Uber app for Heetch and vice versa. There is a clear contrast in the overall design and aesthetics. Most clearly illustrated in their respective color schemes. Figure 1. Start screens of Uber (left) and Heetch (right) Upon opening Heetch, a pink welcome screen with the logo appears before it opens up into the interface. Within the app, this color scheme continues, being mostly white with pink accents on important buttons. The main page is a map showing the user’s location with a search bar on the bottom of the page for inputting the pickup location. Below this is a pink box with “Confirm pickup” written on it. On the top left corner is a button for opening the menu page. Under the menu are the tabs; Profile, Payment, Free rides, and Help. The app allows you to input your location, followed by the destination, asking for confirmation after each entry. It then shows a price estimation for the ride before once again asking the user for confirmation on booking the ride. Once the user does so, the app tries to connect the user with a nearby driver and when connected, shows how long it would take the driver to reach the user’s location with a profile of the driver and an image of their vehicle. Other than booking rides, the app’s other functionality includes allowing the user to edit their profile and change payment methods. It is straightforward and uncomplicated. Uber on the other hand has almost the opposite color scheme. Much like Heetch’s welcome screen, Uber’s is of a solid color screen with the logo in the middle. The difference is that Uber’s color scheme is darker consisting of whites, blacks, and grays. The main page is also of a map showing the user’s location and the menu button on the top left. A key difference is that Uber’s input is at the top of the page and it already assumes the user’s location without needing it to be inputted. The input bar therefore has a prompt that says “Where to?” in which the user then types in their desired destination. In the case of the GPS being inaccurate

and providing a false location, or if the user wants to be picked up in a different location than where they currently are, the pickup location can also be manually typed in. Once both addresses are confirmed the price estimation appears on the bottom of the page for the different types of service the user wants. UberX, which is the cheapest alternative, UberBlack a pricier alternative, UberXL, for when a bigger vehicle is needed, and UberLux, the most expensive service which provides a luxurious town car. In the same page the user can also modify the payment method, and they have the option to schedule the ride if they don’t necessarily want to be picked up right away. At the very bottom of the page is a confirmation button accented in black. Once all the details are confirmed by the user, Uber connects the user with the closest driver and once a connection is made, gives details on when the car will arrive, as well as the profile of the driver with information on their car: model and license plate number. Clicking on the driver’s profile there are other features available, users can split the fair with other users and share the ride, also known as UberPool. It is also possible to add multiple stops to the route. Under the menu options, Uber’s is much like Heetch’s with the exception of a number of other features. Other than being able to edit their profile and changing the payment method, Uber users can look through their ride history with information on who the driver was, the duration of the ride and how much it cost. Uber also has a settings tab where a user can list locations as “favorites” for easy input and set up family profiles. Both Uber and Heetch offer fundamentally the same service, a platform for connecting users with drivers that are available to provide transportation. Heetch has stripped their app down to the bare necessities while Uber has filled their app with other features they deem useful for their users, both drivers and riders. 2.3 Concepts 2.3.1 Usability This study will use the ISO definition of usability which is: “ extent to which a system, product or service can be used by specified users to achieve specified goals with effectiveness, efficiency and satisfaction in a specified context of use”[10]. Usability focuses on the productivity and effectiveness of a user interactions with a product or service. It is more objective in nature, using observations and tools to measure a user’s interaction to analyse the effort required to perform specific tasks. Usability of a product can be measured in a variety of ways, either by using questionnaires about user satisfaction or to measure performance of doing a task, by timing it or other kinds of measurement. The System Usability Scale (SUS) is one such questionnaire that is commonly used. One strength of SUS is that SUS has high reliability coefficient factor above 0.90 [ 14] and it also gives a reliable score with lower numbers of participants [19]. 2.3.2 User Experience The term User experience (UX) started gaining traction in the 1990s, most notably by Don Norman during his work at Apple [15]. UX is a broad subject that covers a wide area and can be hard to characterize as there are no general consensus over a definition or what should be included within UX [12]. For this study, we will be using the iso-standard 9241-210 which defines UX as: “person's perceptions and responses resulting from the use and/or anticipated use of a product, system or service” [11] . UX compared to usability takes a holistic approach towards user interaction focusing on the many different impressions a user might have over time while interacting with a system. Whereas usability evaluates factors such as the effectiveness and task performance, UX is subjective [13] and covers subjects such as such as the aesthetics, hedonic quality and experiential values of a system [8]. The usability of a product is also taken into consideration since functionality of a system is a necessity for a good UX. As such, when designing the UX for a product, usability can be viewed as a subset of UX. Analysing UX can be done by using tools such as using interviews, questionnaires or user observations to analyse the affection, aesthetics, affections or enjoyment of a product [2]. As mentioned earlier while Uber and Heetch provides the same service, they have different backgrounds, design and functionality and considering that, there should be a difference of the UX and usability between the two apps. 2.3.3 Desirability Another approach with analysing user experience is with measuring its desirability. It is the process of measuring the intangible aspects of how desirable, fun, and pleasant a system is [3]. This is commonly done through likert questionnaires but it has the disadvantage of not accurately representing the true attitudes of the respondents due to its limited nature. As an alternative, two researches from Microsoft developed a method that measures desirability in more detail, the Desirability Toolkit. The toolkit consists of 118 reaction words that they have specifically chosen to be relative to user experience testing. Since there is usually bias to positive feedback in other usability testings, they made sure to include negative reaction words as well. The advantages to this method is that it is designed to be flexible and to allow the test subjects to narrate their own story and idea about their impressions of the app. This leads to participants providing a more honest and elaborate response where they can more easily give both positive and negative feedback [3]. The toolkit was designed to measure two types of desirability, functional and emotional. For this study, only the emotional desirability derived from the aesthetics and graphical design of the apps are focused on. The method is therefore altered to only focus on visual design. Out of the 118 words in the toolkit, only the ones relevant to visual design and appeal will be utilized. This cut down the number of reaction cards to 53. 3. METHOD Measuring and quantifying user experience is a broad and complex task that requires various inputs and angles of observation. For the purpose of this study, the particular characteristics within user experience that was chosen to be examined and compared are usability and desirability in the category of visual design and appeal. The first part of the experiment was the analysis on desirability, which was adapted from a method developed by Microsoft for measuring “fun” and “desire”, satisfaction when handling a software [3]. This method makes use of reaction cards where the test subjects will be given a list of reaction words from the “Desirability Toolkit”. This method was chosen due to its

flexibility and being designed to allow the test subjects to narrate their own story and idea about their impressions of the app. For the analysis on usability, a questionnaire adapted from the system usability scale, SUS questionnaire, was chosen. This is a questionnaire consisting of 10 Likert-type statements, with a scale of one to five as answers. One being Strongly Disagree and five Strongly Agree. The questionnaire is designed to have alternating positively and negatively worded questions. Odd-numbered questions are positively worded, and even-numbered questions are negative. 3.1 Participants A total of 16 subjects participated in this study. All of whom are undergraduate students in the Media Technology program of KTH ranging from first year students to third year. Due to this background, everyone who participated are well informed or in some way or another knowledgeable about the topic of user experience revolving around app development. Current, past, and upcoming courses such as Reflective Design Process, Media, Technology and Culture and Human-Computer Interaction has provided the participants with knowledge for analyzing technological services and tools within different aspects. Generally around design optimization, usability, development, and more. All of the participants owns and regularly uses a smartphone. cards, two categories of positive and negative cards were made. That is, words such as ‘Fun’, ’Creative’ and ‘Engaging’ were considered positive cards and ‘Confusing’, ‘Distractive’ and ‘Inconsistent’ were considered negative cards. This provided a measurable and quantitative result from the reaction cards, as well as qualitative from the interviews. The questionnaire has a standardized way of scoring by design. The subjects would have answered each question from one to five, depending on the degree of their agreement to the statement. The scores for the odd numbered questions get one subtracted from them, and the scores for the even numbered questions get subtracted from five. The new values are sum up together to get a total score and finally multiplied by 2.5. From this, the average scores were derived for each app and compared. 4. RESULT The results analyzed are of three categories. The first being the introductory survey, the desirability toolkit, and lastly the usability questionnaire. All the participants ( n 16) participated in all three of these categories. 4.1 Introductory survey 3.2 Implementation All 16 of the participants are familiar with the concept of ride sourcing and the general definition of what this service implies, but of the total number of participants 10 are familiar only with Uber, 4 recognizes both Uber and Heetch, while the remaining 2 have not come across either apps. Before starting with the experiment, the participants were asked to answer a short survey regarding any previous experience with either services. This was done to discover a correlation with each participant’s results. If previous experience with either app show a bias, or if familiarity will have an effect on the SUS score. Among those who have knowledge of the existence of either Uber and Heetch ( n 14), only 11 have used the services. 10 having only used Uber and one participant having experience of using both. The test subjects were given one minute for each of the application to explore the apps and to only consider the general design and aesthetics. Following this, they were presented with the reaction cards and they were tasked to pick the words that they associate with their experience and impression of the apps. Afterwards, a semi-structured interview was conducted asking about the choices they made. An audio recording was made to store and access these interviews for analysis. For the last question on the survey about if they would consider using either services, all but one participant answered yes. The second part of this experiment was the usability questionnaire. Before they could answer the questionnaire, the participants were provided with tasks to accomplish within the applications. These are simple functions that represented the actual process of utilizing the apps. Tasks along the lines of booking a ride from a specific location to another, checking prices, changing one’s profile, etc. A complete list of the tasks is available in the appendix. Upon completing these tasks, the participants were given the questionnaire. The participants had three sections to place the cards; Uber, Heetch, and for both apps. Figure 1-3 shows in the appendix histograms representing the frequency of cards selected for the respective categories. The section under “Both” does not represent the total occurrence of a card being selected for Uber and Heetch in separate occasions, but rather the number of times a participant selects a card and chooses to put it under the section as a reaction for both apps. This means that a card could have appeared in two occurrences of the experiments (from two different participants) once for each app, but this is not indicative of the card appearing under the “Both” section. Both parts were implemented twice, once for each app. It was conducted on a Samsung Galaxy S8 mobile smart phone running on Android. The phone has screen dimensions of 6.2 inches diagonal. 3.3 Data analysis Three data sets resulted from the experiment: the introductory survey, reaction cards result, and SUS score. For the reaction 4.2 Desirability Toolkit- Reaction Cards Among the 53 reaction cards that were made available to the participants to choose from, only 40 were chosen in total by the participants throughout the experiments. The other 13 cards were not used by a single test subject.

Table 1. Most placed cards by category Uber n Heetch n Both Professional 9 Simplistic 6 Clean 5 Confusing 4 Understandable 5 n Simplistic 5 Clean 5 Clear 3 Familiar 4 2 Clear 3 High Quality 4 Clean Trustworthy 4 Fun average as well as the good category and both apps fall short of the excellent category. 2 Understandable 3 Table 1 shows the most placed cards with the name of the reaction card on the left side and the number placed on the right. For Uber alone, a total of 31 words were selected with Professional having the greatest frequency at 9 times. This is followed by Clean and Understandable at 5, High quality and Trustworthy at 4. Heetch on the other hand, had 29 words in total. Simplistic and Confusing being the most used at 6 and 4 respectively. The participants utilized a total of 19 different reaction cards for the “Both” section. Simplistic and Clean being selected the most with a frequency of 5, followed by Familiar at 4. The most used words for the entire duration of the experiment are Simplistic and Clean, both appearing 12 times out of 16. This is followed closely by Professional at 10, and Understandable at 9. It is interesting to note that Simplistic, Clean, and Understandable are distributed almost evenly among the three sections, but Professional was selected 9 out of those 10 for Uber and only once for Heetch. There was no occurrence of a participant choosing it for both. Table 2. Percentage of selected positive and negative cards Baseline Uber Heetch 60,4% 86,2% 73,9% - 39,6% 13,8% 26,1% Table 2 shows the percentage of cards that are considered either positive or negative. That is, words such as ‘Fun’,’Creative’ and ‘Engaging’ were considered positive cards and ‘Confusing’, ‘Distractive’ and ‘Inconsistent’ were considered negative cards. The table shows that both apps got more positive cards than the baseline, but people placed more positive about Uber than Heetch. In fact, people placed close to twice as much negative cards on Heetch compared to Uber. A semi-structured interview was conducted each time after this part of the experiment encouraging the test subjects to explain their thought process in choosing reaction cards. These interviews will be further demonstrated in the Discussion for the acquired quantitative results. 4.3 SUS Questionnaire The questionnaire provided very unambiguous results. As this questionnaire is standardized, there is a systematic method of calculating the scores. Uber had an average score of 80.3 and Heetch 72.0. Research comparing SUS-scores from a thousand surveys shows that the average score falls at 65.9 for mobile devices with the above 71.4 rated good and above 85.5 rated excellent [1]. As shown in figure 4, both apps scored above the Figure 2. Average SUS-score Finally, for the deciding question on which app the participants would prefer to use, 14 out of the 16 participants chose Uber, one chose Heetch and one choose neither of the two apps. 5. DISCUSSION The results show that overall, people felt more positive about Uber over Heetch. One recurring statement from the participants during the semi-structured interview, where the participants were asked to elaborate on their decision making, was how similar and interchangeable the two apps were despite the contrasting results on the most placed reaction cards on respective apps. There were two occurrences of the participant placing the majority of the cards they have picked on the both category, with only one or two distinct cards placed under the apps. The participants expressed how both apps function and follow a similar layout, with both main interface showing a map with animated cars to show proximity and availability. The menu button being on the top-left corner, the menu options being almost identical to each other, and how both apps offer and provides fundamentally the same service. This is demonstrated with the selection of cards placed under the Both category. Cards such as “Clear”, “Familiar”, and “Understandable”. As with the numerous similarities, their differences also abound. These created subtle but impactful distinction with how the participants perceived their experiences with each app. In reference to the section about the differences in design, one such seemingly inconsequential but polarizing design choice of having the search bar only be of one input at a time and placed from the top to the bottom of the screen garnered Heetch most of its “Confusing” card. Coupled with the bright colored Confirm button, a number of participants were left unsure and hesitant to continue. This is due to the Heetch’s design choice to have the user Confirm each input, both pickup and drop off separately, when in the process of booking a ride. Of the four participants who selected “Confusing” for Heetch, all of them related this to the input of location. The second key variation between the apps were their color choices and animation. Another distinction that the participants reacted to. Uber obtained a high number of cards relating to the level of quality it exuded. “Professional”, “High Quality”, and “Sophisticated” being selected by numerous

Stockholm— Uber and Heetch, using a combined method of measuring desirability and usability to analyze which factors influence a user's choice of app. The results showed Uber to have better desirability and usability, and 14 out of the 16 participants chose Uber as their preferred ride-sourcing app. Keywords