Product strategy · Evidence synthesis · May 2026

Strategic product review ahead of a platform rebuild

MVP scoping · User group prioritisation · Evidence synthesis

The brief
Synthesise six parallel evidence workstreams into a decision-ready recommendation for a platform rebuild ahead of an organisational acquisition.
My role
Solo analyst and strategist. Designed the synthesis approach, led all six workstreams, wrote the consolidated recommendations document.
Evidence base
Product analytics (ECD Connect and CHW Connect), 19 usability test sessions across 5 years, qualitative research with 290 participants, 608 categorised support queries, GA4 data.
Output
A structured recommendations document covering problem definition, user group prioritisation, MVP feature options, UX principles, and a research agenda for unresolved gaps.

An acquiring organisation needed to understand what had been built before committing to a rebuild.

The platform had been in development for five years across three products, serving over 4,000 practitioners and community health workers across South Africa. Drawing on five years of accumulated UX research, qualitative fieldwork, and data analysis, I documented six evidence workstream documents in preparation for the synthesis: product analytics, CHW platform analytics, usability test synthesis, qualitative research synthesis, user support query analysis, and Google Analytics data.

The acquiring organisation needed to know what was working, what wasn't, which users to design for first, what the MVP should and should not include, and where the evidence gaps were that required further research before design could begin. I synthesised the six evidence documents into a recommendations document.

Defined the practitioner-level problem.

The acquiring organisation had articulated the sector-level problem well but was less developed on what practitioners themselves needed the platform to solve. The research pointed to two persistent unresolved issues across every engagement from 2020 to 2026: financial instability and professional isolation.

Made user group prioritisation explicit, with tradeoffs.

The new platform's stated scope covered parents, CHWs, practitioners, principals, administrators, coaches, NGOs, and government. Attempting to design for all of them in a 3 to 4 month MVP window would produce a platform that served none of them well. I proposed a clear prioritisation:

MVP: principals and practitioners

The available evidence base is strongest here and the platform is only useful to all upstream users if practitioners actually enter data.

Next phase: coaches

Next phase: coaches Coaches are a potentially meaningful mechanism for practitioner adoption at scale, and the most digitally confident user group in our usability data. In practice, the three white-label organisations we worked with did not prioritise coaching features and uptake was minimal. One organisation has since indicated interest. Organisational appetite needs to be confirmed before investing in design and build.

Defer: CHWs

A competing platform was already in use in some provincial health departments and may become the national standard for CHW data capture. Building a parallel tool risked duplicating infrastructure. The right move was to understand that landscape, and potential partnership opportunities, before making any rebuild commitment.

Future phase: parents

A parent-facing product is a distinct use case requiring its own discovery research before any design work begins. Scoping that into a 3–4 month build window alongside the practitioner platform would not leave enough time to do it properly.

Future phase: administrators, NGOs, government

These groups depend on practitioner data existing in the system first. Build the foundation before the superstructure.

Recommended a narrow MVP against pressure for breadth.

The pattern across the platform's history was consistent: broad launch scope had contributed to drop-off and re-engagement problems across all three products. A bad first experience is permanent in this user population. I recommended choosing one of two primary value propositions for the MVP — community and peer connection, or financial instability support — and building the other into a subsequent phase.

Defined the research agenda for unresolved gaps.

The evidence base was strong on usability, feature preference, and adoption patterns. What it didn't resolve: what would make a financially stuck practitioner find the platform worth opening daily beyond content; whether the community mechanic could be improved beyond what existing social tools already offer; and why high-conversion cohorts still dropped off at the same rate after setup. I designed a specific research protocol for each gap, grounded in questions focused on concrete past behaviour rather than stated preferences.

Recommending against CHW inclusion in the MVP. There was organisational momentum toward including CHWs. The decision to defer was about structural risk: building something that might need to be thrown away if a competing platform became the national standard. Making that call required synthesising the platform evidence with an external sector landscape read, and being willing to say "we don't know enough yet."
Arguing for a narrow MVP. Recommending smaller scope than stakeholders want is uncomfortable. The argument used the platform's own history as evidence: usability tests showed users who encountered a bug early defaulted to paper and didn't return; product analytics showed re-engagement campaigns had no measurable effect on day-one churners. The case for narrowness was grounded in five years of data, not preference.
Flagging what the research didn't resolve. A common failure mode in recommendations documents is false completeness. I included an explicit section on what remained open and proposed a structured research agenda to fill those gaps before design work began. Naming uncertainty is more useful than papering over it.

A 2,500-word recommendations document structured to be actionable:

Problem definition
A clear statement of the practitioner-level problem the MVP needed to solve, distinct from the sector-level framing in the original brief.
User prioritisation
Prioritised user groups with explicit rationale for each deferral, so the reasoning could be challenged and updated as new information came in.
MVP options
Feature options with tradeoffs, not a single prescribed list. Two viable MVP directions with the evidence for and against each.
UX principles
Design carry-forward rules drawn from 19 usability test sessions, formulated as actionable constraints for the new build.
Research agenda
Three unresolved gaps with specific interview questions for each, grounded in past-behaviour framing rather than stated preference.
Process recommendations
Agree on user groups before writing requirements. Programme team sign-off before development. Agree on success metrics before launch.