Editable draft edit inline, then copy back to Aref
ELBERTA LABS

Appendix A: Pilot Plan and Acceptance Scorecard

Prepared forGallery Condominiums Board
ByElberta Labs
Date25 August 2026
Part ofManaged Camera Intelligence pilot package

This appendix sets out how the pilot runs and how it will be judged. The aim is a decision a board can make on evidence, not impression. Every measure below has a baseline, a target, and a method, and each is agreed before launch.

Scope in detail

Included in the pilotCoordination onlyFuture roadmap
Selected existing camera feeds and IntuVision alerts, connected into one reception queue with AI-assisted prioritization, staff training, and reporting. Camera, network, and access-control vendors, engaged when their work affects the pilot. Elberta does not hire or pay these vendors. Access-door event analytics and broader consolidation across the building's systems.

Access control appears only as coordination and future roadmap. It is not a pilot deliverable.

How the pilot runs

PhaseWhat happens
Weeks 1 to 2Baseline measurement and shadow mode. AI recommendations are recorded but do not change the alerts staff act on.
Weeks 3 to 4Supervised prioritization and reception training. The queue goes live under review.
Weeks 5 to 7Measured live operation against the scorecard.
Week 8Results, capacity findings, handover package, and a board go/no-go report.

Acceptance scorecard

Sign-off is by the board or its designee, on the Week 8 report. Where a result is inconclusive, the measurement window is extended within the pilot term at no additional fee before a go/no-go call is made.

MeasureBaselineTargetMethod
Alert volume per camera-day at the front deskMeasured in weeks 1 to 2At least 50% lower after prioritizationQueue counts, before and after
Downgrades judged correct by a human reviewerNot applicableAt least 95% of a sampled setWeekly human review of sampled downgrades
Concerning events wrongly downgradedNot applicableZero tolerance; any miss triggers a tuning reviewHuman review plus incident check
Staffed shifts using the dashboardMeasured at launchAt least 80%Dashboard session logs
Median acknowledgment time by severityMeasured in weeks 1 to 2Improves against baselineQueue acknowledgment timestamps
System availability during staffed hoursNot applicableAt least 95%Health monitoring uptime
Maximum camera loadNot applicable12 cameras at 1080p and about 5 frames per secondLoad test on the pilot appliance
Staff feedback after trainingNot applicableNet positiveStructured survey

Initial proof-of-concept observations

Before the pilot, Elberta ran a short proof of concept at no charge on the Peachtree pedestrian gate to confirm that AI screening works on the existing cameras without per-camera license fees. Over a four-day sample the camera produced about 40 alerts, roughly 10 per day, most of them routine gate traffic and maintenance activity. AI screening separated the routine from the worth-a-look without discarding anything.

These are illustrative observations from one camera over a few days. They are not a general conclusion. The 60-day pilot, measured against the scorecard above, is the actual validation.