Section 03Data source readiness

An honest scorecard of what we can measure today.

Six data sources evaluated on coverage, integrity, refresh cadence, and the decisions each can credibly support. The point is not to grade systems — it's to be precise about which numbers we trust today, and which need investment before they earn the right to drive a decision.

Source assessment summary

As of Q1 2026
SourcePurposeCoverageIntegrityRefreshAPIStatus
OpenBlueAuthoritative space inventory, types, seat assignments98%92%Real-time✓READY
Occupancy sensorsActual presence at seat / room level71%78%5-minute intervals✓PARTIAL
Badge / presence dataBuilding & floor entry, headcount by day100%95%Real-time✓READY
Workforce / HR rosterHeadcount, team, manager, work mode100%88%Daily✓READY
Reservation system (Condeco)Booked rooms, desks, parking85%62%Real-time✓PARTIAL
Wi-Fi associationDevice-level presence, dwell time95%55%Real-time—NOT READY
§ 03.1 — Detail by source

Where each source can — and can't — be trusted today.

OpenBlue
READY

Authoritative space inventory, types, seat assignments

Coverage
98% of expected scope
Data integrity
92% (sampling-based estimate)
Known gaps & caveats
Neighborhood metadata inconsistent across legacy floors. Seat type taxonomy needs standardization.
Decisions this source can support
Space allocationCapacity planningSpace type mix
Occupancy sensors
PARTIAL

Actual presence at seat / room level

Coverage
71% of expected scope
Data integrity
78% (sampling-based estimate)
Known gaps & caveats
Coverage gaps on legacy floors (retrofit pending). Calibration drift in 4% of meeting rooms.
Decisions this source can support
Real utilizationPeak demandSpace type performance
Badge / presence data
READY

Building & floor entry, headcount by day

Coverage
100% of expected scope
Data integrity
95% (sampling-based estimate)
Known gaps & caveats
Tailgating not detected — overstates unique attendance ~3-5%.
Decisions this source can support
Hybrid attendance patternsBuilding-level demandDOW trends
Workforce / HR roster
READY

Headcount, team, manager, work mode

Coverage
100% of expected scope
Data integrity
88% (sampling-based estimate)
Known gaps & caveats
Work-mode field self-reported and stale for ~12% of population.
Decisions this source can support
Demand forecastingTeam allocationWork mode policy
Reservation system (Condeco)
PARTIAL

Booked rooms, desks, parking

Coverage
85% of expected scope
Data integrity
62% (sampling-based estimate)
Known gaps & caveats
High no-show rate inflates booked-vs-used gap. Inconsistent team adoption.
Decisions this source can support
Booking patternsNo-show analysis
Wi-Fi association
NOT READY

Device-level presence, dwell time

Coverage
95% of expected scope
Data integrity
55% (sampling-based estimate)
Known gaps & caveats
Privacy review pending. Multi-device users distort headcount.
Decisions this source can support
Future: dwell time analysis
§ 03.2 — Implications for the framework

Three of six sources are decision-ready today: OpenBlue, badge feed, and HR roster. These three combined are sufficient to deliver Phase 1 of the framework — seat supply, peak utilization, capacity risk, and allocation reconciliation. Sensors and reservations are partial — usable for directional space type performance but not yet for binding decisions. Wi-Fi association data is not ready and should not be relied on until the privacy and de-duplication issues are resolved.

The practical sequence: ship Phase 1 metrics on the three ready sources, layer sensor data into Phase 2 once coverage hits 90%+, defer Wi-Fi to Phase 3 contingent on privacy review.