The Problematic Nature of Gojek’s New UX (Case Study + Redesign)
It’s evolving— just backwards.
UPDATE: Thanks to this case study I got an internship at Gojek! 💚
DISCLAIMER: This case study is written, researched, and designed based on the July update — opinions and problems that are addressed in this article may change.
As a UX designer, I get bothered by an interface quite often. Minor inconveniences such as uneven margins, poor contrast, outdated fonts, and comic sans (had to do this for the memes) are just some of the things I encounter on a daily basis. However, every once in a while, I stumble across a product so unique (read: poorly designed) that I am simply left speechless.
This new Gojek update is one of those.
For readers out there who aren’t familiar with Gojek, it is an Indonesian on-demand multi-service platform and digital payment technology group based in Jakarta. It’s the first Indonesian unicorn company as well as the country’s first decacorn company.
You would think that a company of this calibre would afford to prioritize user experience above all else considering its large user base, but one quick glance at the latest update is enough to prove otherwise.
The latest refresh of the app has received tons of backlash regarding the over-complicated interface.
This case study case aims to break down the new UI update from a technical perspective and provide a better-designed alternative. Without further ado, let’s get straight into it.
Overview
Gojek’s new home screen is a head-scratcher. Put simply, the new update implements bolder and brighter colours and a new floating menu on the bottom. It all sounds good on paper, as the latter is easier for one-handed use and accessibility.
As it turns out, however, a good theory does not necessarily translate to a good result.
Moving the main features of the app to the floating menu on the bottom for the sake of accessibility was, put short, not a good idea. The general consensus is that the problem lies within the fact that by doing so, Gojek is straying further away from their very-own vision: providing services.
The userbase isn’t happy with the direction that the app is heading towards. I, too, am not particularly fond of it. The lingering question that remains now is: why?
Design Thinking
In order to answer that question, I first needed to have a process outline that can help me stay on my lane during the incredibly extensive phases.
I start by observing the problems through my personal experience — all of this is, after all, inspired by my frustration over using the new app. Afterwards, I move on to the Research phase where I spent the bulk of my energy on. This research phase consists of various sub-phases that will be evaluated and addressed both individually and as a group.
The rest of the process is then pretty self-explanatory. The tools that I use for my research, ideation, and prototyping are Notion, Whimsical, and Adobe XD respectively. For usability testing, I conducted the classic pass-the-phone method in which users will benchmark with the current live application.
Observation
As I mentioned before, this whole study happened due to the frustration that I experienced while using the app. In fact, it wasn’t only me — I had friends and family members that explicitly stated out loud that they hated the new update.
My very first observation of the app was: Why did they move their services to the bottom?
My second thought was: Why is the homescreen so crowded?
My friends also seem to share the same opinions as me, and one of them summed up this whole ordeal quite nicely:
“I’m so confused— it’s an eyesore!”
Objectively speaking, however, I acknowledge the notion that Gojek is trying to head into with this update. The new menu is more easily accessible with one hand, and the brighter colours are meant to modernize and refresh the old, dull Gojek interface. Moreover, the news and promotions are now more apparent to increase customer appeal and conversion rate.
These are all understandable from a business point of view, and it is not my place to decide whether it’s wise or not. One thing that’s clear, however, is that they are trying to prioritize accessibility above all else with this new update. @Gojekdesign even tweeted a release note on Twitter that confirms as much.
Credit where credit is due, the design team were really thoughtful to attempt prioritizing accessibility on this update. The keyword being the word “try”.
Unfortunately, attempting and creating are two very different things, and in this particular case, Gojek fell short at trying to deliver the goals they were aiming for. What was even more unfortunate, however, is the fact that in attempting to do so, it created a plethora of fatal flaws that weren’t there to begin with.
Case in point: the new update failed to understand that one-handed operability needs to be consistent across all screens. Take the GoFood page, for example. Users still need to reach all the way across the top for the search bar, and a lot of components are still not accessible with one-hand. Put simply, this update has a lot of half-baked features that are still not ready for public use.
By definition, half-baked means that it “has not been properly thought out, and is stupid or impractical”.
I will not bring myself as far as to say that this update is stupid, but from an observer point of view, it is borderline laughable that the team behind this update failed to foresee the counterintuitive inconsistencies throughout the whole product.
Research
The phase that everyone either really hates or really loves.
Online Reviews
Hopping on to Twitter, I noticed that @gojekdesign ‘s Twitter account is bombarded with user complaints. The major complaint here seems to revolve around the newly-implemented floating menu and the hidden order history page.
Hopping on to a more formal platform (the Play Store and App Store), there isn’t much of a difference. The trend is similar, with users complaining regarding the new update and/or wanting to revert back to the old User Interface.
User Research
To validate and understand why online reviews are as such, I needed to research the users of Gojek. Conducting this research is necessary as we may discover insights and patterns that cannot be explicitly seen.
Since the majority of Gojek’s users are from Indonesia (80%), my research sample will be solely Indonesian users. Based on the data found from Expanded Ramblings, there are around 30 million active monthly users in Indonesia (2019).
Research conducted by Puskakom UI in 2017 showed that the majority of Gojek users are 20 years old (56%), followed by 30 years old (28%), and lastly < 20 years old and > 40 years old (both 7%). From an educational perspective, 54% of users have a bachelor's degree, 18% high school degree, 12% diploma, 7% master’s degree, and the rest have an elementary degree or junior high school degree. These figures are achieved by using pure random sampling method that involved over 10,000 respondents.
Moreover, the study also finds that 83% of users use Gojek use the GoRide service, 69% for GoFood, 50% for GoCar, and 35% for GoSend.
It’s a lot of numbers and percentages, I know. Just hang in there, there’s just one more data and we’ll be over this complicated phase.
Another study done by Kompas in 2017 showed that there were 15 million active monthly users. A quick observation tells us that the number of users has doubled over the past 3 years (15 → 30 million).
With this in mind, we shall now narrow down our scope to only the ones at risk of getting confused with the new design. It’s either: a.) Middle-to-old aged users or b.) Users with a very low level of education.
For the sake of this research, we’ll be utilizing the former as age factor more accurately reflects technological adaptation (or lack there-of) than a level of education one has, especially in these modern times.
From the accumulation of the data above, we can then calculate a very rough approximation of the number of users that are over 40 years old in 2020. Taking into consideration the growth of Gojek from 2017 → 2020 as users have doubled, we can comfortably assume that there are around ≥ 2–3 million of users who are over 40 years of age.
That is a lot of users, in case you didn’t know.
User Interview
For the user interview stage, instead of having an online survey, I decided to interview my circle of subordinates face-to-face (with a mask, of course) to get a more thorough and detailed answer.
Out of the 20 respondents, 5 are aged below 20 and are currently taking an undergraduate degree, 8 are around 25 years old with a bachelor's degree, and 6 are 40 years of age, and 1 is a baby boomer who is over 60 years old.
The questions are as follow:
- What are your thoughts on the new UI?
- Which one do you like better — the old UI or the new UI? Provide a short explanation why.
- Do you read the promotion contents and/or Promo tabs?
- Is the new UI causing heavy problems (slow app load time, laggy animations, etc)? How does it fare compared to the old UI?
- Would you like for Gojek to revert back to the old UI?
- Do you often use Gojek with one hand?
The respondents’ answers were… unsurprising, to say the least.
Out of the 20 respondents, 50% say that the new UI looks good but is way too confusing. The other half does not like the interface as much, stating that it is too colourful and over-complicated.
For the second question, 18 out of the 20 say that they like the old UI better. The other 2 says that they prefer the new design as it is more, quote-and-unquote, modern” and playful. The explanation varies between each of the respondents, although there is an underlying pattern across the majority of the answers: in short, most of them felt that the new update had too much going on and as a result, they found it hard to focus on the information that matters (the services).
Respondents feel that they are exposed to too many information and components on the home screen.
The third question has the most objective answer. 100% of the respondents stated that they have never read any of the promotion contents. This may happen due to a couple of reasons, but my main suspicion is that users who open the application already has a very specific goal in mind that they want to achieve (e.g. I want order Food A through GoFood), resulting in the skip of the home screen altogether to get to their desired destination.
60% of the respondents mentioned that they didn’t notice any performance difference between the old and new UI refresh. However, the other 40% stated that they experienced a noticeable drop in performance after the new update. As this is an anomaly, I needed to follow up and investigate why it is as such. As it turns out, the answers varied greatly due to the phone tier.
60% of respondents who stated that they did not notice any performance difference have a midrange/flagship phone, with the lowest price of 4 million Rupiah and the highest of 12 million Rupiah. On the contrary, the 40% that suffered a performance drop had phones that are on the low-end of the spectrum with an average cost of 1.8 million Rupiah.
This finding is not at all surprising. More CPU power would significantly reduce the probability of performance issues as they are inversely proportional in this context. Cheaper phones tend to use less-capable processors, hence the performance drop. Regardless of the reason, a performance issue for an app of this calibre is just simply not acceptable. Hence, I will be further investigating this aspect in my comparison phase.
Moving on to the penultimate question, 16 out of the 20 respondents answered that they’d like the update to be reverted. 4 of the respondents, when asked why they do not want to go back to the old UI, state that the update will grow on them after the adjustment period. This introduces us to another aspect that we have not taken into consideration until now: Learning Curve. I will be discussing this down the line, but for now, let’s move on to the last question.
The last question was included due to the fact that the new update is repeatedly said to be done for the sake of one-handed operation. 70% of the respondents state that it depends on the situation and the other 30% state that they never operate their phone one-handedly. Following up on this, 100% of the respondents feel that one-handed operation is a very niche thing that they simply do not care enough about.
Comparison
New UI vs Old UI
In other words: Visual Appeal vs Good Usability
New UI
What Went Right:
- Revamped GoPay component design
- Streamlining of navbar into 3 section avoids redundancy
- The use of brighter colours is modern
What Went Wrong:
- Core features aren’t apparent at first glance since it is located at the bottom.
- It is more cluttered (too many promotions and news which users barely even read).
- Why is there a slide down feature that does absolutely nothing?
- App is heavier and low-end phones are suffering performance issues
- Order History is hidden deep in the menu
- Bad UX
Old UI
What Went Right:
- Clear visual hierarchy.
- Core features can be clearly seen at first glance.
- App is more lightweight
- Orders are more easily accessible
What Went Wrong:
- Outdated design components and colours
- Unclear difference between Inbox and Chat feature on the navbar
Competitor Comparison (Grab)
Grab is a Singapore-based technology company offering ride-hailing transport services, food delivery and payment solutions. In other words, it’s Gojek’s biggest rival.
Grab’s UI
- Hierarchy is similar to Gojek’s old UI (Clear and concise)
- Good use of whitespace — not too much colour and content promotion is easier on the eyes. This allows users to focus on what really matters: using their services.
- Super easy access to view past orders.
Sometimes less is more, and more is less; Grab managed to land a perfect spot right between the interchanging dimension of good UX and UI.
Performance Comparison
For this phase, I will be doing a series of test that aims to compare the performance between the new Gojek app with the old one. To make things fair, I did the tests on a series of different phone tiers and did each test three times. The time displayed below is the averaged time each app needs to take before displaying the homescreen.
Testing Method: Clear Cache of Gojek App → Clear All Apps From Memory → Time how long it takes for Gojek app to boot up on a clean state (1st Run)→ Erase from memory → Reopen (2nd Run) → Repeat 3x → Calculate Average
Samsung Galaxy S10+ (Flagship) (2019)
Specs: Octa-Core Snapdragon 855, 8GB RAM, 238GB Free StorageNew UI:
Result (1st Run): 2.87s
Result (2nd Run): 2.42sOld UI (v. 3.53.1):
Result (1st Run): 2.74s
Result (2nd Run): 2.51s
Samsung Galaxy J3 (Low-End) (2016)
Specs: Quad-Core Exynos 7570, 2GB RAM, 1.2GB Free StorageNew UI*:
Result (1st Run): 9.24s
Result (2nd Run): 7.89sOld UI (v. 3.53.1):
Result (1st Run): 6.29s
Result (2nd Run): 5.43s*Slight animation lag (swipe gestures)
Samsung Galaxy Note 3 (Old Flagship) (2013)
Specs: Quad-core 2.3 GHz Krait 400, 3GB RAM, 2.7GB Free StorageNew UI*:
Result (1st Run): 28.31s
Result (2nd Run): 24.67sOld UI:
Result (1st Run): 14.26s
Result (2nd Run): 16.66s*Very laggy animations (freezes and lags)
The performance difference is negligible on flagship phones. However, on old and low-end smartphones, there is a clear disparate in performance (app load time and animation smoothness).
On low-end phones, the animations need a couple of seconds to register in the first place. On old flagships, on the other hand, the animations were just straight-up a pile of unadulterated trash of lags and freezes.
This result then begs another question, is the worse performance really worth the aesthetic gain? If I am to take a guess, the longer app load time in low-end devices is due to more graphics, components, and swipe interactions needing to be rendered in the new UI as it loads up, compared to the simpler and lighter old UI.
With this said and done, we are finally over the Research phase. Phew.
Adjustment Period (Learning Curve)
As promised, I will take into consideration this factor during my case study.
As is any other UI overhaul, I acknowledge that there is bound to be a certain learning curve in the beginning. And yes, there is also bound to be some complaints coming from the users — it’s human nature to despise learning, after all.
With this particular update, Gojek is trying to take a gamble by introducing a very steep learning curve in hopes that its users will grow to love the new update after constant use.
However, it is not much about the learning curve as it is about the fundamentals of a proper redesign in this case. A well-thought-out redesign would not generate this much negative reviews and out-lashes in the first place. Take the MasterCard brand redesign, for example.
Forcing an update to users with the mindset that they will grow to love it is as bad — or even worse — than not updating it in the first place.
Users will adapt, definitely, but perhaps the more important question that we should be asking is this: will they enjoy it?
Or will they be forced to enjoy the app due to the lack of alternatives?
There is a very fine line between enjoyment purely coming from experience, and enjoyment coming from the lack of options.
Problems
Based on the research that I had just conducted, I am then able to pinpoint the major problems of this new update. Instead of elaborating on each and every pain points in long paragraphs, for the sake of efficiency and practicality of this already long case study, I will instead narrow it down to a list:
- New location of core features is wrong. It is less-apparent at first glance since it is located at the bottom.
- Service buttons are more accessible, yes, but it requires another extra swipe to get to the other services. The old UI has 7 front-page presets, compared to the current 4. It’s counterintuitive.
- The home screen is over-cluttered with promotions and news (which users rarely read, if ever).
- The slide down implementation on the home screen is unnecessary and causes performance issues
- The app is heavier and low-end phones are reportedly crashing often.
- Why is Order History hidden so deep in the menu?
To show just how fatally wrong moving the core features to the thumb area is, I designed a similar case. Imagine Google Maps, but when you first open it it shows you the recommended places instead of the map itself. Instead, it tucks the map (which is the core feature of the app) to the thumb menu.
That wouldn’t make any sense, would it?
This design principle is similar to what Gojek is currently implementing in their app. Similarly, the core features (service buttons) are put in the expandable thumb menu.
While this implementation is easier for one-handed use, it completely neglects an albeit more important general UX rule: focal points. Focal points are areas of interest, emphasis or difference within a composition that capture and hold the viewer’s attention.
A good focal point should prioritize the core features first and foremost— it should make said features stand out as the element with the most dominance. This update, however, completely misses this crucial point entirely.
Either it’s purposefully done to generate more income through banner promotions or not, one thing is for sure: the users are not having any of it.
Design (Ideation & Prototype)
One thing to note is that I am redesigning this as a tweak instead of an overhaul. Therefore, I try to stick with Gojek’s brand identity — not mine. This means that I am closely following Gojek’s Asphalt design system and am using similar colours and/or fonts that are already being used in the app. Moreover, since the nature of this case study focuses heavily on the home screen above all else, I do not intend to design anything other than such (e.g. the GoFood screen).
Lo-Fi Design
In a nutshell, I went with the notion of blending the good things about the new and old app design. The hierarchy is closely related to the old UI, whereas the colours and identity resemble more of the new one. Here are my design notes:
- Core features are now back at the top of the screen, where our natural eyes rest. This hierarchy brings our focal point to where it really matters; the services.
- The 3 top app bars are now Orders, Home, and Chat instead of Promos (this is done based on research findings that state that users rarely read Promotion contents).
- Orders now have a dedicated page and can be accessed more easily from the home screen (one left swipe). It will also be universal; this means that all the different service orders will be under one page for maximum convenience.
- Promos are now moved on the home screen instead of having a dedicated page.
- Since Gojek is trying to prioritize one-handed use with this update, I included a floating universal search bar at the bottom, so that users don’t need extra taps/swipes to search, like the current implementation.
- Search is state-specific. This means that content will be prioritized on the screen the users are currently on (e.g. When a user is in Order History, search results will display your orders at the very top).
- NEW: Filter feature. This will allow users to easily filter and sort their orders.
- Pull-down gesture implementation is still present, albeit a little more subtle this time. It’s used only to go back from the Search screen and the Filter card.
Design System
Following Gojek’s Asphalt design system, I created a mini design system for the redesign. The colours include but are not limited to the ones displayed above— although, I will be mainly using those 5 for the primary components.
Hi-Fi Design
As I mentioned before, this Hi-Fi redesign retains most of Gojek’s branding and identity. Components such as the bottom floating menu along with many colours and/or CTA buttons are inspired directly from the current live app. Here are my high-fidelity design notes:
- Use of similar colours with the addition of subtle gradients retains the current playful direction that Gojek is taking.
- Streamlined look— with the addition of more white space and fewer complex illustrations.
- History is now more easily accessible and is texts are colour-coded to specific services for easy readability and/or differentiation.
- Subtle swipe interaction implementation makes the overall app feel more dynamic.
- If the search text box is pressed, the state will immediately change to input mode so that users do not need to reach to the top to press the search bar again.
One screen alternative that I did not include in the final product is the Center-aligned Search page. After back-and-forth consideration, I find it hard to justify using a different alignment solely for this page as it breaks the overall cohesiveness of the app.
Validation
Usability Testing
The basic methodology of this testing is to let users benchmark the prototypes with the live app on 2 different devices simultaneously, and then recording their feedbacks. To decrease the probability of variating factors, the devices used for this testing are similar in nature and specs (iPhone XS and iPhone 11).
These are the positive feedbacks:
- The prototypes are much easier to navigate than the current live app. (9/10 users)
- The dedicated “Orders” tab is much more preferred than the current “Promos” tab. (10/10 users)
- The alteration of the Promos tab into the home screen is much more accessible. (8/10 users)
- The history page is informative and well-designed. (10/10 users)
- The search screen is more practical and clean. (10/10 users)
- The filter feature is a must-have. (10/10 users)
- The subtle gradient is beautiful to look at (6/10 users)
These are the negative feedbacks:
- The colours are still too colourful. (4/10 users)
- The icons are a bit outdated. (3/10 users)
- The gradient is too distracting — flat colours are much more preferred (4/10 users)
TL;DR : 9 people out of 10 prefer the redesigned prototype over the current live app.
Conclusion
As Jared Spool once said:
“Good design, when it’s done well, becomes invisible. It’s only when it’s done poorly that we notice it. Think of it like a room’s air conditioning. We only notice it when it’s too hot, too cold, making too much noise, or the unit is dripping on us. Yet, if the air conditioning is perfect, nobody will say anything and we focus, instead, on the task at hand.”
Good design is not something you look at and say, “wow, that’s a great design!”— instead, it is the one that goes unnoticed.
Gojek, unfortunately, is on the other end of the spectrum.
It’s apparent that they are trying to establish their very own unique identity in this ever-populated service market by creating a stand-out design. However, in doing so, the decacorn company has mistakably over-prioritized its visuals over its usability.
This new update feels like Gojek is stuck in a limbo of in-between; they have put themselves in a position where neither the usability and visual design are great.
What’s more questionable is that Gojek is being eerily quiet regarding the whole ordeal. Considering the fact that so many of its users are against the update, maybe it’s time for Gojek to consider the possibility of rolling back the new update and move on. However, looking at the direction in which the company is going as a whole, this plausibility seems very unlikely.
This then leads us back to square one:
Why is Gojek so hell-bent on pushing this update?
Based on the limited amount of study we had just conducted, one can only make calculated assumptions. The first possibility is that the company is trying to shift gears into generating more revenue through this update(this is reflected by the prioritization of promotional contents over the core feature). The second possibility is that maybe the company wanted to stand out from the rest of the market with its colourful (too colourful, I might say) brand & identity.
The third and last possibility is that the company simply did a mistake and hired too many UI Designers and not enough UX Researchers.
The true answer, however, we will never know.
Thanks for reading! Oh, and if the Gojek team is reading up (or should I say, down) until this point, give me an internship maybe? Haha jk… unless?