What is the open web, anyway?

Today’s the last IAB Tech Lab installment with the OMSDK for mobile measurement. And we take on the question of “what is the open web, anyway?”

IAB Tech Lab: The OMSDK

In this final installment of the IAB Tach Lab interviews, we deep dive on the OMSDK, the bridge that allows measurement companies access to in-app environments without custom SDKs.

Due to the generosity of the IAB Tech Lab, these interviews will remain outside the paywall for all users.

Podcast: Kerel Cooper from Group Black

Kerel Cooper comes on the pod and gives us fascinating insights about how brands and CMOs are reaching the African-American audience in the US. From programmatic to whisper campaigns in barbershops, you will learn a lot!

Listen to the pod now:

• Spotify

• Apple

• Anywhere you listen

What is the open web anyway?

I’ve been writing a bunch on the challenges of the “open web”, and how closed systems like Meta have huge advantages for marketers. For a good discussion of this I would recommend my conversation on the MobileDevMemo podcast with Eric Seufert where we really dive into these issues and where I refer to the “logged in web”, which is a term I think I made up on the spot. And this week, here came The Trade Desk with their “premium internet” branding.

I’m literally the last person to get hung up on semantics or to sit on a panel and say “well first, we need to define our terms.” This type of thing makes me want to vomit. But I’ve been fielding a lot of queries that go something like “well, is Netflix part of the open web given they are going to support programmatic buying?” Or “aren’t most of TTD’s premium publishers actually in CTV?” So I feel like we need a framework to talk about this stuff.

There are two competing and overlapping definitions of “open”:

• Open to consumers to view content without a login or payment;

• Open to advertisers to buy ads programmatically using the tools of their choice.

These get confused because the traditional understanding of the “open web” was open for both, and the traditional understanding of the “walled gardens” was closed on both. But the lines have blurred.

With that, dear reader, let me humbly propose the following 2×2 matrix I call The Internet Advertising Map:

We’ve mapped the internet advertising world into four (or five) clusters of businesses, starting on the lower left and going counter-clockwise:

• The “Open Internet” is where most journalism, blogs, apps, and other content lives. These businesses are under the most pressure from declining referral traffic and signal loss. We don’t talk about this enough, but the majority of free mobile apps are also in this segment. The strategic goal is to move to the right into…

• The “Logged-In Internet” is where the consumers are logged-in or pay for content. The Trade Desk calls it the Premium Internet with their own methodology based on many signals, but I think just logging in to get content is a great way to segment these publishers from the open internet as presumably no one logs in for non-premium content. It is true that the majority of CTV companies are in this segment — good for them!

• “Walled Gardens” is a term we all throw around, but should be a bit more strictly formed, around both a majority of known, registered consumers, and a closed ad ecosystem. This is the elite club, where you have power over demand and supply and you exploit it.

• “Native Content” is comprised of media properties that allow consumers free access and monetize with native (primarily text) ads. These companies encourage logins and payments, but get enough value out of anonymous traffic to monetize well. Much of Commerce Media (including Criteo’s commerce products) are in this segment as is Google search. Interestingly, this segment is in the most danger from the growth of AI and decline in search referral traffic.

• Bonus segment: You may notice there’s a cluster of companies sitting between Native and Walled Gardens. This reflects the push-and-pull dynamic whereby companies would prefer to be Walled Gardens but either make so much money on non-logged in traffic (YouTube, commerce media) or don’t have the clout to force everyone to use their platform (Pinterest).

Is this framework useful?

Creating a framework like this is only useful if it also explains other phenomenon. Like a scientific theory, it should have predictive value, so I’ve outlined a number of questions below and shown how this framework applies.

First, from the buyer’s perspective, how do the quadrants align with the ways buyers buy media? This maps very cleanly between vendor APIs (i.e. the Meta API) vs programmatic, and further divisions between open exchange vs PG, etc.

How do these publishers generate and sustain audiences? There’s another clear pattern where the left quadrants are much more search and referral dependent, while the right side has more control over their destiny (and their consumers).

Strategically, what is each quadrant’s greatest challenge?

Is the Open Internet in decline?

Coming back to how this conversation started, the answer is definitely yes, the Open Internet is in trouble. Defined as properties that are mostly anonymous traffic and that allow broad access from advertisers, the prognosis isn’t great.

This lower-left quadrant has a litany of ails:

• Too much undifferentiated supply

• Loss of signal (cookies going away, ATT, etc)

• Competition from MFA and low-quality sites and placements

• Reduction in referral traffic from social

• Reduction in referral traffic from search given AI

• Difficulty collecting email addresses (and Apple isn’t helping)

In a sense, The Trade Desk is right to focus on “The Premium Internet” as it gives advertisers signals to target, avoids most low quality placements, and is much less dependent for traffic on the internet giants. But that definition is arbitrary and controlled by one company, so we should be more clear.

The challenge for Open Internet publishers is to move to the lower-right corner. The only way to do that is to build relationships with consumers and get those emails!

Reading list

• Dr. Mark cancels his Uber and goes to PayPal.

• TTD ranks the top 100 pubs for the “premium internet”. Here’s the PDF with the rankings.

• Klarna cuts marketing expense with AI

• Perion closes ContentIQ a week after AdExchanger exposes them as an MFA farm via Sincera

• Google’s search API algo is leaked and many public statements by Google are contradicted

• Adam Heimlich’s Google anti-trust video. From the Civil War to the Sandbox.

• Worst Memorial day box office ever

• Auren Hoffman on the lessons of the LiveRamp acquisition