Seismic Collections
Improving industry-leading software solutions for marketers and sales enablers by simplifying and optimizing industry-standard tasks.
ROLE
UX Designer
UX Researcher
TEAM
Matt Lackey - Product Manager
Ricky Nguyen - Product Manager
Peter Pchelin - Lead Developer
DATES
May 2021 - Aug 2021
Hypothesis
Our platform has historically catered to Enterprise customers with large teams dedicated to content curation and visual design. Many of our mid-sized customers struggle to curate similar experiences due to a lack of resources, we believe we can improve our existing model for both basic and advanced content curation experiences.
Our project goal is to streamline content curation and customization for users, making the experience both accessible and powerful for all levels of experience. We can leverage existing folder structures, reducing customer's need to replicate their content management within our ecosystem.
These improvements will not only save time but also allow users to focus on creating visually impactful content. As we build a new platform, we have the opportunity to address known friction points—like loading times—by overcoming previous technical limitations.
Process
Collections were initially envisioned as a standalone product with a focus on improbing publishing, curating, and controlling content in a unified way. I started by highlighting Jobs to Be Done (JTBD) and their respective stories.
With the help of my team, we identified the three use cases that I would need to solve for ; I want a streamlined way to activate content in Seismic, I want to quickly bucket content into groups visually and permission it to audiences, and I want to be able to edit content and a curated view in one location.
My first task was to first improve our content activation flow with a visual focus.
Jobs to be done (created with Lisa Piper)
Proposed solutions for dev review
Given the project's focus on content activation, I explored three implementation options. The Product Manager and I chose the "visual-first" model (Option 2), which lets users use their existing folder taxonomy with minimal curation. Developers confirmed it was feasible, so I began creating wireframes for internal testing.
Initial proposal wireframes
I shared the wireframes with internal marketers and sales enablers for feedback. They unanimously supported reflecting the folder taxonomy to save time on content organization in Seismic. Feedback also emphasized prioritizing new or updated content for better discovery. After presenting the findings, the Product Manager asked me to align the designs with our design system, which I updated.
Updated designs to align with design system
To reflect folder taxonomy, the Product Manager and I discussed adding metadata tagging to both the content and Collection. This would allow marketers to organize and permission content using their existing folder structure, reducing the need for multiple content locations. We explored using metadata to create single Collections with content tailored to user permissions.
Permission Model Explorations mapped
Based on our initial feedback, I felt that the single Collection with unique views for each user persona made the most sense, and after some discussion and validation of the concept with our Product Manager and developers I began mocking up some concepts.
Initial design mock ups
Following interest in our metadata permissioning model, the VP of Product requested exploration of a single Collection experience with content based on user permissions. However, testing showed marketers were concerned about sellers navigating too much content with limited control. In response, the Product Manager suggested adapting the metadata model to apply tags based on content usage within Collections, which we explored through wireframes.
Collections with automatic metadata tagging
Three months in, we faced a setback: we could no longer build a standalone product and had to integrate the Collections model into the existing architecture. As a result, most of our explored solutions, including metadata tagging, folder taxonomy, and the visual layer, were unfeasible. To stay on track, I reassessed our JTBD with user feedback and mapped out IA flows to ensure we could implement viable solutions.
Updated JTBD with user quotes
IA flows
Model of current latency Issues
Using these assets, I quickly developed an MVP solution by adapting our existing flow and mapping previous designs to a new one. After rapid internal testing, despite some disappointment over lost features, we received positive feedback on the content activation flow and the simplicity of the new model.
Marketer Collection creation flow
Seller Collection view
With a strong understanding and presentation of the existing ecosystem and ideal workflows, I was able to quickly pivot on a short timeline and get quick approval from the stakeholders. My Product Manager and our developer team were critical in getting day to day feedback about potential issues and solutions that would be feasible within the current architecture constraints.
Highlights
I have never had to pivot so late in a project and still have the expectations of delivering an impactful MVP product. This project was tumultuous to say the least, but I learned a lot about the value of understanding the technical stack and potential limitations. While we were unfortunately not able to make as large of an initial impact as expected, we have been able to adapt a lot of our previous features into roadmap items including; metadata tagging, content views based on permissioning, and enhancements to our existing visual layer.
Learnings
Our product is currently launching in EA and we expect to have plenty of actionable feedback by our new user base. I am very optimistic that our product cycle will allow us to release features in a meaningful way so that by our product's release date later this year we are close, if not at our initial features we explored. As a designer I definitely found that I was disrupted quite often by either new requests or asks for additional feature enhancements, I believe I could have mapped out the IA flows earlier to show the value of potentially building out a new solution as opposed to the short-term savings on developer overhead. Overall I am happy with this project and believe it developed my skills in IA, UX, and UI immensely.