#reflection03

Task 1: Finding a specific genre

For new users:

(1) Open Libby App (2) Search for  the ‘Romance’ genre (3) Find a book that is in the ‘Romance’ and ‘Fiction’ genre

We found that finding the ‘genre’ section is quite difficult due to the way the homepage was being arranged. The discoverability of the genre section is quite low and when the user manages to click into the page, it shows a generic ‘romance’ genre but fails to tell the user that besides that genre, it nested other genres such as ‘fiction’, ‘historical fiction’ etc. This is due to the way it was placed, which users might miss. The buttons are also clickable which does not look like it could at a glance. Thus, we feel this could be something that can be improved for better user experience. 

From this, we go on a step further and identify insights which could help us generate HMWs:

Task 2: Finding similar titles based on previous reads

For existing users:

(1) Open Libby App (2) Find similar book recommendations based on previous reads

We find that Libby’s book recommendations are very generic as it will bring us back to the genre section based on our previous read. Libby currently does not go the extra mile to suggest similar titles. It is not really a big problem but it is a ‘nice-to-have’ which could allow better user experience.

Based on the presentation's feedback, we were advised to look up on data to support the evidence that people really want to find titles that are similar to what they've read previously. Even though there are current sites utilising the personalised recommendations feature, it is always good to be backed by research so that our justification for implementing the solution is solid.

Thus, with this in mind, I went back to the root of why I asked this question in the first place. Is it a matter of better user experience? Why do we need preferences anyway? In the recent years, studies have shown that there is an emphasised need to move beyond usability and to understand and design for more engaging experiences (Hassenzahl & Tractinsky, 2006; Jacques, Preece, & Carey, 1995; Laurel, 1993). Having just the usability of a product that does not attract users’ attention and fails to engage users might prompt users to stop using the product. O’Brien (2008) posits that current successful technologies are “...not just usable; they engage users.” Thus, I think to engage users and motivate them to continue using the app, there has to be a level of personalisation to retain users. There are a number of ways to emulate engagement with several studies suggesting: media presentation, perceived user control, choice, challenge, feedback and variety (Jacques et al., 1995; Said, 2004; Webster & Ho, 1997). These engagement emulations affect the physical, cognitive and affective elements of user experience (O’Brien, 2008). With this, the use of personalised recommendations can be said to demonstrate the user’s ‘perceived user control’; the personalisation the product allows users to choose books based on previous reads, and perhaps ‘feedback’ as the product takes into account the users’ reads and then generate suggestions based on past history.

With the above in mind, I think having suggestions of similar titles could elevate the experience of existing users and improve user retention. However, this could be purely subjective and there are other ways to increase user engagement as stated. Using personalised recommendations may be one solution and are also commonly sighted in multiple products such as our NLB website and Amazon website. 

Moving on to the insights from this particular task:

Task 3: Confusing popularity rankings

For existing/new users:

(1) Open Libby App (2) Find a currently popular book on Libby

This was a very simple task yet we found that there were users who did not understand the labelling (the stars rating & no. in series) placed on the books. I think existing users who have used the app for a while may have an inkling, however, the explanation they gave was very vague and there were differences in responses. This could mean that the labelling is unclear and users are unsure of what it actually means. Several other labels within preferences are also very ambiguous and are most likely unused. Also, we could improve the screen’s real estate as currently, the title takes up about 1/3 that only one book is shown to the user (1st screen on the left). 

The insights we have come up with this particular task are:

‘How Might We’s

After compiling the insights from the three tasks, we came up with a few HMWs to try and tackle the problems listed for each one. The HMWs should not be direct as we did not want to zoom into the solutions straight away but it should give us room to think about other possible ways to solve them. 

Possible Solution Sketches

With our HMWs, we can use them as guidelines when sketching for possible solutions. The sketches should try to answer our HMWs, yet we should also bear in mind that this should not be our final version as we still need to test it out and gather user feedback if our lofi prototype is intuitive and solves the pain point of the problems above. 

Here we try to simplify the homepage and also arrange the headers such as ‘genres’, ‘audiobooks’, ‘movie books’, ‘recommends’ etc at the top to reduce scrolling and for better visibility for both new and existing users. 

Here we try to improve the user experience by adding ‘similar titles’ when the user taps into their previous reads. We have also talked about how the books are being arranged and changed the layout to a left alignment once a user taps into a title so that we keep the consistency of the app’s alignment. 

Lastly, we tackled to improve the preference filter inside a particular genre section. We found that the preferences Libby has already incorporated seem a lot like a tag function, however, the discoverability for a user to click into the tags and nest them together is very low. Thus, we decided to reduce the header’s size and placed a clear tag section where users can ‘add’ and ‘delete’ tags within a particular genre. The books sorting are also reduced in size so that users are able to see more than 2 books. The fourth book would be halved so that the user would know that it is scrollable.

What I have learnt