#8 - Are You Paying for Ads Nobody Watches?

Probably, but there is a better way to compare the effectiveness of ad types.

Last week, we saw how campaigns calculate vote goal and why audience targeting matters. The next step is putting budget behind channels that actually deliver quality views to the right voters.

So how should campaigns compare mediums?

We’ve talked about where to reach voters. But there’s a second metric we obsess over at Cross Screen Media when allocating budgets.

eCPM - effective cost per 1,000 on-target viewable 5-second views

Why is eCPM important?

Voters consume video content across Linear TV, Streaming, Desktop/Mobile, and Social.

The problem is these channels transact on different currencies, offer different targeting capabilities, and define a “view” differently.

That makes budget decisions an apples-to-oranges exercise.

When we started CSM, we chose to set aside channel-specific metrics and define a single standard to compare all video inventory. Alongside reach, eCPM governs how we build media plans. As the market shifted and new channels emerged, we reallocated dollars quickly using real, comparable data.

How do we define an eCPM?

The cost to reach your audience with 1,000 viewable 5-second views.

Let’s break that down:

Most media is transacted on a CPM, the cost to serve 1,000 impressions, with an impression defined as an ad served, no matter how long someone watched the ad.

Why 5 seconds?

Most digital ads aren’t actually watched.

85% of digital video ads are viewed for less than 2.5 seconds. Industry standards didn’t help reduce the noise. MRC set the bar at 2 seconds, Facebook used 3, and Google used multiple definitions.

We raised the bar to 5 seconds and applied it consistently across all digital video. One standard, one metric.

For Linear TV, we assume ads are watched to completion. No bathroom breaks.

How does this impact video ad channels?

When we looked at the Nielsen Gauge, we saw how apps varied in time spent based on ad-supported time and political policies.

The same is true when you look at the eCPM of different video channels.

Let’s take a look at the average CPM, targeting capabilities, and 5-second viewability %.

These are directional averages to illustrate how eCPM behaves, not exact benchmarks. We’ll assume if a channel allows audience targeting that the on-target percentage is 100%, we’ll see later that is not always the case.

How does this impact eCPM?

Remember, we still have to transact in CPMs on digital and Points on Linear.

Let’s assume, we have an audience that is 20% of the adult population in a geography. Much larger than the 9% of the population that are persuadable voters.

In this example, the most expensive inventory by CPM is the most efficient by eCPM.

Every buyer can make some of their own calls here. YouTube doesn't allow for precise audience targeting, but maybe you think their contextual segments are great and that should raise the on-target percentage. Or maybe you think Mobile/Desktop has a lesser impact than the big screen.

There's room for judgment, but it should be rooted in a clear foundation for comparison.

What does this mean for political advertisers?