Go Back
Artificial Intelligence
Gen AI, Custom LLMs, Machine Learning & Computer Vision
Data
ETL & Storage, Visualization, Processing & Enrichment
Cloud
Infrastructure, DevOps, APIs, Automation & Scalability
In early 2018, a few Google data center teams ran a proof of concept to evaluate the effectiveness of switching to a queue-based execution model for maximizing the throughput of build project work. The proof of concept proved that significant improvements could be made to increase workplace efficiency.
These data center teams used various software tools to gather relevant information and achieve specific tasks. The goal? To increase task success by creating a product suite that boosts overall happiness and our operation teams’ productivity.
The Google Opus team wanted to transform both business processes and software systems with a new product suite that would maximize work throughput, accelerate the onboarding of new products into data centers, guide all operations teams towards the most critical work on the floor, better orchestrate work between teams, and minimize the number of task-specific tools. The team recognized the challenge as a perfect candidate for a design sprint, so Google’s Product Management team reached out to Rootstrap to begin the process.
Creating a Framework and Mapping out Requirements
Rootstrap delineated the goals, logistics, team, and schedule. The Sprint Challenge was defined as finding new, innovative ways of transforming processes and tooling so that our operations teams can reliably, accurately, and quickly meet teamwork order goals. Before starting the sprint, we delved into existing Opus requirements and data center software tools. While a traditional design sprint would be five days, they had only two days due to scheduling constraints. The goal was to rapidly generate actionable ideas that could be completed in a more leisurely timeframe afterward.
Facilitating a Design Sprint Workshop, Day One
On the first day, we quickly dove into lightning talks to better understand the product goals, the design evolution of the existing software tools, and current user journeys. We then promptly identified each persona’s path from start to end goal and mapped out all main user journeys to provide the team with ideas on reducing friction and improving UX. This was a relatively large and quick Sprint, with upwards of a dozen people working together for just two days.
Facilitating a Design Sprint Workshop, Day Two
The second day was dedicated to expanding each participant’s ideas into presentable solution sketches they would share with the rest of the team. The group discussed each sketch and highlighted the best features, ultimately voting on the best solution to prototype.
Presenting our Findings
The team was then divided into two groups: one to create a final storyboard of the winning solution sketch and another to discuss the technical feasibility and other solutions to any of the technical challenges they identified. The Opus Team presented their results to the Google team members unfamiliar with the project, and they discussed how to implement the solutions moving forward.
At the end of the 2-day Design Workshop, Rootstrap defined a clear vision and action steps for the coming year. The Google Opus team began user testing and iterating on their solution from the Sprint process and used the defined foundation to expand the full Opus product suite. Solutions included introducing collaboration tools, documentation, and a set engagement model, and the brainstorming sessions revealed the need to codify internal communications and processes within the Google Opus team.
