Portfolio case study / October 2026
Learning inside
the decision.
A short microlearning prototype for partner learning coordinators reviewing a request to adapt facilitator-training materials.
Status: independent concept and functional prototype. The programme, people and request are fictional. This is not commissioned, endorsed or field-tested by War Child or another organisation. It does not teach a mental-health intervention.
The performance problem
A team needs to make learning materials work in a different context. Approving every change can remove the practice a learning objective needs; freezing every detail can create unnecessary barriers. The sample asks learners to judge a change against a clear brief, involve users and recommend a small pilot.
Audience and objective
Intended audience: partner learning coordinators and trainers who already understand their programme and have access to its technical guidance.
Observable objective: given an adaptation request and a programme brief, choose a recommendation that addresses an access barrier, retains the intended practice, involves intended users and specifies a pilot before wider use.
Why this is in-product microlearning
The learning activity appears beside the request it supports, inside a simulated partner workspace. The learner moves from “Needs review” to “Pilot recommended” through short decisions. It demonstrates a point-of-work interaction; it is not integrated into a live partner platform.
Storyboard and learning sequence
| Step | Learner action | Design purpose |
|---|---|---|
| Context | Read a request and explicit adaptation boundaries. | Make the decision and relevant evidence visible. |
| Decision 1 | Distinguish access adaptations from removing practice. | Align the change with the objective. |
| Decision 2 | Choose a pilot involving varied intended users. | Include partner knowledge and observation. |
| Decision 3 | Respond to a gap between positive reactions and observed performance. | Use evidence to guide the revision. |
| Transfer check | Review a new offline-format request. | Apply the approach beyond the first example. |
| Takeaway | Review targeted reminders and open the job aid. | Support discussion after the lesson. |
Feedback and adaptation
Each choice receives a specific explanation. An incorrect choice adds a focused hint and another attempt. The final recap highlights only the decisions missed on the first attempt. This is simple rule-based remediation, not an AI tutor. The activity reports first-attempt decisions separately from completion so repeated attempts cannot appear as independent mastery.
Inclusive delivery choices
- Text-first content, no autoplay, no timed response and no required media downloads.
- Native radio controls, visible keyboard focus, feedback announcements and focus moved to new content.
- Responsive layout and a print-friendly job aid for facilitated or offline discussion.
- Local fonts with system fallbacks. The interactive lesson requires JavaScript and an initial connection; it does not claim offline caching.
The prototype is English only. A production version would require partner-led language adaptation, review of meaning and testing with intended users. Accessibility and assistive-technology testing would form part of that validation.
Proposed co-creation and testing
- Agree the boundaries. Work with a technical lead and partner trainers to define the objective, required practice and local approval process.
- Explore the real context. Ask what is difficult in current materials, which examples work, and how people access learning.
- Test a small prototype. Invite approximately five to eight intended users with varied relevant experience. Observe unaided choices, unclear wording, navigation and use of the job aid. This is formative testing, not an effectiveness study.
- Review together. Compare first-attempt choices, reasoning and task barriers. Agree changes and retest the revised material.
- Improve after use. Invite trainers and a community of practice to share de-identified implementation observations. Review recurring issues with partners and maintain a versioned change log.
Evidence to collect in a real pilot
Time to make the recommendation, first-attempt decision patterns, the learner’s explanation of what must be retained, assistance needed and whether the job aid supports a later adaptation conversation. No field results or impact metrics are claimed. This public demo stores responses in memory only and sends no learner responses to a server. The hosted portfolio records basic page usage, link clicks and active time; see the privacy settings.
Build and ownership
HTML, CSS and JavaScript. Concept, content and implementation developed by Kevin Gakarau with AI assistance. This sample was not authored in Evolve and does not claim SCORM, xAPI or LMS integration.