Named client
US Open (tennis) · PMY Group
Measured every US Open since 2021
- 220,621 unique devices at the 2025 edition
- 36% engaged · 39% returning · 28 min average engaged dwell
- Recurring since 2021 · delivered via PMY Group
More than 1,400 events supported directly
We measure human behaviour in your physical space.
Which zones held your visitors, which were walked past, and what to change before the next event — answered zone by zone, day by day, from passive WiFi sensors that anonymise everything on the sensor itself. At Palfinger’s two-floor Las Vegas booth: 33,464 unique devices measured across five days.
- No app, no tags
- No cameras
- Anonymised on-sensor
- 1–3 s
- Live since
- 2017
How it works
Sensors in, numbers out, nothing identifiable in between.
Place the sensors
AP-class, listen-only hardware on existing structures — a 220V outlet is the whole requirement. Each sensor hears phones within ~10 m; three or more seeing the same device gives approximate location.
Anonymise on the device
The sensor hashes each device identifier with a random salt in 1–3 seconds. Raw identifiers are never stored — on the sensor, in transit, or on the server. The salt is destroyed when your event closes.
Read the dashboard
Zone visits resolve into faraway, passer-by and engaged (≥ 30 s dwell, stated). Daily cleaning removes staff and fixtures; then the final daily report publishes. Counts are unique devices, not people.
What you receive
The flow explorer: every zone-to-zone path, weighted by preference.
A production dashboard view. Each line is a path visitors actually took between zones; thickness is how many took it.
- Which entrance feeds which zones — and which zones feed nothing
- Where to move a product, a desk or a person to intercept real traffic
- Whether this year’s layout outperformed last year’s, on identical definitions
In production
Measured at the events you’ve heard of.
Named where clients have agreed; anonymised proofs where they haven’t. Every figure below was read from the production dashboard.
Named client
Disney · Mirrored Media
One campaign, two activations, one comparable read
- Convention 2024: 164,586 unique devices · 54% engaged · 7 sensors
- VidCon booth: 13,475 unique devices · 43% engaged · 45 min dwell
- Cross-activation measurement in one campaign · via Mirrored Media
Anonymised
A European motorsport event
Race-weekend crowd measurement at circuit scale
- Recurring race-weekend deployment
- Same-day daily reporting to the operations team
- Client under confidentiality — shared as an anonymised proof
Products
Three ways to read a space.
Short events · roadshows · long events · permanent installsPassive WiFi analyticsFootfall, dwell, engagement and flow across your zones, from passive listen-only sensors — no app, no tags, no cameras, no venue infrastructure.Conventions · venues · long events · permanent installsCapacity monitoringLive occupancy against your limits: how many devices are inside each area right now, with alerts when zones approach capacity.Short events · activations · venuesInteractive captive portalOpt-in visitor registration over event WiFi: QR to portal to double opt-in, with per-person visit journeys for registered users only.
Reports
One deployment. Answers for every team.
The same sensor layer answers the exhibitor’s question, the organiser’s question and the venue’s question — as separate reports, each sold to the team that needs it. Three samples are public.
Partners
Agencies run it under their own brand.
Partners and agencies deliver 100% of our transient-event work. The dashboard whitelabels, and EYEframe embeds reports in your own portal in about a day and a half of integration — agency partners include PMY Group and Mirrored Media.
Start small
A baseline deployment first. Expansion only if the numbers earn it.
The recommended path: a 1–2 month fixed-cost baseline deployment, then expand to the areas the data says matter. You hold the hardware and activate it when you measure.
Sensor hardware€250per sensor, one-time — yours to keep and redeploy
Active days€80per sensor per day while measuring
Minimum€350per month, in active months
4G/LTE data traffic included. Standard list, individually negotiated. Permanent installations are priced per site.
A worked example
A 10-sensor booth at a 4-day trade show: hardware 10 × €250 = €2,500 one-time — the sensors are yours to keep — plus activation 10 × 4 × €80 = €3,200. First deployment about €5,700; the next event on the same kit, about €3,200. In months you don’t measure, you pay nothing; in active months the €350 minimum applies. 4G included; standard list, individually negotiated.
Straight answers
The questions buyers actually ask.
Are your counts people counts?
No, and we will not pretend otherwise. Counts are unique devices, not people. Phones with WiFi switched off are not seen; some visitors carry two devices; modern phones randomise their identifiers, which our statistical model accounts for. We do not apply a hidden multiplier to turn devices into “visitors” — you get the device numbers and the definitions, and the comparison that matters is like-for-like across your own zones, days and editions.
Can you identify or track an individual person?
The passive pipeline is built so that we do not. Each device is assigned a pseudonymous identity, irreversibly de-linked at campaign end: the salt that links a device to its hash is destroyed when your event closes (or at campaign end by agreement), and destruction irreversibly eliminates linkability in production systems. No report cell ever represents fewer than 12 devices — in production since 2019 — and outputs are aggregate statistics only: no profiles are created, stored, or delivered, and no re-marketing use.
How accurate is it?
We publish one calibration point rather than a blanket percentage: at a major international Grand Prix, our weekend figure of 289,661 unique devices compared with a box office of 301,837 tickets — within about 4%, with the caveats stated in full in our case studies (devices are not people; tickets include no-shows; agreement is venue-dependent). Any accuracy figure without its method and bounds is marketing, not measurement.
Do I get the raw data?
Yes, as standard. The anonymised, consolidated data behind your reports is exportable — it is your deployment. Anonymous aggregates and delivered reports are retained for 12 months.
Scope a deployment
Tell us about the space you want to understand.
We reply with a scoped proposal: zones, sensor count, and the reports you’d get. Or write to info@advancedradiomapping.com.