Digital insertion Rides Simulcast Wave


This being the week of the Cabletelevision Advertising Bureau’s annual gathering — during a month when more and more cable systems are gearing up for digital simulcasts — it’s probably a good time to check in on digital program insertion.

Digital program insertion, or DPI, is the stuff that splices local ads into digital video channels. It didn’t coincide with the launch of digital video because it was a tough sell to advertisers. (“You have how many customers watching this digital video product?”)

The last time this column delved into DPI, way more people were watching analog video than digital video. That had a predictable impact on how quickly advertisers didn’t beat a path to the local avails in the digital lineup.

That was two years ago, though. Now, there’s an accelerator: Digital simulcast, the sudden priority of 2005.


Across the nation, systems are gearing up to digitize their analog video tiers — partly to have a quantifiable “all digital” offering, partly to give HDTV owners a consistent viewing experience on all channels, partly to lay a strategic framework for related services. Like advertising.

Most MSOs that crank up simulcast will crank up the digital-into-digital advertising machine at the same time, or very soon after. That means local advertisements that can potentially slip into all channels, whether they’re being viewed in analog or digital; standard-definition or high-definition. The total available market really is everyone watching cable.

A brief recap of the DPI story follows.

It came into being because digital devices needed a way to hear the cue tones transmitted by program networks, when it’s time for a local ad to pop in on top of the Ginsu Knife ad.

Digital things generally can’t hear analog things. They live on bit rivers, not sine waves.

The trick with digital ad insertion was finding the right place to splice into, and out of, the river of bits.

Enter the Society of Cable Telecommunications Engineers, which in 2001 published two technical standards on the topic: ANSI/SCTE 30 and 35. In essence, they, as the reference documents for DPI, describe how to splice pieces of digital video into other pieces of digital video.

Back then, the growth potential for digital-into-digital advertising was only as fast as overall digital subscriber growth. That was before digital simulcast was even a topic amongst technology people, let alone a top-three priority.

Simulcast, at least in theory, accelerates digital subscribership – especially if it becomes the entry point for new video customers.


Before we get to the “next” in digital ad insertion, a quick look at the “now.”

Local advertisements are stored on digital file servers, same as video on demand. But when they’re switched out into the system, they’re carried in plain old RF analog. Meaning, they’re not crammed into a QAM (auadrature amplitude modulator), like VOD programs, because they’re not intended for digital ears.

(Analog receivers can’t hear digital any more than digital can hear analog.)

With the necessary “your mileage may vary” caveat, here’s how a typical cutover to DPI will go, in a system that’s simulcasting.

First, the DPI gear gets hooked up, in parallel with the analog gear. The gear is likely a module within the equipment that’s doing the bit organizing for all the simulcast (formerly analog) channels.

Tests happen. Is the splicer mechanism listening for cue messages on all ad-supported networks? Is it asking the file server for the right spots? Is the file server responding properly?

In some cases, there’s an additional twist: The conversion of a stored ad back to analog for insertion into the channels that will remain on the analog tier.

The twist is that the same ad is spliced once, into both the digital and analog networks.

Specifically, it works like this: Program network feeds enter the headend, where they’re encoded (squished) into a digital multiplex. At that time, the “digital” cue messages are spliced in, and the whole shebang plunges into the plant for the ride to homes.

Near the edge, the path forks, so that the channels that need to be delivered in analog can be converted back to analog — ads and all. So that gets tested.

Other implementations will spill one copy of an ad out of a digital spigot, and another copy of the same ad out of an analog spigot. And that gets tested.

No wrath matches that of a cable ad salesperson when a client’s ads aren’t running right. Once everything is working, it’s cut-over time. Out with the old, in with the new.


The intersection of DPI and digital simulcast puts a fresh timestamp on “the digital transition.” Another part of analog fades to digital — the equipment that listens for analog cue tones and switches stored ads into analog channels.

The most recent big milestone in ad-insertion technology happened about a decade ago, when those three-quarter inch professional-grade VCRs (video cassette recorders), used to store and play out local video ads, got bounced by digital file servers.

This is the next milestone. After the push to launch simulcast and DPI is a frenetic memory, there won’t be a need for that analog ad-insertion gear.

And all along the way, adjacent to all things touching the video food chain, the distant drumbeat grows nearer, and louder: Shrink analog. Grow digital.

Stumped by gibberish? Visit Leslie Ellis at