Shopping malls
Mall operators selling screen time in atriums, corridors and lift lobbies.
What it measuresAd attention analytics for digital signage
A small camera on your screen counts glances, viewers and engaged viewers, and maps where on the ad people looked. Everything is processed on the device. Only anonymous numbers reach your dashboard.
Glances
412
Viewers
236
Engaged
97
Sample data, for illustration
Why it matters
Proof of play
Your ad ran 1,240 times this week.
What most signage reports stop at. It says the screen was on, not that anyone looked.
Proof of attention
How many people glanced, how many viewed, how many stayed to read, and where on the ad they looked.
What advertisers are actually paying for.
Want to see your own screen's numbers?
Book an on-site demoHow it works
A small camera sits centred on the top or bottom edge of the screen, facing out. A Raspberry Pi plays your playlist on the same screen over HDMI.
For every face it works out distance, head direction and, up close, the eyes. If the line of sight meets the screen, that person is looking. Frames are wiped from memory straight after.
Only counts and gaze seconds per area are sent. You see them per ad in the dashboard and export a PDF report.
We do the setup for you, on your site, in about 15 minutes.
Book an on-site demoThree levels of attention
A person counts as looking when their line of sight meets the screen, worked out from where their head is and which way it points. Someone at the side who turns to the screen counts. Someone in front who looks past it does not.
Counted when: Their line of sight touched the screen, even for a moment while walking past.
Your true reach: how many people the screen caught, not how many walked by.
Counted when: They looked for at least half a second.
People who actually saw the ad, not just the screen.
Counted when: They kept looking for 2 seconds or more.
The audience that read your message. The number creative teams optimise for.
What you measure
Three levels of attention, counted in 10-second windows. Each level includes the next, so you see where the audience drops off.
How long people looked, per ad and per window. Glances are kept together across blinks and missed frames.
Where on the ad people looked, drawn over the ad itself on a 32 × 18 grid.
Draw boxes for the logo, product, price or button. You get gaze time for each, even for areas drawn after the data was collected.
Time on screen, plays, viewers and engaged viewers for every ad in the playlist, optionally per device.
Estimated age band and gender per ad, only where your privacy notice covers it. Groups under 5 people are hidden.
Where people looked
Gaze time per area
Areas are drawn on the ad in the dashboard. Draw one later and it is estimated from the stored heat map.
See the heat map for your own ad, live.
Book an on-site demoPrivacy first
Frames are analysed in memory and wiped straight after.
Nothing image-like is written to disk or sent.
No face recognition, no identities, no tracking across visits.
The core guarantees are enforced by automated tests in the product.
Use cases
Mall operators selling screen time in atriums, corridors and lift lobbies.
What it measuresRetailers running in-store screens at entrances, aisles and checkouts.
What it measuresSelf-service kiosks and freestanding totems with an attract loop.
What it measuresQuick-service restaurants and cafés with digital menu boards.
What it measuresNetworks selling in-store screen inventory to brands.
What it measuresWe bring the device to your site, mount it in about 15 minutes and show you live results on the dashboard.
Book an on-site demo