Staq platform modernization

Context

The Staq modernization programme was created to deliver measurable returns within two years on a twelve-month investment. Rather than rebuilding the platform, the strategy focused on modernizing approximately seventy-one ChMS pages in place, improving user experience while preserving existing functionality and minimizing engineering disruption.

The first modernization initiative carried the greatest UX risk. Ministry leaders were experiencing friction caused by inconsistent interaction patterns, outdated interface conventions, and unnecessarily complex workflows. While feature capabilities remained competitive, the overall experience was contributing to lower satisfaction and eroding NPS.

The broader modernization portfolio extended beyond visual refresh. It included unifying identity and data across ChMS and Donor Management, introducing the first cross-sell module, and surfacing partner-platform context directly within Staq profiles.

The challenge was not whether modernization should happen. It was how to prioritize the opportunity space, establish a scalable delivery model, and generate validated design work at a pace the programme could absorb.

---

Approach

Click study audit as the prioritization engine

I initiated and led a click study audit across the ChMS pages in scope.

The audit combined qualitative usage research with quantitative Pendo analytics, allowing behavioral data and customer-reported friction to be evaluated side by side. The resulting framework became the primary prioritization lens for modernization decisions.

Using the audit, leadership could distinguish between areas that required:

• Like-for-like modernization for consistency and visual alignment
• Workflow optimization to reduce user effort
• Consolidation or deprecation of low-value functionality

The work also identified the ten highest-frequency workflows where click reduction and workflow simplification would create the greatest customer impact.

Rather than prioritizing based on opinion, the programme gained a repeatable, evidence-based model for investment decisions across approximately seventy-one pages.

Designing the modernization delivery pipeline

To make modernization scalable, I designed the end-to-end delivery pipeline used across the programme and refined it through implementation.

The pipeline established clear outcomes, decision points, and ownership at every stage:

Research and plan Customer feedback from Dovetail, Pendo usage data, session replays, and support documentation synthesized into research findings and JTBD hierarchies. Owner: UXPM

Prototype and review Feature scope definition, North Star concepts, AI-generated prototypes, engineering feasibility review, and peer review. Owner: UXPM

Design and review High-fidelity Prism 1.5 designs, design system coordination, peer review, and CX review. Owner: UXD

Test and validate Internal CX testing for lower-impact changes and unmoderated user testing for medium- and high-impact changes using JTBD-aligned evaluation methods. Owner: UXR

Iterate Design refinement, validation review, and workflow sign-off. Owner: UXD

Collaborate and build UX and engineering collaboration using AI-assisted workflows to accelerate implementation and contribute improvements back into the shared component library. Owner: Engineering

The pipeline became the operating system for the modernization programme. It created alignment across teams, reduced ambiguity around ownership, and allowed multiple workstreams to move simultaneously without sacrificing quality.

End-to-end UXPM across multiple workstreams

Alongside designing the process itself, I acted as concept designer and end-to-end UXPM across several modernization initiatives in parallel.

I owned the concept development, prototyping, design specifications, validation strategy, stakeholder decisions, and readiness process through to development handoff.

This reduced the number of handoffs on the critical path and allowed feature teams to consume validated, implementation-ready work rather than partially defined concepts. As modernization scaled, this role became less about producing artifacts and more about maintaining continuity from discovery through delivery.

AI-assisted prototyping as a scale multiplier

AI-assisted prototyping became the primary speed lever within the programme.

Using Claude-generated HTML prototypes, existing product screenshots, workflow definitions, and North Star concepts, I established a rapid concept-development workflow that significantly reduced the time between idea generation and stakeholder review.

The resulting prototypes were intentionally grounded in production reality. They leveraged Prism 1.5 components, realistic data structures, and interaction patterns aligned with engineering constraints. Rather than functioning as disposable explorations, they served as early implementation drafts that accelerated both design and engineering conversations.

The increase in output was significant enough that the team ultimately doubled its design review cadence to keep pace.

That shift was an important signal. The bottleneck had moved from design production to organizational review capacity, demonstrating that the programme’s limiting factor was no longer generating solutions but evaluating and operationalizing them.

Decision governance through the decision register

I maintained the modernization programme’s decision register as the primary source of truth for scope-level decisions.

The register was intentionally distinct from Figma. Visual design decisions remained within design artifacts, while the register captured decisions that altered scope, affected integrations, deprecated functionality, introduced dependencies, or changed user workflows.

Each entry documented context, stakeholders, implications, ownership, and status.

Examples included autosave behavior on settings pages, deprecation of the ChMS Financial section, and removal of the Leadership tab from Campus Settings pending Metrics dependencies.

As modernization expanded across multiple teams, the register became a governance tool that maintained visibility, alignment, and accountability without bloating design specifications.

---

Reflection

The most valuable lesson from this programme was that AI-assisted prototyping was not primarily a design acceleration tool. It was a constraint-discovery tool.

The immediate effect was obvious: concept-to-review cycles compressed dramatically, allowing us to produce validated design work at a rate that would not have been possible through traditional design workflows. The more interesting effect came afterward. As design throughput increased, the bottlenecks shifted elsewhere. Review capacity became constrained. Decision-making cadence became more visible. Design system gaps surfaced faster. Engineering teams needed clearer guidance on how to consume increasingly realistic prototypes.

In practice, the challenge stopped being how quickly we could produce design concepts and became how efficiently the organisation could evaluate, align on, and absorb them. Doubling design review sessions was the most visible response, but the more important interventions were operational: maintaining a disciplined decision register, clarifying ownership through the modernization pipeline, and ensuring work moved through validation and sign-off without accumulating hidden ambiguity.

If I were starting the programme again, I would invest earlier in the review and governance model. The increase in throughput was predictable, but the downstream impact on review capacity and decision-making was not planned for early enough. A more structured review system, with smaller decision-making groups, clearer asynchronous pre-reads, and predefined approval criteria, would have allowed the organisation to capture more of the value unlocked by the accelerated prototyping process.

Ultimately, the success of the effort was not that we designed faster. It was that we learned how to convert increased design velocity into a repeatable delivery system that multiple teams could execute against in parallel.