Pivotal, Parrit App
PRODUCT MANAGER & PRODUCT DESIGNER
Open Source Pair Programming Tool MVP Launch
While working at an agile software company the need for streamlined tools is a necessity and if there is not a tool in place, one is made.
Parrit was incepted out of a need to streamline the daily rotation of pair programming project teams. Some teams used spreadsheets to manage their pairs and others an analog board that had manual rotation with sticky notes, however these methods were not entirely efficient. What about something that is automated? And tracks the history of pairs?
With a validated hypothesis in mind the Product Owner, a software engineer at the company, put together a proof of concept demo which led to the formulation of a dedicated project team to help grow Parrit. I led design and product management for this growth including, lean user research, branding, visual design, front-end development, and data analysis to develop the React based app and help gain adoption. Work was/is implemented in an agile fashion and kept in digital format to align with public availability for the open source community.
Feel free to use the project name 'test' and psw 'test' to play around with Parrit.
Top Skills: Qualitative Research, Service Design, Persona Development, Design Strategy, Product Design, Product Development, IA/UX/UI, Agile Practices, Project Management, React, CSS
Project Goal
Build a tool for pair programming teams while learning new technology and experimenting with process.
PROBLEM SPACE
Project teams were using a great deal of cognitive load to remember pairing history for these daily manual boards.
Fun yet how do you know what happened last week?
This team started with whiteboard, then moved to spreadsheet.
Project BUY-IN
A low fidelity proof of concept (POC) was developed in code by founding engineer and tested with one team to validate usage.
POC (pictured on the projector screen) met user needs but was hard coded for one team and had bugs. Above taken from a tech talk given after product launch.
Next Steps for Parrit
Buy-in from the POC led to a dedicated project team with myself in a solo Product role along with two Engineers.
LEAN RESEARCH
Facilitated work sessions with project team for alignment upon goals for what to explore and validate through user research.
Started by brainstorming assumptions.
Mapping assumptions about user needs to prepare for user research.
Top Hypotheses to Validate
The primary user of Parrit is an Anchor Engineer on the project team who manages the daily pair rotation.
The POC app will meet needs of more than one team, plus will not be cumbersome to use for especially large project teams.
NOTE: Interview subjects were individually selected based on role and experience anchoring large and small project teams.
HYBRID METHODS
Research lead for open-ended discussion and usability sessions along with synthesis and analysis of feedback.
To remain lean, tried out using an online tool for synthesis and analysis, See Board >
ASSERTIONS FROM RESEARCH
We validated the primary persona from our hypothesis to be an Anchor Engineer.
Fred is the primary persona yet two others were outlined to validate in future research, See More >
Key Research Insights
Deciding on pair rotations is generally a quick decision making process because it needs to happen at the beginning of the day, directly after project team stand-ups.
Keeping track of pair history is important to ensure 'newness' of pairs for codebase knowledge, but on large project teams it is difficult to remember past pair rotations.
DESIGN EXECUTION
A sketching session helped identify the signup flow and basic user interface layout.
Whiteboarding our solution before heading into Sketch App for interface design.
LAYOUT + INTERACTIONS
Wireframes were created for story writing to drive engineering, and determine interactions.
PRODUCT MANAGEMENT
A backlog was created to keep the engineers working and we followed user-first prioritization by addressing persona needs.
We built up a backlog for engineers to pick up when they have time, See Pivotal Tracker >
IDENTITY BRANDING
In preparation for interface design of the product build, design focus shifted to create a logo.
Decision Making
A consensus of feedback from passersby determined which logo design to move forward with. The project team then narrowed it down to the below two color palettes.
VISUAL DESIGN
To iterate upon design a Pattern Library was developed from common visual interface elements.
Our research highlighted desktop as primary use and phone for secondary users.
MVP LAUNCH
To launch the product a company-wide email was drafted, approved and sent to all global employees.
Company-wide email blast led to complications:
âž¡ 333 new users in first few hours
âž¡ Crashed our free PWS account
âž¡ Plus attracted legal concern!
PRODUCT ADOPTION
Some months after product launch the team gave a lunchtime Tech Talk about Parrit.
SIDE NOTE: The audience was especially interested in how the logo was designed which led to a design discussion about negative space.
Open-Source Contributions
After product launch the team dispersed but continue to contribute (even today) as a community.
Below represents work after leaving the company.
Communication Channels
With open-source our project became part of a community with a need to maintain documentation across channels.
Parrit app continues to be iterated upon with code base available on GitHub.
We collaborate over Slack plus take feedback from users through a Help channel.
PRODUCT GROWTH
Launch included a survey to collect feedback and overtime this was cross-synthesized with GitHub issues for growth strategy.
We cross synthesized survey feedback with GitHub Issues six months post-launch, See Board >
Continuing to maintain our backlog on Pivotal Tracker.
UI UPDATES
Refinements were made to contributor designs and new features created for product planning and backlog.
Always User-First
The feature updates designed for above mockups (plus added to the backlog) came from synthesis of feedback through GitHub issues, survey submissions and Slack.
NOTE: The team sees an opportunity of showing by example to the open-source community by applying user-first methods.
DESIGN System
A product management need arose to create design guidelines for assistance in maintaining brand continuity with contributors.
Design Patterns
The pattern library was created for engineers and moved from Zeplin to InVision for public ease of use.
The design system is a work in progress feel free to, View InVision Project >