Service Canada Labs
Service Canada Labs: Unblocking usability testing in a stalled approval process
Summary
The Service Canada Labs website was envisioned as a central hub where product teams across the department could showcase prototypes, receive public feedback, and recruit participants for user research. The idea was to provide early, continuous feedback to improve federal services.
However, showcasing incomplete prototypes was new to government, and the IT "authority to operate" approval processes were built for fully launched, data-processing products. By the time I started, progress had been stalled for 8 months, as many teams couldn’t get their prototypes on the site. I identified the root cause, escalated the issue, and resolved it within 8 weeks—unblocking many product teams across the department.
Context
The team aimed to create a public-facing platform for testing prototypes and gathering feedback from Canadians. The idea was that teams across the department would use the site to engage the public in early-stage research.
But the platform became stuck in a lengthy technical assessment process. Broader scrutiny of government timelines also underscored the need to move faster.
My role
As product manager, I was responsible for securing authority to operate. My focus was identifying the bottleneck and guiding the product through approval.
This meant:
Diagnosing the cause of the delay
Escalating to leadership
Advocating for a reassessment approach
Collaborating with a new assessor to finalize the evaluation
What I found
The problem wasn’t the platform—it was the evaluation approach.
The assessment treated the site like a traditional, on-premises system that stored personal data. In reality, it was a cloud-based site that simply linked out to prototypes and didn’t store personal information.
This mismatch introduced irrelevant requirements and delayed progress. The evaluation framework didn’t align with the product’s risk profile.
What I did
I escalated the issue to the Director of IT Security, recommending a reassessment by someone with cloud expertise and a revised evaluation that matched the platform’s actual risk.
Once a new assessor was assigned, we worked closely to:
Clarify architecture and real risks
Align on what needed review
Move quickly to final approval
Outcome
Reduced the timeline from 8 months to 8 weeks
Unblocked numerous product teams who were waiting for their prototypes to go live
Helped shift internal understanding of how modern digital services should be assessed
Reflection
This experience showed me how delivery can stall when evaluation processes don’t keep up with modern product development. My role wasn’t to redesign the process, but to spot the misalignment, challenge it, and bring the right expertise in to keep us moving forward.
This created repeated deadlock, as every option introduced tradeoffs that prevented consensus.
Insight
The team wasn’t stuck because the decision was complex, but because they hadn’t identified a viable alternative that could satisfy multiple requirements at once.
Given the Prime Minister’s role and the representational expectations of the page, no single image could fully satisfy all stakeholders or reflect the breadth and diversity required.
Once that became clear, the opportunity shifted from forcing consensus on one option to expanding what was possible within the system.
Solution
I proposed replacing the single-image decision with a curated set of approved images that rotate on page refresh.
This shifted the decision from:
“Which one image best represents the Prime Minister?”
to
“Which set of images adequately reflects the breadth and diversity of the Prime Minister’s role?”
I worked with the design team to ensure accessibility requirements were met, including a 33% black overlay to maintain text contrast for WCAG compliance.
Working with engineering, we implemented responsive focal-point handling so images remained properly centred across devices.
I also worked with political and communications stakeholders to define a set of images that reflected both institutional needs and the breadth and diversity required for public representation.
Outcome
The project moved from months of stalled discussion to launch
Stakeholder alignment improved by reducing perceived risk
The solution was simple to maintain and required no ongoing operational overhead
The final experience better reflected the breadth of the role and its public context
Reflection
This reinforced for me that many design problems are actually constraint and decision problems.
When a team is stuck, expanding the solution space—rather than optimizing within an assumed constraint—can be what enables progress.