Scroll ↓

Summary

Increased speed and efficiency of user interactions with the Ally mobile app by designing a shortcut widget to be incorporated on the homepage.

Reduced number of clicks to reach users’ most common tasks by 50%.

Role

Product Designer

Duration

4 weeks

Team

Just me!

Tools

Figma, ChatGPT

Final solution

Final solution

Watch this quick demo of the final widget!

And an example scenario: Accessing Zelle

Before

After

Background

Background

In a world where mobile banking is the #1 banking method among consumers, it’s critical for Ally to ensure the usability of its own app.

However, given the complex nature of banking and finances, keeping user interfaces minimal and easy to navigate is no simple matter.

After encountering problems of my own while using the Ally app, I sought to discover how it could be improved.

My 4-week project plan, following the well-known Double Diamond model.

Research

Research

I used three different research methods to gain a holistic understanding of the app's current usability.

In addition to each method’s unique strength, these were chosen for their speed and cost-effectiveness, due to time and budget constraints.

Heuristic evaluation

Get an idea of potential problem areas to focus on

Competitive analysis

Learn from competitors’ strengths or discover opportunities to shine

User interviews

Great for exploratory research and provides more context than surveys

Based on the heuristic evaluation and competitive analysis, I initially concluded that Ally could provide a unique value to users by incorporating shortcuts in the app.

To validate my hypothesis, I conducted 5 user interviews—but I had to get creative.

I didn’t have access to Ally users… so I made my own! I took screenshots of the Ally app (redacting my information where necessary), and quickly pieced them together in Figma to create an interactive prototype.

During interviews, I first gave participants a few minutes to familiarize themselves with the prototype, so that they would be able to answer questions about the app’s usability.

This workaround solution may have even worked in my favor, because actual users may be familiar with the app enough to navigate it well, regardless of any usability issues.

I swear this didn’t take as long as it looks 😅

Overall, participants’ responses strongly supported a need for quick access to common tasks.

Furthermore, I extracted 3 important findings regarding users’ behaviors to inform later designs:

Each user wants the interface to be tailored to their unique needs.

“[This message] does not apply to me and I would rather see my banking info.”

Any task requiring more than 1 step risks causing frustration, or even failure to successfully complete the task.

“I gave up after trying twice [when a feature didn’t work].”

Users pay the most attention to the homepage, and especially to information above the fold.

“I’m used to seeing amounts [i.e. account balances] at the top as soon as I log in.”

*Although I prompted ChatGPT to conduct a thematic analysis as well, it didn’t catch anything I might have missed—in fact, I found its analysis to be far less nuanced than my own.

After prompting ChatGPT to create a guiding user persona and journey map, I defined the problem to be solved using a “How Might We” statement:

“How might we help users quickly access their most-used features, so that they feel satisfied and confident using Ally?”

Ideation

Ideation

At the end of brainstorming, I found myself torn between 2 different ideas.

Option A: Ring

Appears around the user’s finger when they tap and hold on the screen, allowing users to access features with a flick of the finger.

Option B: Widget

Located on the homepage and lets users create and add their own shortcuts, offering a new level of personalization.

After great consideration, I decided to move forward with option B.

Although ChatGPT’s ring idea was novel and exciting, its implementation could potentially interfere with users’ normal gestures.

Additionally, users would likely need a tutorial on how to use the ring, thus adding to their cognitive load and defeating the purpose of a shortcut.

Design

Design

After completing lo-fi designs, I discovered that my solution… did not work.

Good thing I asked others for feedback early on! I found that while the concept was good, the execution needed improvement.

Originally, I envisioned that users could choose which pages to star, and then shortcuts to those pages would appear in the widget—similar to the concept of bookmarking.

However, multiple respondents said that the starring system was unintuitive, easy to miss, and could potentially cause decision paralysis.

Lo-fi prototype of my initial idea, which I used to gather feedback.

Since I hadn't invested too much time in the previous design, I was able to rapidly iterate upon it.

Instead of placing the burden on the user to create shortcuts, the widget would now automatically display shortcuts to the user’s most-visited pages: no effort required from the user!

I polished my designs to hi-fi, and was faced with one last decision of choosing between several widget layouts.

Once again, I sought feedback from others.

At last, the victorious layout #3 was selected, for its compactness and horizontal structure, which was consistent with other parts of the app.

Results

Results

After examining the Ally app’s original structure, I determined that my solution would reduce clicks to get to users’ most common tasks by 50%!

The widget would save even more time for uncommonly-used features buried in the app by making them instantly accessible from the homepage, thus enabling users to complete their banking tasks with ease.

“I could absolutely see myself using this in my banking app to save time if there was a feature like this.”

Reflection

Reflection

Finally, my takeaways from this journey:

Ask for feedback even if you think your idea is great.

I was ultimately glad that I didn’t skip this step, because other people surfaced problems with the design that I hadn’t thought of.

AI tools excel at doing things faster, but produce subpar deliverables due to their lack of nuance and reasoning.

Therefore, I choose to keep parts of the design process fully human, but if I use AI, I always review and edit its products before using them.

Research has funny ways of surprising you, so don’t skip it!

Some of the insights that influenced this project’s direction for the better were completely unexpected.

Lastly, if I were to improve this project, I would love to bring on teammates. I have no doubt that a mix of different perspectives would have culminated in an even better solution. Here’s to hoping for plenty of opportunities to create and collaborate with others in the future!

Create a free website with Framer, the website builder loved by startups, designers and agencies.