Designing a categorization tool to boost feature adoption for a B2B startup.
I was a Product Design Intern at Pendo, a rapidly-growing startup offering a suite of tools that help businesses understand what drives their success. One of Pendo's key products is Guides, which businesses can deploy to fulfill a number of purposes - onboarding new users, announcing new features, and communicating important updates to their product. During my internship...
I explored ways to reduce customer frustrations.
I focused on increasing customer awareness of promising new features.
I immersed myself in every part of the design process.
I participated in research sessions, collaborated with designers and developers, and produced deliverables in Figma and Whimsical.
I boosted conversions for a new feature by 125%.
I presented my work to the company, and saw my ideas making a positive impact in the product.
Background
After finishing my first project, I dove deeper into the need for efficient guide management demonstrated by Pendo’s users. Research uncovered strong concerns from customers about overwhelming their users with guides, leading many to preemptively limit their active guides.
Recognizing this, my team proposed a solution: guide categories, each with predefined rules and reusable layouts. This showed a lot of promise, allowing customers to save and reuse visual and stylistic elements and organize their guides. However, some issues had emerged:
Layouts could only be identified by name.
There was no other way for users to differentiate between saved layouts, creating confusion and inefficiency.
There was no way to sort or filter between layouts.
Without these capabilities, customers could have unwieldy lists of layouts, making it difficult to locate the right template quickly.
Customers cannot edit a saved layout.
This forced customers into a cumbersome duplicate-and-delete process, opening the doors to version control issues and a cluttered dashboard of outdated layouts.
These limitations were creating a paradox: a feature designed to accelerate guide creation was actually causing a bottleneck. The lack of organization was bound to lead to inconsistent user experiences, the very problem guide categorization was meant to solve.
Recognizing the importance of categorization for guides, my team prioritized this as a critical enhancement to unlock the full potential of Pendo's guide management system.
Design Process
When I started to ideate, I knew that it was important to leverage existing UI patterns to streamline design and development. I decided that a dropdown, which was already established throughout Pendo’s product ecosystem, would be perfect for selecting categories for a layout.
I also decided, through discussions with other designers on the team, to focus exclusively on category assignment for new layouts, deferring the more complex editing of existing layouts to future iterations of the tool.
I began with low-fidelity sketching sessions, exploring various complexity levels for category assignment. Through iteration, I landed on a solution featuring:
A dedicated ‘Category’ field with helpful explanatory text
A multi-select dropdown enabling efficient assignment of one or more categories in a single interaction
Using Whimsical, I translated the sketched concepts into detailed wireframes that expanded on the core interaction design, preparing the foundation for more detailed prototyping.
Prototyping
As I started prototyping the design in Figma, I explored the different dropdown variants in the Pendo design system to identify the most suitable solution for our specific use case.
First Option
Initially, I opted for a checkbox-based dropdown used in different parts of Engage, Pendo's main product.
This dropdown organizes options with headers for clarity and categorization. When a user picks a category, it moves to a "Selected" section, allowing users to easily view and remove selected categories.
However, when a category moves to the "Selected" section, it loses the context for various use cases provided by the headings in the list.
Second Option
I then tested out a dropdown with pill-shaped tags, also found throughout Engage. Unlike the first dropdown, selected options stay highlighted in their original position on the list of categories.
When gathering feedback from my team, we realized that the absence of checkboxes might make users unaware that they can choose multiple categories. Before a user selects anything, this dropdown appears identical to the single-select dropdown in Pendo’s design system.
To address this, I chose to combine the styles of both dropdowns into a custom component.
Solution
The final design directly tackles the inability for customers to organize and quickly identify their layouts.
When a customer creates a new layout, they're immediately presented with an area to add categories. Clicking into the dropdown, users encounter an interface that combines the best of Pendo's existing design patterns. The hybrid design features:
Clear Multi-Select Affordance: Checkboxes immediately communicate that multiple categories can be selected, eliminating confusion.
Contextual Organization: Categories remain in their original positions with visible headers, preserving the organizational logic behind the categories.
Visual Confirmation: Selected categories are highlighted in the list and simultaneously appear as tags in the input field, providing immediate feedback.
For power users managing layouts that apply to many categories, the design gracefully handles multiple selections. When space in the input field becomes constrained, a numerical truncation system displays the first few categories followed by the total count (e.g., "Marketing, Onboarding +3").
With this approach, users can maintain consistent organizational standards with their guides.
Impact
As my internship came to a close, I handed off my design to the engineering team for development. I advocated for my design decisions and emphasized the feature’s value to customers, while collaborating closely with engineers to ensure accurate implementation of all design elements.
By solving core usability bottlenecks in layout management, the feature should drive higher adoption rates among customers who previously stopped using layouts. And for enterprise customers, the new system will reduce the likelihood of duplicate layouts, translating to measurable time savings and improved consistency with guides.
While user feedback for the new feature is pending, I’m confident that it will significantly improve how Pendo’s customers organize layouts and create guides.
Lessons Learned
My time at Pendo was such an enlightening experience, especially as a young designer. Here are some of my takeaways:
Embrace constraints.
One of my biggest takeaways was that embracing constraints early, rather than fighting them, can lead to more innovative solutions. Navigating a short timeline and technical limitations led me to explore a variety of approaches to the problem at hand.
Feedback is essential.
Pendo’s weekly design huddles proved essential to my success. Having other designers and PMs critique my designs and challenge my assumptions taught me the value of diverse perspectives in any project. Even when the feedback was brutally honest, opening myself up to it ultimately pushed me to explore my successful solutions.
Stay organized.
By keeping my Figma files organized and tracking meetings and iterations in my Notion workspace, I was able to speed up communication, deliver an airtight presentation of my work, and preserve my sanity.
