Designing the dashboard of a Customer Service Management Tool.
Internship @ Nintendo | Summer 2021
Designing the dashboard of a Customer Service Management Tool.
A new and improved interface for Nintendo’s customer service agents, designed to streamline communication and boost motivation. It is a web application, built using Oracle Service Cloud.
Responsibility:
I was the solo designer on this project, in charge of research, design, and product strategy. Although I took responsibility and drove the project, I collaborated with other designers, stakeholders, developers, and users throughout my process.
The Original Dashboard.
Navi is a software used by Nintendo customer service agents to receive, manage, and complete customer incidents. Currently, the Navi dashboard houses Bulletins, or communications, detailing to agents information on the most recent announcements, power outages, and more.
Research.
After gaining understanding of Navi and covering business requirements with stakeholders, I moved onto user research. The goal of my research was to understand how agents interact with Navi and the Navi dashboard, and discover problems they face with the software.
01
User Interviews
To better understand agents and their interactions and feelings with Navi, I recruited 5 people for user interviews.
02
User Survey
For broader data across more agent types, I sent out a survey on behaviors and attitudes towards Navi, receiving 9 responses.
The User Problems.
Based off my research, I was able to identify four key problems with agents’ experience using Navi:
01
Cluttered and difficult to navigate
With a text-heavy interface and a lack of information hierarchy, users often didn’t know which bulletins they needed to click on.
03
Little time to read the dashboard.
Overall, agents usually don’t use the dashboard because of how long it takes. They prefer Outlook or Teams, although those forms of communication are less standardized and reliable.
02
No indication of new bulletins
Besides the time stamp, users don’t have a way to see which bulletins they haven’t viewed yet. This makes the process of locating bulletins to read both inefficient and complicated.
04
Low motivation and accountability.
This issue is more general than Navi-specific, but agents often experience drops in motivation and accountability, especially during afternoons or peak season.
Design Principles.
01
Engage.
How can I create an engaging communication experience for agents to encourage use of the dashboard?
02
Streamline.
How can I streamline the reading process to make it more efficient?
03
Incentivize
How can I incentivize and motivate agents to increase productivity and accountability?
Lo-Fi Prototypes.
By conducting a competitive analysis of Outlook and completing ideation exercises, I was able to, through iteration, create lo-fi prototypes in order to map out features and possible layouts. My goal for these prototypes was to show them to the development team to understand my technical constraints regarding features.
Development Constraints.
From a demo with our development team, I was able to learn about what constraints and freedoms I had. I learned the following:
No Unread/Read states
In talking with the devs, I found out the Navi is unable to have persisting states between sessions due to limitations of OSC. Thus, I was unable to have read or unread states in my designs, instead moving toward time-based states.
Modern, New Design System
Using research findings, I was able to convince the development team to use a new, more modern design system, as the old dashboard’s visual design was noted by many agents to feel “antiquated” and contributed to clutter on the interface.
Hi-Fi Testing and Feedback.
Using the direction provided from the dev meeting, I created and iterated on a high fidelity prototype. I then created a testing plan with the prototype, where I conducted testing with five customer service agents of different teams and agent types.
The new designs allowed agents to locate bulletins much faster than the original interface.
They described the interface to be intuitive and the filters as extremely helpful.
Positives:
1
2
3
4
5
Critiques:
1
Add “All” bulletins inbox.
Agents wanted an "All" inbox to more efficiently view all the bulletins at once and locate new bulletins.
2
More visually engaging metrics.
"The metrics are a little text-heavy..."
3
Unsure about search bar’s scope.
It was not clear what the search bar would search through. All the inboxes? Filtered results?
4
“How do I remove a filter?”
Agents had an add/subtract mindset towards filters, which contradicted with how my designs worked.
5
No use case for time-based sorting.
Agents said they wouldn't ever want the bulletins in any other order than newest first.
Conflicting Needs.
Around the time I was conducting user testing, I held a stakeholder review where I presented my high-fidelity prototype to three different business leaders to discuss if it had met their business needs.
I discovered that the feedback from stakeholders conflicted with the feedback from users.
Agents (users)
Agents valued features that allow them to streamline their work processes. In this case, searching and filtering to quickly locate most relevant bulletins.
Business Leaders
Business leaders wanted to push away from searching/filtering and similar “shortcuts” in fear that it would cause agents to miss some bulletins.
While I could see where business leaders' fears were coming from, the results I got from user testing and feedback contradicted their concerns. As a result, I was very passionate about keeping searching and filtering, as I believed it helped agents identify better the bulletins most relevant to them.
Thus, I decided to make a case for my design decision, using user feedback and testing findings to show stakeholders why it would be beneficial to keep search and filtering. Rather than remove the functionality entirely, I focused on a design that was encourage agents to not rely heavily on filters.
A Redesigned Customer Service Tool Dashboard.
The Solution
Metrics.
Promoting transparency and increasing motivation for agents through selective exposure to data.
Collapsable Time Groups.
In place of time-based sorting, collapsable time groups help organize and de-clutter the inbox, make reading the date more efficient and legible, and are easy for agents to understand and use.
Filters.
Filters are now in a modal, where they can be added and subtract to emphasize an optional experience and better align with how agents view filtering. The search bar has been transformed into a "filter by search term" option, as stakeholders were especially opposed to the search bar. Having it as more of a filter than a search provides the functionality to search, but keeps agents from over-relying on it.
Team Lead View.
An additional view for team leads of agents to keep their agents accountable. Offers transparency into both how many bulletins the agents are reading, and what they see in terms of agent metrics.
The Impact.
Final usability testing results and feedback.
Redesign Usability Rating:
4.7/5
The original dashboard: 2.6 / 5.
Time to read all new Outage bulletins:
300% faster
8 seconds with the new dashboard vs 24 seconds with the original.
“This could make reading bulletins so much easier...it’s such a simple approach but I love the new things you can do.”
- Interviewed Agent
“If we build this, we don’t need to have [employees] send out an email every time there is a new bulletin posted.”
- Leadership (CS Department)
Reflection.
01
Communication > Assumptions.
The value of communication, and even over-communicating, to get clarity and direction is much better than assuming and needing to re-do designs or meetings later.
02
Working with Cross-Functional Partners.
In a large enterprise environment, learning how to work with individuals from different backgrounds, teams, and expertises and juggling complex and often times conflicting information.
03
Learning to defend Design Decisions.
In the case of conflicts, learning how to use user testing, research, and UX principles to defend and advocate for design decisions that I am passionate about.