review
visual
journey-map / experience-map
2026-06-18
Experience Map Findings - AFPS Tracker
Stage 2 framework findings review for the Adaptive Path Experience Map. This page renders the complete proposed non-canonical packet for `research/afps-tracker/_working/preliminary-experience-map-research.md`; canonical `research/afps-tracker/journey-map-experience-map.md` remains gated until compiled approval YAML is provided.
Approval Context
The pasted YAML approved the prior JTBD Timeline framework findings and canonical target. The parent loop has promoted that approved JTBD packet to `research/afps-tracker/journey-map-jtbd-timeline.md`, then recalculated the selected framework queue from `research/afps-tracker/_working/journey-map-run.yaml`. The remaining selected framework is `experience-map`.
| Framework | Intermediate | Current State |
|---|---|---|
| JTBD Timeline | `research/afps-tracker/journey-map-jtbd-timeline.md` | Canonical intermediate written from approved packet. |
| Experience Map | `research/afps-tracker/journey-map-experience-map.md` | Pending approval; this page reviews the working packet. |
Scope
This packet maps the hypothesized end-to-end experience for AFPS power users and AI workflow operators who run repo-local, alignment-first product research and need to understand branch state across AFPS artifacts.
The approved journey-map scope excludes `exec-loop-tracker`, development execution tracking, generic PM replacement, active recommendation engines, and broad research repository workflows. The map describes the operator's experience, emotional arc, pain moments, delight moments, and channel transitions; it does not prescribe UI, architecture, or implementation.
Executive Finding
The AFPS Tracker experience is likely a valley-then-peak arc. The operator begins with confidence in source-faithful AFPS files, drops into frustration and anxiety as product paths, alignment pages, working packets, and task docs multiply, then either recovers trust when a source-linked map proves the current state quickly or abandons the tracker if it feels like another parallel summary.
The emotional center is not generic productivity stress. It is the shift from "the truth is somewhere in these files" to "I can see the state and verify why it is true." The product wins when it reduces rediscovery and preserves provenance; it loses when parsing ambiguity, write-back anxiety, or manual-maintenance burden creates more doubt than the native files.
Experience Phases
| Phase | Doing | Thinking | Feeling | Intensity |
|---|---|---|---|---|
| Idea or branch creation | Starts an idea brief, ICP, competitive-analysis, or journey-map pass; creates or updates product-path files and alignment pages. | "The workflow is producing useful structured evidence. I can trust the files because they are the source." | Focused optimism; low anxiety. | Medium |
| Artifact accumulation | Opens `research/.progress.yaml`, product-path markdown, working packets, alignment pages, and `tasks/todo.md` to understand what exists. | "The state is technically here, but I have to assemble it across several artifacts." | Mild friction rising into frustration. | Medium-high |
| Session restart or handoff | A new Codex/Claude session, future self, collaborator, or client needs the current path state and evidence trail. | "Do I need to reread everything? Will the agent miss a branch or stale approval state?" | Anxiety and impatience. | High |
| Manual reconstruction | Uses editor search, markdown previews, `alignment/index.html`, `.progress.yaml`, and agent rereads to reconstruct active/deferred/archived/promoted state. | "Native files are accurate, but this costs attention every time." | High cognitive load; confidence depends on thoroughness. | High |
| Active evaluation | Compares staying native, using generic PM/workspace tools, asking an AI agent, or trying AFPS Tracker. | "A tracker helps only if it stays source-native and does not become another board." | Skeptical curiosity; trust guarded. | Medium-high |
| First tracker read-through | Runs or inspects a passive source-linked view over real AFPS artifacts. Checks active paths, pipeline stage, approval state, evidence refs, and source links. | "Can I verify every displayed state back to canonical files?" | Relief if correct; immediate distrust if wrong. | High |
| Ongoing portfolio review | Returns after additional skill runs or branch changes to scan current state and decide where to resume. | "Can I understand the portfolio faster than opening five files, and can I explain why each path exists?" | Confidence and control if fresh; annoyance if stale. | Medium-high |
| Handoff or explanation | Explains an active, deferred, archived, revisit-candidate, promoted, or out-of-scope path to future self, collaborator, client, or another agent. | "Can I show the evidence trail without reconstructing it live?" | Professional confidence when provenance is visible; concern when gaps remain. | Medium |
| Drop-off or rejection | Returns to native files, ad hoc summaries, or agent rereads if the tracker misreads state, requires manual upkeep, or overreaches into execution tracking. | "This adds a second source of truth or weakens my review gates." | Disappointment, distrust, or indifference. | High |
Required Gate: Experience Phases
Emotional Arc
The likely arc starts above neutral because the operator values AFPS source artifacts and the alignment workflow. Emotion drops as artifact volume grows and the operator repeatedly reconstructs branch state across YAML, markdown, HTML review pages, and task history. The lowest point occurs during session restart, handoff, or manual reconstruction, when the user fears missing a branch, stale approval state, or provenance link.
The recovery moment is a real repo read-through that shows current product paths, stage progression, evidence refs, and approval provenance with direct links back to source. Relief becomes confidence only if the operator can verify the map and see that it remains subordinate to canonical files. If the map is wrong, stale, manually maintained, or too eager to write back, the arc falls below the native-file baseline and the user rejects the tool.
| Phase | Valence | Reason |
|---|---|---|
| Idea or branch creation | Positive | Source workflow is productive and trusted. |
| Artifact accumulation | Near neutral | Useful evidence begins to feel scattered. |
| Session restart or handoff | Negative | Rediscovery and stale-state risk spike. |
| Manual reconstruction | Most negative | Native files are trusted but cognitively costly. |
| Active evaluation | Guarded | Tracker appeal is blocked by trust concerns. |
| First tracker read-through | Most positive if correct | Source-linked state creates relief. |
| Ongoing portfolio review | Positive if fresh | Repeat use builds confidence when parsing remains accurate. |
| Handoff or explanation | Positive | Visible provenance supports explanation. |
| Drop-off or rejection | Strongly negative | Wrong, stale, or overreaching state breaks trust. |
Required Gate: Emotional Arc
Pain Moments
| Phase | Pain Point | Emotion | Intensity | Evidence |
|---|---|---|---|---|
| Artifact accumulation | State spans `.progress.yaml`, product-path research files, alignment pages, working packets, task docs, and archives. | Frustration | High | ICP pain map identifies scattered state across YAML, markdown, HTML, and task files as high severity/high frequency. |
| Session restart or handoff | Agent or future self must rediscover context after compaction, restart, or a gap. | Anxiety | High | ICP trigger events rank session restart/compaction and repeated rediscovery as urgent; JTBD timeline names handoff as a push force. |
| Manual reconstruction | The operator uses source files because they are trustworthy, but the cost of reconstructing state is repeated. | Cognitive overload | High | Competitive synthesis identifies native AFPS artifacts as the strongest substitute and zero-price benchmark, with low navigability. |
| Active evaluation | The operator worries that a tracker could become a generic PM board or parallel source of truth. | Skepticism | Medium-high | ICP post-purchase dynamics and JTBD anxiety analysis warn against duplicate manual state and scope drift. |
| First tracker read-through | Any mismatch in active paths, statuses, approval state, or evidence refs breaks trust immediately. | Distrust | High | ICP evaluation criteria and JTBD journey risks identify parser fidelity as a decisive risk. |
| Ongoing portfolio review | A stale map after new skill runs forces the user back to source reconstruction. | Annoyance | Medium-high | JTBD satisfaction stage says repeat use depends on correct parsing across future skill runs and no drift. |
| Drop-off or rejection | Write-back or automation appears to weaken approval gates or corrupt canonical files. | Alarm | High | ICP and competitive synthesis both flag write-back trust as high risk. |
Delight Moments
| Phase | Delight Point | Emotion | Intensity | Evidence |
|---|---|---|---|---|
| First tracker read-through | The operator sees active and parallel product paths, current pipeline stage, and evidence refs without opening multiple files. | Relief | High | ICP likely aha moment and JTBD consuming stage both emphasize seeing a branch tree and pipeline-stage view from real artifacts without manual data entry. |
| Manual reconstruction avoided | The map answers "where is each path and why?" faster than `.progress.yaml`, alignment pages, and task docs. | Confidence | High | ICP stated value drivers include seeing every branch without opening five files and knowing why a path is active/deferred/archived/promoted. |
| Handoff or explanation | The operator can point future self, collaborator, client, or another agent to source-linked provenance. | Professional control | Medium-high | ICP goals and JTBD social job emphasize explaining branch decisions with evidence links. |
| Ongoing portfolio review | The tracker remains current after additional skill runs and keeps native files as the source of truth. | Trust | Medium-high | Competitive synthesis says the market gap is passive, source-linked state continuity; JTBD satisfaction depends on no drift. |
| Scope clarity | The product stays research-stage and does not absorb `exec-loop-tracker` or generic PM expectations. | Reassurance | Medium | Idea brief and approved journey scope explicitly exclude development execution tracking and generic PM replacement. |
Required Gate: Pain And Delight Moments
Channel Transitions
| From | To | Phase | Friction Level | Evidence |
|---|---|---|---|---|
| Markdown idea/ICP/competitive files | `research/.progress.yaml` | Artifact accumulation | Medium | Idea brief and ICP describe workflow state spread across markdown and YAML artifacts. |
| `research/.progress.yaml` | Alignment HTML pages and `alignment/index.html` | Manual reconstruction | High | ICP alternatives say `.progress.yaml` is authoritative but not ergonomic and `alignment/index.html` is a page-level index, not state model. |
| Alignment page review | Compiled YAML approval and canonical markdown | Session restart or handoff | Medium-high | Journey-map staged workflow and current approved YAML show the workflow moves decisions through HTML review into canonical files. |
| Local repo/source files | AI agent session | Session restart or handoff | High | ICP frustrations include agents forgetting, compacting, or needing rediscovery. |
| Native files | Generic PM/workspace mirror | Active evaluation | High | ICP alternatives and competitive synthesis identify manual duplication and drift risk. |
| Native files | AFPS Tracker source-linked map | First tracker read-through | Medium-high | JTBD deciding stage says trust is earned only when the map reflects canonical files with source links. |
| AFPS Tracker map | Canonical source file verification | First tracker read-through | Medium | ICP says users inspect whether the tracker faithfully maps `active_paths`, `product_paths[]`, statuses, evidence refs, and alignment pages. |
| Research-stage branch | Build/development tracking | Handoff or explanation | High | Idea brief and journey scope exclude development execution tracking and keep `exec-loop-tracker` separate. |
Required Gate: Channel Transitions
Indifference Zones
| Zone | Why It May Feel Flat | Implication |
|---|---|---|
| Single-path research | Native files and agent rereads may be sufficient when only one path exists. | Validate against multi-path repositories; avoid overfitting the first experience to simple workflows. |
| Generic task/status labels | Labels alone do not solve evidence provenance or approval-state confidence. | The experience must show why status is true, not only what status is. |
| Broad portfolio dashboards | A polished dashboard without source fidelity is less valuable than canonical files. | Treat source links, freshness, and parser trust as experience-critical. |
| Managed hosting before trust | Hosting convenience is secondary before users know the tracker is faithful. | Keep monetization and hosting claims separate from journey proof. |
Experience Principles Inferred From The Map
These are experience requirements, not UI or architecture prescriptions:
- The first successful moment must use real AFPS artifacts, not sample data.
- Every important state claim must be source-verifiable.
- The tracker must feel subordinate to canonical files, especially before write-back trust exists.
- The highest-value path is multi-branch or multi-artifact review, not simple single-path tracking.
- The product should reduce rediscovery anxiety without broadening into generic PM or execution tracking.
Evidence Matrix
| Claim | Evidence Source | Evidence Type | Inference | Confidence |
|---|---|---|---|---|
| The emotional low point is repeated reconstruction across source artifacts, especially after restart or handoff. | `research/afps-tracker/icp.md`; `research/afps-tracker/journey-map-jtbd-timeline.md` | Approved research synthesis | Trigger events, pain map, and JTBD push forces all identify rediscovery and scattered state as core pain. | High |
| The likely delight moment is a correct source-linked repo read-through. | ICP likely aha moment; JTBD consuming stage; competitive synthesis market gap | Approved research synthesis | Relief comes from seeing current state quickly while preserving provenance. | High |
| Parser fidelity is emotionally decisive. | ICP evaluation criteria; JTBD journey risks | Approved research synthesis | A mismapped branch state changes the experience from relief to distrust. | High |
| Generic PM/workspace mirrors create a high-friction channel transition. | ICP current alternatives; competitive-analysis synthesis | Approved research and competitive synthesis | Duplicate manual state weakens source fidelity and creates drift anxiety. | High |
| Write-back should not lead the early experience. | ICP trust risk; JTBD anxiety and minimum switching conditions | Inference from approved research | Users need source confidence before they accept state mutation. | Medium-high |
| Multi-path complexity is the strongest urgency threshold. | ICP trigger ranking; `research/.progress.yaml`; JTBD first thought/passive looking stages | Repo artifact and approved research | Single-path workflows may not generate enough emotional urgency. | Medium-high |
| External commercial urgency remains unproven. | ICP WTP caveats; competitive-analysis synthesis | Approved research caveat | The emotional map is hypothesized from internal/dogfood and adjacent evidence, not interviews. | High |
Required Gate: Evidence And Completeness
Assumptions And Confidence Register
| Assumption | Status | Confidence | What Would Change It |
|---|---|---|---|
| The operator's dominant negative emotion is rediscovery anxiety rather than general disorganization. | Evidence-backed from ICP and JTBD, not interview-observed | Medium-high | Direct user interviews show the pain is mostly planning, collaboration, or UI preference instead. |
| Relief appears when the tracker shows source-linked branch state faster than native files. | Hypothesized from stated value drivers | Medium-high | Prototype tests show users still prefer raw files or agent summaries. |
| A wrong parser result destroys trust more than a missing feature. | Evidence-backed inference | High | Users tolerate minor parsing gaps if source links and error states are clear enough. |
| Write-back anxiety makes a read-only/read-mostly first experience emotionally safer. | Evidence-backed inference | Medium-high | Design partners demand editing as the first useful action and accept approval-gated mutation. |
| Multi-path repositories create the clearest emotional contrast. | Evidence-backed but untested | Medium-high | Single-path users report equally high state anxiety or multi-path users remain comfortable with native files. |
| Future collaborators, clients, or other agents will increase the value of explanation/handoff. | Hypothesized from ICP social job | Medium | Users remain solo and do not need to explain provenance outside their own sessions. |
Source Coverage Gaps
- No observed AFPS Tracker prototype sessions or emotional-reaction data.
- No direct AFPS user interviews confirming the named emotions, intensity, or sequence.
- No measured time or confidence delta between native reconstruction and a source-linked map.
- No parser-fidelity audit across multiple historical AFPS artifact versions.
- No evidence yet that external users will value handoff/provenance enough to pay.
- No separate Experience Map for `exec-loop-tracker`; it remains out of scope.
Proposed Canonical File Change After Approval
| Path | Action | Timing |
|---|---|---|
| `research/afps-tracker/journey-map-experience-map.md` | Create canonical Experience Map intermediate from this approved working packet. | Only after final compiled artifact approval YAML. |
| `research/afps-tracker/_working/preliminary-experience-map-research.md` | Archive then remove active working packet. | Only after final compiled artifact approval YAML. |
| `alignment/journey-map-afps-tracker.html` | Convert from framework findings review to confirmed/archive state for this framework handoff, preserving approval record. | Only after final compiled artifact approval YAML. |
Required Gate: Proposed Canonical File Change
Compile Responses
Answer the required gates above, or use section feedback for partial revision requests. When every required gate is answered with no unresolved negative feedback, the compiled YAML is ready for the parent `$journey-map` loop to consume.
No responses compiled yet.
Parent Routing Payload
agent_routing:
workflow: pattern-a-research-loop
parent_skill: journey-map
command: "$journey-map research/afps-tracker"
product_path: research/afps-tracker
gate_owner: parent-orchestrator
gate_type: framework-findings
framework_slug: experience-map
framework_mode: inline-subskill
run_manifest: research/afps-tracker/_working/journey-map-run.yaml
next_resolution: parent-resolves-from-yaml-and-filesystem