Digital signage analytics

Digital signage analytics shows what you can report to an advertiser, what make-good you owe, and which screens to fix first. Every figure comes from player logs, counted per screen per slot and set against the schedule.

Last reviewed

Scheduled against played, example

94.7% of scheduled plays ran: offline screens caused 330 of 534 misses

Example data

Share of each day's scheduled slots played. A blank cell is 100%.

MonTueWedThuFriSatSunPlayed of 1,260
Screen 1100%100%100%100%100%100%93.3%, schedule error1,248
Screen 280.0%, content missing100%100%100%100%100%93.3%, schedule error1,212
Screen 3100%100%100%100%100%83.3%, screen offline93.3%, schedule error1,218
Screen 4100%80.0%, content missing100%100%100%100%93.3%, schedule error1,212
Screen 5100%100%100%100%100%100%93.3%, schedule error1,248
Screen 6100%100%33.3%, screen offline0%, screen offline100%100%93.3%, schedule error948, 75.2%
Screen 780.0%, content missing100%100%100%100%100%93.3%, schedule error1,212
Screen 8100%100%100%100%100%100%93.3%, schedule error1,248
Network1,3681,4041,3201,2601,4401,4101,3449,546 of 10,080, 94.7%
Offline all scheduled hours: 180 slots missed, 15 hours down
Schedule error: Sunday's schedule ended an hour early, 12 slots per screen
Screen 1Screen 2Screen 3Screen 4
Mon100%80.0%, content missing100%100%
Tue100%100%100%80.0%, content missing
Wed100%100%100%100%
Thu100%100%100%100%
Fri100%100%100%100%
Sat100%100%83.3%, screen offline100%
Sun93.3%, schedule error93.3%, schedule error93.3%, schedule error93.3%, schedule error
Played of 1,2601,2481,2121,2181,212
Screen 5Screen 6Screen 7Screen 8Network
Mon100%100%80.0%, content missing100%1,368
Tue100%100%100%100%1,404
Wed100%33.3%, screen offline100%100%1,320
Thu100%0%, screen offline100%100%1,260
Fri100%100%100%100%1,440
Sat100%100%100%100%1,410
Sun93.3%, schedule error93.3%, schedule error93.3%, schedule error93.3%, schedule error1,344
Played of 1,2601,2489481,2121,248

Network: 9,546 of 10,080, 94.7%

Offline all scheduled hours: 180 slots missed, 15 hours down Schedule error: Sunday's schedule ended an hour early, 12 slots per screen
Shortfall by cause
Slots missedShare of shortfall
Screen offline33061.8%
Content missing10820.2%
Schedule error9618.0%
Total534100%
Example data: an invented network of eight screens, scheduled 07:00 to 22:00 with 12 slots an hour, for the week of Monday 9 to Sunday 15 November 2026. Played slots are counted from player logs. Uptime in scheduled hours across the network was 96.7%.
Show the numbers
Slots played as a share of slots scheduled, example data. 180 slots per screen per day.
ScreenMonTueWedThuFriSatSunPlayed of 1,260
Screen 1100%100%100%100%100%100%93.3%1,248
Screen 280.0%100%100%100%100%100%93.3%1,212
Screen 3100%100%100%100%100%83.3%93.3%1,218
Screen 4100%80.0%100%100%100%100%93.3%1,212
Screen 5100%100%100%100%100%100%93.3%1,248
Screen 6100%100%33.3%0%100%100%93.3%948, 75.2%
Screen 780.0%100%100%100%100%100%93.3%1,212
Screen 8100%100%100%100%100%100%93.3%1,248
Network1,3681,4041,3201,2601,4401,4101,3449,546 of 10,080, 94.7%

Digital signage KPIs: the numbers that run a screen network

Delivery rate counts plays against the schedule and uptime counts scheduled hours only, so every figure in an advertiser's report leads back to log rows.

KPIFormulaA good readingThe common mistake
Delivery rateSlots played in full, from player logs ÷ slots scheduled, per screen per day Read per screen, with every missed slot given a cause Taking the schedule as the record of play. A screen that was off still has a full schedule.
Uptime in scheduled hoursScheduled hours the player was online and playing ÷ scheduled hours Read for each screen on its own, since a network figure can hide one screen that was down all week Dividing by 24 hours a day, which counts a screen switched off overnight by design as a fault
Time to recoverHours from the first missed heartbeat inside scheduled hours to the screen's next logged play, per fault Falling, and read by site and by cause Starting the clock when someone reported the fault, which leaves out the hours before anyone noticed
Plays per screen per daySlots played ÷ (live screens × days in the period) Steady by screen, with any drop matched to a cause in the delivery grid Dividing by every screen on the asset list, including ones installed and not yet live
Fill rateSlots booked by paying campaigns ÷ slots available for sale, by site and hour of day Read by site and hour, so the hours that sell out and the ones that don't are both visible Counting house content and filler loops as booked
Make-good owedPlays booked in a campaign's order − plays delivered on the booked screens and dates, from player logs Agreed in writing with the advertiser before the campaign report goes out Counting plays on screens outside the booking toward the total, which hides a shortfall on the screens that were sold

There are more examples in the KPI library.

A screen switched off overnight by design has not failed, so uptime is only ever measured in scheduled hours.

Where digital signage data usually lives

The content management systemPlaylists, schedules and campaigns, and the slot each was booked into. It records what should have played.
The player logsOne row per play, with the file, start time and whether it finished, and the only record of what did play. Players that upload late make a day's figures rise afterwards, so data strategy and infrastructure treats the logs like any late table, with a check per player that flags overdue uploads.
Device monitoringHeartbeats from each player, which give uptime and the start of each fault. Player clocks and the monitoring tool can run in different time zones, so every timestamp is converted to one zone before plays and faults are joined.
The booking systemCampaigns sold, the screens and dates booked, and the plays promised. If its screen IDs differ from the content management system's, one mapping table links them.
The audience sourceFootfall counts, traffic data or a survey, depending on the site. These estimates stay in a column of their own, with the source named.

Analytics services for screen networks

Proof of play is only as current as the last upload, so data monitoring and support checks each day that every player's logs arrived and names the screens that have gone quiet.

A play counts only when the player's own log shows it ran to the end.

AI in digital signage analytics

AI-led analytics answers "which screens missed the most plays last week, and why?" once every missed slot already carries a cause in the data, such as offline, content missing or schedule error. Each answer shows its query.

Questions about proof of play and uptime

What counts as proof of play?

A row in the player's log for each slot, with screen, content file, start time and whether it ran to the end. Only full plays count, so one that stopped part way shows as missed.

How should screen uptime be measured?

In scheduled hours only. A screen set to switch off from 22:00 to 07:00 isn't down overnight, so those hours drop out of both sides. A fault's clock runs from the first missed heartbeat inside scheduled hours to the next logged play.

Can audience figures go in the same report as plays?

Yes, in their own column, marked as estimates with the source named. Plays are counted from logs, and audience figures are modelled from footfall, traffic data or a survey, so we never multiply the two into unlabelled impressions.

Do you build network reporting dashboards for digital signage?

Yes. The core view is the screen by day grid above, with the shortfall split by cause, beside uptime in scheduled hours, time to recover and make-good owed per campaign.

What happens to plays logged after a screen comes back online?

They're dated to when they ran, so an earlier day's figures can rise after you've reported them. The report states when it was run, and a campaign's figures close only once every screen in it has uploaded past the campaign's end.

Could you defend your proof of play to an advertiser?

On the call we set a week of your schedules beside the player logs and show where the missed slots went.

Book a call

Free 30 min call