BLOOMBERG
Streamlining Enterprise Collaboration
Bloomberg's Collaboration Products
Over the last 25 years the Bloomberg Terminal has evolved from an information service to a collaboration environment. The change occurred haphazardly, one feature enhancement at a time, resulting in an inconsistent and bloated experience. In 2015 I took on the mission of leading a team of UX designers to streamline the collaboration experience across 9 of the Terminal's flagship applications.
My Role
I lead the Collaboration UX Team. I work with product owners to prioritize and define the scope of UX work, inspire appreciation for UX and design among Product and Engineering teams, inspire a business-minded approach to design among my designers, and manage the delivery of my team’s work.
Design at Bloomberg
• Historically the firm has not been design-driven
• The tech stack makes use of an extremely constrained front-end
• There is only one font
• The UI is based on fixed row heights
• The color palette is high-contrast and limited
Given the constraints, design in this environment is an exercise in streamlining workflows based on a precise understanding of user needs.
Case Study: The SPDL Redesign Project
In the early days of the Terminal a group of enterprising engineers developed a small application for managing contacts within the system. Within months, usage of the Contacts App skyrocketed and users were asking for more sophisticated features, especially a way to create groups of contacts and share them with teams. The Contacts App (called SPDL) worked well and became one of the most-used functions on the system. Engineering efforts moved on to more urgent areas.
In 2016 we took a hard look at the Contacts App and discovered that it had become a mess. The interface and workflows that had worked well at the beginning had incrementally become unusable.
Goals
1. Make the Contacts App simple again.
2. Give users the powerful features they need so they don't have to hack the groups feature.
3. Create a scalable UI that can grow.
Process
1. Research: Interviews with SME’s, cognitive walkthroughs
2. Ideation: Brainstorm including Product, Sales, & Engineering
3. Prioritization: Design roadmap
4. Design Iteration: Two design tracks, V1 and beyond
5. Test: Usability and concept testing
Formative Research
We stepped back from the legacy application and conducted a round of formative research, talking with 20 users representative of key types, to learn the current fundamentals of their business needs and workflows.
Ideation
Building on a refined understanding of the key workflows, I ran a workshop with engineers, designers, and product managers to brainstorm new features tailored fit to client needs.
Prioritization
The features generated in the brainstorm phase were too ambitious to include in a single release. I worked with product and engineering counterparts to create a high-level prioritization to focus our design efforts.
Design
The V1 redesign of SPDL focused on key views driven by specific user needs.
The Contacts View
User Need
As someone keeping track of my colleagues, I need to know find them quickly and determine their online status.
A variety of UI's were tested for the Contacts View. Given the specific need to find contacts quickly, see their status, and given our users' familiarity with tools like Excel, it's not surprising that a simple data grid was the preferred UI.
The List View
User Needs
• As someone thinking about contacting clients, I need to know which mailing lists I have access to.
• As someone thinking about using a list, I need to understand:
• where the list came from
• what’s in the list
• what I can do with it
The Contact Detail View
User Need
As someone thinking about contacting a colleague or a client, I need to understand the context of our relationship.
Usability Testing
We identified 12 critical tasks in the redesigned version of SPDL and conducted lab based usability testing. As a UX team we had concerns about some tasks even though they had relatively high success rates. These tasks revealed a a high degree of confusion on the part of users despite the fact that they eventually were able to complete them.
Refining the design
Following usability testing we refined several of the key workflows. Shown here is a refinement to the create-a-new-contact flow.
The Design As Tested
The create contact view included a first and last name field for new contacts. But the autosuggest menu could only work on one field (the first name field). Users who failed to select a contact from that menu in the first name field were stuck filling out fields that should have been filled out for them.
The Refined Design
The refined design included only a single name field instead of a first and last name field. This made it much more accurate for users' selecting a contact from the autosuggest menu. But it made the form less precise for users entering new contacts not currently in the autosuggest. The tradeoff was worth it since most users select contacts from the autosuggest.
