The Problem
We get a high volume of client support tickets while our existing Help Center does not get a lot traffic. The challenge is to redesign the Help Center so our clients can use it to resolve their support issues before they open a ticket. Additionally, we want to incorporate the Roostify brand into the Help Center design to build trust with clients.
My Role
Product Design Lead
Time Frame
3 Months
Collaborators
Product Manager (VP of Product), Customer Success Team, Zendesk Implementation Partner
Project Summary
I began by interviewing users and analyzing previous year support data and Zendesk usage metrics. As I gathered more insights, I refined the direction of the project. Turning insights into opportunities, I explored different design solutions, and hosted a cross-functional workshop with client support, implementation and product teams to define the information architecture. I tested the low fidelity design in front of users to get their feedback. With input from the marketing and design team, I completed the the final UI design. We partnered with a Zendesk implementation partner to build the site and later onboarded a technical writer to rewrite the articles. The site launched in August 2021. We saw a 10% increase in article views 30 days after the launch.
Understanding the 2 Dimensions of the Problem
The User Problem vs. the Business Problem
The User Problem
Let’s take a moment to distinguish the user problem from the business problem. The target user in this case are our clients’ employees, the bank administrators. I’ll refer to them collectively as “clients” for simplicity. The problem is that our clients do not use the Help Center before they open support tickets. We want them to confidently use the Help Center to resolve their questions instead of opening a support ticket right away.
The Business Problem
With a constant, high volume of support tickets coming in, our client support team is overwhelmed. There’s a high turnover rate and we need to continuously hire more supports agents. This is very costly to the business operations. In order to solve the business problem, we need to get clients to self-service help using the Help Center.
Research: Some Questions to Uncover
I broke the problem into smaller pieces to frame the direction and scope of the research.
Who are the different people that access the Help Center?
How does the current support process work and where does the Help Center fit in?
How much impact can an improved Help Center provide?
Why are target users not using the Help Center?
How might we increase the Help Center’s usage?
I began by interviewing our customer support team and clients (clients will refer to bank employees). While interviewing our internal team, I asked each person to refer me to someone else who might offer helpful information. I ended up speaking with members across multiple teams, including implementation, engineering, sales and marketing.
1. Who Are the Different People that Access the Help Center: Understanding the Target Audience
2. How Does the Current Support Process Work?
Currently, clients can open a support ticket in two ways, using the “submit a request” function in the Help Center or sending an email. Most choose the latter option. When a client support specialist receives the ticket, they go through the process below to resolve the issue.
3. How Much Impact Can an Improved Help Center Provide?
To understand the potential impact, I looked through the CS team’s excel database of support ticket data as well as metrics on Zendesk. I synthesized the data and found the most insightful pieces to be the Support Ticket Subcomponent, which is what topic each ticket is about, and corresponding topic searches in Zendesk. With the help of a support agent, I determined what percentage of topics a Help Center article can really solve.
Redefining the Scope
Since a redesigned Help Center can help between 17% - 52% of support tickets, our success metrics should focus on these areas rather than on the total ticket count. I aligned with the VP of Product to redefine the this project’s focus and the metrics for success.
4. Why are target users not using the Help Center?
In addition to interviewing users, I conducted an audit of our existing Help Center to put their feedback into context.
Audit of the Existing Help Center
5. How Might We… Increase the Help Center’s Usage?
Let’s take a step back and recap the key research insights.
Actual impact: An improved Help Center can help between 17% - 52% of support tickets in select topics. We need to inform our technical writer to prioritize these topics because they have the most potential for impact.
Currently, clients don’t use the Help Center because it’s too easy to open a support ticket. We need to partner with CS to promote the new Help Center and train clients to use it.
The design strategy should focus on making it easy for clients to find information, in particular, motivating them to stay engaged to finish an article and easily find related content if the first article did not provide sufficient information.
Design Opportunities and Strategy
The research insights informed me to follow the following design guidelines.
Motivating concept: create many “deflectors” in the Help Center so it becomes really easy for the user to find what they’re looking for.
Organize the content in a way that is easy and clear for users to find the information they are looking for, i.e. prominent search bar, separate general admin and loan team personas.
Use visual design to delight, engage, and motivate users to stay on the page until they have found the information they’re looking for.
Avoid overly-technical terms. Use human language that is easy to understand. (We brought on a UX writer who can whisper magic into words.)
Exploring Different Design Solutions
Competitive Analysis
According to Jakob’s Law, users spend more time on other websites and apps than on our own. It’s important to understand what they are expecting when they access a Help Center. It’s also useful to see how other products solve similar problems since we are not trying to reinvent the wheel. Below is a tiny sample of the products I studied for competitive research.
Cross-Functional Collaboration to Explore and Define the Information Architecture
I hosted a cross-functional online whiteboarding session with the client support, product, and implementation teams to define the information architecture for the Help Center.
Ideation, Wireframing, and Visual Design
Starting with pencil on paper, I began rounds of ideation and developed a series of low-fidelity wireframes. Incorporating feedback and new insights, I evolved the wireframes to higher fidelity designs. Working within brand guidelines, I created a palette that combined components from both our brand guide and UI Kit. I also illustrated an icon set.
Final UI Design
Results
Measuring Success
Since the site launched in August 2021. We saw a 10% increase in article views 30 days after the launch. Unfortunately, I wrapped my time up at Roostify 2 months after the launch. If I spent more time there, I would collect the following metrics to determine success. I would compare the number of support ticket subcomponents with the corresponding article views. For example, an increase in “User Education” article views coupled with a decrease in “User Education” support tickets in the same time period would indicate positive Help Center impact.
Reflection
Sometimes, design is just a part of the solution. Building trust and alignment with cross-functional teams is crucial to a project’s success. Because I built relationships with members on the CS team, they helped me uncover so much key information and insights that I might’ve missed otherwise. They also volunteered to host webinar trainings to onboard clients on using the new Help Center.
Finally, having connected with everyone who had touch-points with the support process, I was able to identify the key people who needed to be in the cross-functional whiteboarding session that I hosted. The project helped me truly understand the power of collaboration and I’m very grateful for it.