Post-Trip Feedback Survey (Lightweight Mobile Flow)¶

Post-trip feedback survey mobile screen with trip rating, petal activity ratings, discovery prompt, and final actions Figure 8. Post-Trip Feedback Survey collecting overall trip rating, per‑petal feedback, discovery input, and follow-up actions to enrich personalization + Popular Petals ranking signals.

1. Purpose¶

Capture high-signal qualitative + structured feedback immediately after trip completion (or upon user returning to the app) while cognitive context is fresh—fueling: personalization vector updates, Popular Petals scoring, quality moderation, and future trip suggestions.

2. Entry Triggers¶

Trigger	Condition	Debounce
Auto prompt	Trip end date passed + user opens app (≤ 48h)	1 per trip
Manual	User taps “Give Feedback” on trip detail	None
Reminder	No response after 24h (push/email optional)	Max 1 reminder

3. Survey Sections¶

Section	Elements	Notes
Header	Trip title, supportive subtext	Dynamic: “weekend near {Region}”
Overall Rating	Petal (1–5) or emoji scale	Accepts skip (stores null)
Recommend Toggle	Yes / No	Only shown if rating ≥ 3 or user opens details
Activity (Per Petal)	Mini card: name, rating 1–4, optional note	Collapsible after rating chosen
Discovery Prompt	Open text + autosuggest petals	Suggest list filters out already in library
Final Actions	Submit • Save Trip (if not) • Plan Follow-Up	Follow-Up launches generation seed using feedback deltas

4. Data Model (Submission Payload)¶

{
  "tripId": "forest_lake_weekend_2025_09_13",
  "overall": { "rating": 4, "recommend": true },
  "activities": [
    {"segmentId": "seg_1", "petalId": "baneasa_forest", "rating": 4, "note": "Shady & cool."},
    {"segmentId": "seg_2", "petalId": "snagov_lake", "rating": 3, "note": "Windy midday."}
  ],
  "discoveries": ["hidden_boardwalk", "local_bakery"],
  "freeformDiscovery": "Found a small lakeside kiosk with great lemonade.",
  "clientMeta": {"device": "mobile", "lang": "en"}
}

Normalization¶

discoveries list items resolved to Petal IDs via fuzzy match or created as candidate petals (pending moderation).

5. Signal Derivation¶

Signal	Source	Uses
petal_quality_score	Per activity rating	Ranking suppression / boost
trip_satisfaction	Overall rating	Personalization weight shift
recommend_flag	Recommend toggle	Social proof metrics
discovery_yield	Count of new discoveries	User explorer persona weighting
friction_flag	Notes containing patterns ("muddy", "closed")	Data freshness / POI status review

6. Telemetry Events¶

Event	Payload Example
feedback_view	{tripId, activities:7}
feedback_rate_trip	{tripId, rating:4}
feedback_rate_petal	{tripId, petalId:"baneasa_forest", rating:3}
feedback_discovery_add	{tripId, candidate:"hidden_boardwalk"}
feedback_submit	{tripId, overall:4, activitiesRated:6}
feedback_followup_generate	{tripId, seed:"follow_up"}

7. Ranking & Personalization Impact¶

Component	Update Logic
Popular Petals	Increment weighted counters (recent ratings decay exponentially)
User Vector	Fine-tune interest dimensions (e.g., water, forest, cultural) based on high-rated petals
Negative Signals	Down-weight low-rated categories in next generation candidate pool
Cold Start Aid	Aggregated anonymous averages inform baseline priors

8. Privacy & Compliance¶

Freeform notes passed through mild PII scrub (emails, phone numbers masked).
Discovery submissions flagged for moderation before surfacing publicly.
User can delete feedback (endpoint DELETE /api/trips/{id}/feedback).

9. UX Interaction Details¶

Pattern	Description
Progressive Disclosure	Only first 3 activities expanded; others lazy-expand on scroll
Skip Friendly	“Skip for now” returns to trip view; keeps reminder eligibility
Micro Animations	Petal rating selection pulses ±5% scale
Optimistic Submit	UI disables controls, shows inline success check ✔

10. Accessibility¶

Ratings group uses radiogroup semantics with clear aria-labels.
Notes text areas support dictation / screen reader hints.
Focus order: header → overall rating → per-activity groups → discovery prompt → actions.

11. Error Handling¶

Scenario	Handling
Network fail on submit	Queue payload locally (IndexedDB) retry w/ backoff
Partial ratings only	Accept; missing ratings omitted (null stored)
Discovery unmatched	Create tentative candidate (status=`pending`)
Duplicate submit	Idempotency key hash of (tripId + userId) prevents double ingest

12. Future Enhancements¶

Sentiment analysis on notes → adaptive clarification prompt.
Inline photo attachment for activity evidence.
“Auto plan follow-up” wizard seeding delta preferences (e.g., more water activities next time).

Figure 8 added (image pending upload as trip-feedback-survey.png).

Cross-links: trip-in-progress.md, itinerary-refinement-detail.md, navigation-picker.md