TIL podcast advertising shifted *away* from programmatic last year

Not what I expected

There’s a very common story people in the podcasting industry tell themselves. Some are excited about it. Others hate it. But pretty much everyone thinks it’s happening. The story goes like this:

Right now we live in the golden age of podcast advertising, because hosts actually read the ads, which makes them way less annoying than normal advertisements are. The problem is, that model doesn’t work for programmatically inserted ads. And programatic is the future, because brands want to be able to target specific types of people, test different creative, and spend exactly as much as they want to spend. So as the technology of podcasting matures, the world is going to move towards programatic.

Turns out, the world has been headed in the opposite direction:

(This is a slide from the IAB’s annual podcast revenue report, which came out earlier this week.)

Why did it happen?

Nobody seems to know. I can only find one mention of this slide on twitter or anywhere else on the internet, and it comes from Nick Quah, publisher of Hot Pod:

Weirdly, the share of automatically inserted ads — i.e. dynamically inserted ads — as a delivery mechanism dropped from 56.4 percent to 41.7 percent between 2016 and 2017. I’m personally surprised by this, and will do some digging around on the issue.

My theory

I think there’s a couple things going on:

  1. The vast majority of “dynamically inserted ads” aren’t all that dynamic to begin with. You still have the host reading the ad, and it’s still intimately tied to a show. It’s not like a TV ad where it’s totally separate from the content. The only reason for using dynamic insertion is so the podcast network can keep tighter control over how many downloads they deliver so they’re not over (or under) serving their customers. A secondary reason is to make more efficient use of the back-catalog episodes.

  2. From the above we can infer that there is only marginal benefit of using dynamic insertion technology, so it’s important to also consider the cost. You need someone to manage the dynamic insertion system, and many newer podcasters (even with very big audiences!) may not want to deal with that. (For example, I know one of the largest daily news podcasts does not use dynamic insertion.)

  3. Another important thing to keep in mind: the share of “branded content” revenue in podcasting rose from 1.5% of the market in 2016 to 6.5% in 2017. It would make no sense to dynamically insert an ad there, so some of the shrinkage of programatic can be partially explained by that.

Honestly though, it’s a pretty interesting mystery! I would never have guessed that dynamic ad insertion was shrinking.

Going forward, I think the main driver of the technology will be if any programatic ad network (like Megaphone Targeted Marketplace) starts to take off. With that, the benefit goes from being marginal (help podcasters serve their customers in a more organized way) to existential (target specific demographics of users). Certainly lots of companies are trying to make it happen.

And many hope it never will:

nick q@nwquah

the dude @pkafka articulates a thing I’ve been trying to say for a while but better

“You may well see podcast advertising move much faster — but it also means podcast advertising will be more unpleasant. Let’s hope the podcast guys keep puttering along.”https://t.co/LSmr9iVwIH

June 11, 2018

It’ll be interesting to follow!

(PS - if you have any information / theories as to why dynamic ad insertion shrank, let me know!)