Support Center

I was charged with standing up the first knowledge base and contact form for Lyft Business customers. With a 10 week timeline and limited resources, I wore multiple hats—handling UX research, information architecture, system configuration, content strategy, UI/design and orchestration across a cross-disciplinary team.

Stakeholder team: Head of Customer Engagement & Advocacy, Director of Lyft Business Product

Partner team: System Tools Administrator, Product Marketing Manager, Customer Support, Account Manager, Engineering, Sales Enablement, Copywriter

Goals

• Empower customers to self-help/troubleshoot product issues in order to reduce their need to contact support
• Establish a centralized escalation channel in order to aggregate and structure data related to customer issues
• Eliminate 15+ ancillary email aliases currently in-use for customer support

Constraints

Time constraints: 10 week timeline to understand and build for a complex product landscape (multiple B2B products x multiple industry verticals x multiple use cases x multiple provisions)

Resource constraints: Content Specialist role had not been filled—also requiring my participation with interviewing candidates over project duration

Technical constraints: Single routing/no reroutes
, no conditional fields/logic, no intake data

Gathered requirements from key stakeholders (CEA, Product, Marketing, Sales, CET)

Attended customer-facing training sessions to get a feel for the expectations we set and level of training provided to new users

Learning: Attended internal sales enablement syncs to see what questions our AMs were unable to answer for customers. I learned engineers were often the only person in the meeting who knew the answer to these usability questions. I knew we had to do a better job of centralizing and surfacing this info by making it available internally and externally.

Associate workflows

Shadowing support associates to understand how they navigate their internal toolset in order to resolve customer issues.

‍

Project timeline

‍Backed-out of deadline (loosely defined timeline leaving room for scope changes)

Configuration evaluation

Partnered with systems architect team — prepared an analysis of 4 possible build methods against criteria and decided on shared backend and custom frontend for a "best of both" approach.

Competitive analysis

I looked at the relationship of properties commonly used in the B2B space (Help Centers <> Contact Forms <> Marketing Sites — stand alone vs. integrated)

‍

User definitions

Because this knowledge base had to work for all customers using a variety of B2B products across many use cases and industry verticals, I created persona-based user stories to benchmark assumptions against.

‍

‍

Content audited across disparate sources and grouped across customer lifecycle. This involved gathering one-off FAQs, static troubleshooting docs, training decks, getting started PDFs, etc. This helped us gauge the content types to architect around. As we explored IA and decision tree logic for selection pathways —relationships between issue types were A/B tested with active users for findability (Optimal Workshop, Google Forms).

Contact form

With the constraint of no conditional logic for beta, our contact form had to work for any issue type—meaning we needed to include the right balance of inputs that would help an associate resolve most issues upon first-touch. In order to look at the inputs required to solve any issue and score their importance/frequency, I created an evaluation matrix.

When it came to the order of inputs for the contact form, we modeled the structure after the natural flow of an associate support interaction on chat or phone—using their investigation steps as a template.

‍

• Launched a first-generation knowledge base (support center) and contact form in less than three months
• Enabled customers to know where to go for support and empowered them to self-help
• Enabled business to monitor contact volume + issue types with structured data
• Built camaraderie and engagement across LB team by involving partners throughout the business = via  shared success
• Earned trust across the team and a proven ability to deliver

Tradeoffs made:

• ‍Limited time meant a tradeoff of ideally validating the SR/RR classification system we were building on top of and instead operating on good faith that the data structure was viable
• Working with an off-the-shelf theme and applying custom styling helped deliver on-brand and quickly
• Our inability to require sign-in and ingest User ID data upon initial launch meant our contact form had to work for all cases resulting in more inputs than desired
• Team commitment as a fast-follow to audit and sunset all ancillary email aliases by redirecting customers to the contact form

‍

‍