CRYSTABEL WOO REITER
Resume                        

I’m a product designer based in the DC metro area, with 6 years of experience, specializing in the dynamic B2B startup realm. 

I love the challenge of demystifying complex topics into intuitive, visually helpful products. 

When I’m not designing, I’m reading, doodling, winning plushies from claw machines, and cuddling with my corgi-pomeranian, Yeppy. 

Home




VA.gov Documentation 

 
A KNOWLEDGE HUB FOR THOSE BUILDING ON VA.GOV. 


THE PROBLEM
We created a myriad of documentation that was designed to help our customers find what they need to work seamlessly and iteratively, and build great products on VA.gov. 

We have observed that the resources as they exist today aren’t effectively enabling that, which  is frustrating our customers, and is potentially resulting in products on VA.gov that don't meet standards. 

 
ROLE

UX/UI designer

TIMELINE

December 2019-March 2020




Discovery sprint 
schedule


DAY 1: RESEARCH
Prioritize what to focus on

DAY 2: IDEATION
Explore possible solutions

DAY 3: PROTOTYPE
Decide on MVP and start prototyping

DAY 4: TESTING AND ANALYSIS
Validate MVP

DAY 5: RECOMMEND A SOLUTION
Present our findings to stakeholders




   Day 1 Research



Research goal

Learn how we might improve our documentation resources so that our customers are more successful. 

Study type

User interviews
Existing research
Proto personas

Participants

4


Pain Points

‘
Documentation is hard to find or nonexistent 
Documentation is incorrect or untrustworthy
No feedback loop
Teams do not know what documentation they are responsible for creating and maintaining
Barriers to keeping docs up to date (technical, logistical, and cultural)





Proto - Personas


With the interviews we gathered, we were able to brainstorm a couple of proto-personas. They were about 8 in total, but we decided for this iteration to go with just 2. 





Day 2 Ideation






Crazy 8s

I led a workshop where we did the classic design sprint exercise crazy 8s. Drawing 8 ideas in 8 minutes. 

From there, we explained our ideas, voted, and were able to create 1 board of features/ideas we’d like to include in the future. 





Day 3: Prototyping  




Wireframing


From synthesis of our discovery interviews and past research, we came up with this hypothesis: 

Creating a documentation homepage with a table of contents and powerful search functionality for existing documentation will enable people to find what they need and ramp up quickly.

I created a lo-fi prototype of a homepage, documentation page, and an editing interface for the documentation page. 



Day 4: Testing & Analysis  


Testing


Research questions

  • Are users able to find documentation when provided with a landing page that brings together all types of docs?
  • What is our users' comfort level with markdown editors?
  • Do users utilize features that allow them to help keep documentation up to date?

Study type

Unmoderated click test
Participants

43 participants

Tools

Chalkmark



Findings



Result 1:
Findability: Most users do not first click to search -- average 9% first click search on findability tasks

What did that mean?
  • Search was still gonna be a part of MVP - it was a high severity pain point we noted earlier in research
  • May show that users don’t trust the existing search 
  • Users may return to search after first click 
 Result 2:
Findability: Without search, findability success varied widely -- first click success 31-72%
What did that mean?
  • More research was needed on the Table of Contents links on sorting 
  • Users may return to search after first click


 Result 3:
Findability: Preference to access all docs from one location -- 88% stated this would help findability 
What did that mean?
  • MVP has all docs accessed from single entry point
  • Some user interviews stated differently -- will need to validate this approach as we go


 Result 4:
Editing: Users will comfortahble editing in Markdown -- 81% were moderately or very comfortable
What did that mean?
  • WYSIWIG was not part of the MVP


 Result 5:
Editing: Users vary widely in likelihood of updating inaccurate docs -- 34% unlikely; 53% likely
What did that mean?
  • Cultural and process clarifications needed regarding who should update what docs


 Result 6:
Editing: Users were willing to identify inaccuracies or suggest change
What did that mean?
  • Users were most likely willing to contribute to maintenance given clear instruction



Day 5: Recommend a Solution  



    Our MVP Approach

    Our MVP was to improve findability and trustworthiness of existing docs 
    • We used Algolia for our search technology
    • We would make a custom homepage with links to key resources and basic insights from Algolia and Google Analytics
    • Standardize and require use of ”Owned by” and “Last updated or reviewed” conventions 



    Visual Design




    Hifi designs



    This was a very fun project to me and I got to use the VA.gov design system but also create a bit of a sub-brand. 

    The illustration in the landing page header was inspired by collaboration and work done between the platform team and the product teams building on VA.gov. 







    Project Monitoring



    Was the project a success?The documentation homepage decreased average weekly support requests by 6% 


    The Platform Customer NPS increased by 2 points that quarter


    Job Search case study