Case study · Broadcast & streaming

Every ad break on 52 live channels, checked around the clock.

Server-side ad insertion fails quietly. No error, no alarm, just lost revenue. We built the monitor that catches it for a major Latin American pay-TV operator.

52
live channels watched continuously
2
streaming protocols covered (DASH + HLS)
Every break
classified: commercial, promo, or slate
Minutes
from failure to alert, not days
Client
Latin American pay-TV operator
What we built
Ad-break fleet monitor
Watches
52 linear channels, 24/7
Alerts via
Live dashboard, Slack, hourly digest
Status
In production, managed
The problem

Ad breaks that fail without failing.

The operator runs 52 live linear channels with advertising stitched into every break server-side. When a break goes wrong, nothing errors. The stream keeps playing.

The failure is invisible unless someone is watching that exact channel at that exact moment. The only defence was manual spot-checking, which can never cover 52 channels at once.

×A break fills with slates instead of paid commercials, so revenue is lost and nobody is notified
×A break fills with the operator's own promos when paid ads should have run
×A break over-fills and cuts back into programming late
×Tracking beacons go missing: the ad played, but nobody can prove it
What we built

A monitor that watches every break on every channel.

Continuous, protocol-aware, and tuned to the four ways an ad break actually goes wrong.

Full-fleet coverage

All 52 channels monitored continuously on both DASH and HLS, the two ways the streams actually reach viewers.

Every break classified

Each ad break is analysed and labelled as commercial, promo, or slate, so fill quality is measured rather than assumed.

The four failure modes, detected

No-fill, promo-only fill, over-fill, and missing tracking beacons each raise their own targeted alert.

Alerts where the team lives

A live dashboard for the operations floor, Slack alerts for immediate action, and an hourly digest with downloadable reports for the record.

The outcome

From spot checks to certainty.

Before

  • Manual spot checks on a handful of channels, when someone had time
  • Failed breaks discovered days later, or never
  • No record of fill quality across the fleet
  • Ad revenue leakage invisible until reconciliation

After

  • Every break on all 52 channels classified, around the clock
  • Failures alerted in minutes via Slack and the live dashboard
  • Hourly digests and downloadable reports as a permanent record
  • Fill problems caught while there's still time to fix the next break
Running in production

Going live was day one. It's been looked after every day since.

Streaming platforms change under you: encoder updates, ad server behaviour, new channels. We run this monitor as a managed service, so when the platform shifts, the monitoring shifts with it.

StatusIn production
MonitoringWatched around the clock
Maintained byThe team that built it

Tell us about the process that's eating your week.

A 30 minute scoping call. No pitch deck, no obligation. Just an honest answer on whether we can fix it.

Book a Scoping Call