DIREKT Location and Addressing Research

Status: Open; no map provider or public location-precision decision approved.

Objective

Determine how customers and providers in the pilot context actually describe, share, verify and protect locations. DIREKT must improve discovery without publishing unsafe or falsely precise information.

Location concepts to separate

  1. Customer task location — where service is needed; private by default.
  2. Provider public area — broad discoverable area or service zone.
  3. Provider fixed premises — location customers may visit, subject to consent and verification.
  4. Provider private evidence location — coordinates/photos used for verification but not necessarily public.
  5. Provider service area — where a mobile/hybrid provider is willing to travel.
  6. Field-visit location — operational route data restricted to authorized staff.
  7. Search origin — customer’s current or manually selected area; minimize retention.

These fields must never be collapsed into one coordinate.

Research questions

  • Which address elements are commonly known and stable?
  • How are compounds, townships, roads, plots and landmarks used?
  • Do people trust map pins, live location or Plus Codes?
  • What happens when the pin points to the wrong entrance or centre of a property?
  • How often is a call required to complete directions?
  • How do mobile providers define coverage?
  • What public location detail is safe for home-based providers?
  • How do providers change premises or operate from shared locations?
  • What makes customers believe a provider has a real local presence?
  • What should a “premises visited” check prove and not prove?

Observation tasks

Customer task

Ask participants to describe how they would send a provider to a fictional or consented location. Record method and friction, not the coordinate.

Provider task

Ask providers to mark:

  • where they are based at broad level;
  • whether customers visit;
  • normal travel radius or named zones;
  • areas they avoid or charge more for;
  • how they explain location when maps fail.

Field-agent task

Simulate locating a consenting or fictional provider using the information a reviewer would receive. Record travel time band, calls, ambiguity and safety concerns.

Candidate public representations

Test with customers and providers:

  • neighbourhood/compound only;
  • approximate pin with privacy radius;
  • verified premises pin;
  • service-area polygon or named zones;
  • “travels to customers” without premises pin;
  • distance band rather than exact distance;
  • landmark-based text;
  • fixed/mobile/hybrid icon and explanation.

Accuracy states

Proposed states for research:

  • PROVIDER_DECLARED_AREA;
  • PIN_SUBMITTED_PRIVATE;
  • PREMISES_EVIDENCE_REVIEWED;
  • FIELD_VISIT_COMPLETED;
  • PUBLIC_PREMISES_APPROVED;
  • LOCATION_STALE;
  • RECHECK_REQUIRED;
  • LOCATION_DISPUTED.

These are research concepts, not approved database states.

Public-copy examples to test

  • “Serves customers in [named areas]. Provider supplied this service area.”
  • “Fixed premises privately confirmed by DIREKT on [date]. Exact evidence is not public.”
  • “Customers may visit this location; confirm opening hours before travelling.”
  • “Mobile provider: travels to customer locations.”
  • “Location check requires renewal.”
  • “A location check does not confirm workmanship, availability or ownership of the premises.”

Privacy rules

  • home address hidden by default;
  • customer exact location shared only for a justified interaction;
  • coordinates logged at no greater precision than needed;
  • field-route data restricted;
  • public map pins require explicit provider approval and evidence rules;
  • a premises visit must not expose family members or unrelated occupants;
  • photos must avoid faces, vehicle plates and neighbouring private property where possible;
  • location history must not become employee/agent surveillance.

Map-provider evaluation criteria

Phase 1A provides inputs; selection occurs only after cost and technical review.

Criterion Evidence needed
Coverage and POI quality in candidate areas field comparison
Geocoding of local place names test dataset
Pin and entrance accuracy observations
Offline/cache capability device tests
Android SDK size/performance prototype benchmark
Pricing and free-tier stability current provider terms
Data residency/privacy terms legal/security review
Usage restrictions and attribution legal/technical review
Reverse-geocoding quality field test
Vendor lock-in and abstraction feasibility architecture review

Staleness and correction

Research must define:

  • provider confirmation frequency;
  • customer stale-location reports;
  • relocation workflow;
  • evidence required for correction;
  • temporary closure;
  • disputed pin handling;
  • removal of public pin after suspension;
  • audit history without exposing prior private locations.

Exit outputs

  • location-method frequency by segment and area;
  • pin/direction task findings;
  • approved terminology for fixed, mobile and hybrid providers;
  • public/private precision recommendation;
  • field-visit and renewal recommendation;
  • map-provider test dataset using synthetic or consented locations;
  • unresolved legal/privacy questions;
  • updated design and architecture requirements.