Bristol, Stokes Croft.
§ CASE STUDY · TEMPLATE · MID-MARKET VENUES
Ticket touts flagged at the gates. Multi-camera reach across foyer, smoking area, side exit.
An independent music venue in Bristol's Stokes Croft, 400-cap, with a door team of two and a security director who also runs box office on the night. The pattern that keeps coming back is ticket touting — the same handful of resold-ticket sellers turning up gig after gig, working the queue before doors. The template engagement is shaped to flag the pattern at the gates the second the seller reappears, and to give the security director cross-camera reach without expanding the door team.
§ ENGAGEMENT SNAPSHOT
What the template engagement covers, at a glance.
-
Vendor type
Independent mid-market music venue (1k-cap-and-below cohort)
-
Site covered
1 venue — Bristol Stokes Croft
-
Capacity
400
-
Camera estate
~18 cameras (gates + foyer + smoking area + side exit + stage-right)
-
Duty model
2-person door team + 1 security director + box-office volunteer
-
ICO posture
ZA-class controller; Music Venue Trust DPIA template; embedding-only ReID
-
Outcome headline (template)
Repeat ticket touts surfaced at the gates the second time they show up.
§ WHAT THE ENGAGEMENT COVERED
Four work-streams. One posture.
- 01
Embedding-only ReID across the gate, foyer, smoking area and side exit
Cameras at the main gate, the foyer, the smoking-area gate and the side exit all stream to the edge node. Each detected person is reduced to a vector; the cross-camera trace across the four zones surfaces in the security director's pane as a thumbnail strip — same vector, different zones, different gig nights.
- 02
Ticket-tout pattern flag — same vector, multiple gig nights
When the same embedding vector is seen on the gate-area camera across three or more gig nights without ever entering the venue (or always exiting after a short stay), the system surfaces the pattern. The security director sees a single row — 'this vector has been at the gate on five Friday nights, never gone in past the foyer' — and can talk to the seller directly the next time they show up.
- 03
Music Venue Trust DPIA posture
We arrive with the DPIA scaffold pre-filled against the MVT (Music Venue Trust) template. The security director (acting as DPO) stamps the legitimate-interest balancing test, 14-day vector expiry, no facial-recognition store. The venue's regulars get a one-paragraph posters-only summary at the door.
- 04
Punter-facing posture: no face image, no name, no ticket-record link
The system never crosses into the box-office system. The vector is not tied to a punter name, a ticket-purchase record, or a regulars list. Door signage names what the system does ('cross-camera matching by clothing and gait at the gates') and what it does not do. The punter experience is unchanged.
§ HONEST FRAMING — PENDING PILOT
What this page is and is not.
- This page describes a template engagement. There is no Bristol venue signed to it today.
- We have had pilot conversations with one Stokes-Croft-shaped independent and one Manchester equivalent; neither has signed a paid pilot.
- Camera counts and tout-detection figures above are sized to the realistic shape of a 400-cap independent — they are not measured.
- When a real pilot lands, this URL will be replaced with the named-customer case study and a CHANGELOG note will record the swap.
§ WHAT WE WILL PUBLISH WHEN REAL
Four commitments. The shape of the real case study.
When this template is swapped for the real customer, these are the four sections the new page will carry. The point is to be measured against them now — buyers can read what we will publish before we have published it.
- 01
The named venue + the MVT-template DPIA
We will publish the venue name (with consent), the security director's name, and the Music-Venue-Trust-template-stamped DPIA in redacted PDF.
- 02
The measured tout-pattern outcome
We will publish the count of repeat ticket-tout patterns flagged in the first six gig nights, the false-positive rate (the security director's judgement of 'flagged but actually a confused regular'), and the count of touts confirmed.
- 03
The cross-camera reach outcome
We will publish the count of incidents the security director surfaced cross-camera with the system in place vs. the previous 90 days. We will publish whether the door team's effective coverage grew without adding headcount.
- 04
The honest miss-list
We will publish what the system missed in the first six gig nights that the door team did catch by eye. We will not publish a success-only template.
Read the rest of the stack.
Read the mid-market venues brief → All templates → Schedule a 20-min call →