As mmhmm’s Knowledge Manager, I took on the exciting challenge of redesigning our Help Center. This project was all about making our users’ lives easier by transforming a cluttered, outdated resource into a sleek, user-friendly hub that truly supported our growing suite of products.

Transforming mmhmm’s Help Center

UX Research
Engineering
Customer Support
Video Production
Localization
Design

👷🏽

🤝

Collaborated with

Role

Lead designer and
project manager

3 years

⏱️

Timeline

About mmhmm

mmhmm is a SaaS (software as a service) company that’s aimed to help people create engaging presentations and videos, making remote work more effective and enjoyable. mmhmm’s Help Center is designed to deflect support tickets and allow users to self-serve by finding answers to their questions.

Before the redesign, our Help Center wasn’t living up to its potential. Here’s what we were dealing with:

The starting point

  • Finding the right information was like searching for a needle in a haystack. The site architecture was cluttered, making it difficult for users to find relevant answers to their questions.

  • With mmhmm evolving fast, many of the help center articles were old and confusing.

  • Users were reaching out for help on issues they could have solved themselves if the Help Center was easier to use.

  • Without standardized guidelines, Help Center documentation lacked unified writing style and visuals.

Goals and objectives

I set out with clear goals to tackle these issues:

Boost user experience

Improve navigation and accessibility to help users find information quickly.

Keep our help articles fresh and aligned with our latest products.

Improve content quality

Encourage self-service to reduce the number of monthly support tickets.

Reduce support cost

Create consistent guidelines for writing and visuals to maintain quality across the board.

Standardize content creation

The redesign journey

Diving into Research

  • Gather user feedback: I partnered with our UX researcher to collect insights through surveys, feedback forms, and support tickets to understand what users were struggling with.

  • Audit existing content: I went through over 100 help articles to check what was still relevant and what needed updating.

  • Analyze data: Using Google analytics and Zendesk explore tools, I tracked how users were interacting with the Help Center to identify the most popular topics and pain points.

Teaming Up Across Departments

  • Partner with UX Researchers: I collaborated with our UX researcher to map out how users should navigate our Help Center.

  • Coordinate with designers and developers: I worked closely with designers and developers to implement a new, streamlined site theme.

  • Keep stakeholders in the loop: I sent out regular updates to product managers and executives ensured everyone was on the same page and their feedback was incorporated. Below is a snippet from a company-wide update video I created.

Rolling Out the Content Strategy

  • Redesigning information architecture: Reorganized the Help Center layout to make it more intuitive and user-friendly.

  • Setting quality standards: Developed style guides and templates for the written copy and visual assets to ensure all content was consistent in tone and appearance.

  • Content creation and updates: Continuously updated and curated articles to reflect our latest product features and user needs. View writing samples.

Embracing Technology

I implemented the following technology into our Help Center:

  • AI Algolia search: Improved our search functionality, making it easier for users to find what they need. Some features that were implemented:

    • Rules - allowed us to pin or hide items in search results, or temporarily promote certain articles to the top of the search results page (great for launches when we want to highlight certain new features).

    • Dynamic Reranking - helped rerank search results using user search behavior trend. It helps identify and demote results that are not relevant to search query.

    • Dynamic synonyms - identified queries that our users search and proposes synonyms for them and allows us to apply those synonyms as rules.

  • AI Chatbot: I integrated a chatbot to answer common queries instantly, providing quick support to users without the wait.

  • UX testing and user feedback: I worked with our UX researcher to conduct public-facing testing sessions to ensure design changes met user needs and expectations. In addition, working alongside another product designer, we created a custom feedback form for our site. Whenever someone filled it out, I received an email notification, allowing me to quickly update articles based on their suggestions. This hands-on approach helped me keep our Help Center in tune with what our users really needed.

Facing challenges

  • We transitioned from a five-person writing team to a single contributor (me). To keep up the output of high quality documentation, I prioritized tasks effectively and automated routine processes using Asana, and set an internal SLA (service level agreement) to set expectations with stakeholders as to when documentation would be updated.

  • I helped the team adapt to new processes and technologies. I outlined new procedures for giving feedback for Help Center articles and showcased how it helped me prioritize feedback so I can incorporate it into article updates.

Winning outcome

Thanks to our improved site navigation, fewer users needed to reach out for help, which cut down our support costs. This is represented by calculating the Help Center’s self service score.

The self service score gauges how effectively the Help Center allows users to resolve their issues on their own. To calculate it, take the number of times the Help Center is viewed each month and divide that by the number of support tickets received in the same period:

Self service score = (monthly site views) ÷ (monthly support tickets)

A reduction of support tickets leads to lower operational costs. According to Zendesk’s benchmark, the self service score for the IT industry is 4.6, and the software industry averages around 7.1.

mmhmm’s Help Center score has an average of 64.4 (with an all-time-high score of 103.9).
This is 15.7 times better than the IT industry average, and 9 times better than the software industry average.

Simply put, for every 64 Help Center views, we would only get 1 support ticket. You can see this impressive efficiency in the charts below, where our self-service score keeps climbing while support tickets are steadily dropping over time.

Key takeaways

Putting users first leads to more effective customer support.

User-centric design matters

Leveraging AI tools drastically improves both user experience and operational efficiency.

Technology is a game changer

Collaboration is key

Working with different teams brings in diverse perspectives and expertise.

Continuously gathering feedback and analyzing data helps maintain and enhance the Help Center’s effectiveness.

Keep improving

Wrapping up

Redesigning the Help Center was a major milestone for mmhmm. It not only boosted user satisfaction and engagement but also highlighted the importance of strategic content management and teamwork.

Throughout this project, I honed my cross-collaboration and project management skills, and deepened my understanding of UX design and technical writing. I also improved my data analytics abilities and sharpened my technical skills with Markdown, CSS, and HTML. This experience reinforced the need to stay adaptable and innovative in knowledge management, making sure we can always meet and exceed our users’ needs.