Paul Bannister
pbannist.bsky.social
Paul Bannister
@pbannist.bsky.social
Adtech, privacy, the web.
Net/net, content creators and marketers need to start adapting - now. Search trends are changing rapidly, and if you don't get ahead of the trends, you stand to lose ground quickly. 8/8
February 6, 2025 at 9:28 PM
Google has the most to lose right now as user behavior starts to shift. They're working hard on features like AI Overviews and new Gemini models, but you'd think Google would want allies, and it seems like they're just thumbing their nose at the market. 7/
February 6, 2025 at 9:28 PM
Google hasn't been doing anything. Aside from a deal with Reddit (which is different) and a side deal with AP, Google has been purposefully avoiding doing any AI-based deals with content creators. Considering the pressure that Google is under, this is odd. 6/
February 6, 2025 at 9:28 PM
OpenAI has taken a semi-constructive approach to working with publishers, signing deals that include compensation but, more importantly, guidelines around how citations in AI search would work. Perplexity has been doing similar deals. 5/
February 6, 2025 at 9:28 PM
Publishers and marketers take note - it's entirely possible this will become a material source of traffic within the next 12 months if ChatGPT can provide a credible alternative to Google. Figuring out how to optimize for generative search needs to be a key focus. 4/
February 6, 2025 at 9:28 PM
Since the prototype and release, we've seen significant growth in traffic from ChatGPT; it's already driving 4x as much traffic as Perplexity - which has been consistently growing over time. These are still very small numbers (<.1% of our traffic), but the growth is real. 3/
February 6, 2025 at 9:28 PM
This follows the prototype announcement last July and the release to registered users in October. This could be a big deal as it's OpenAI's first real stab at dethroning Google from search. 2/
February 6, 2025 at 9:28 PM
All of this will accelerate Google's "zero-click" search strategy as it will make no money from the web and spin more traffic back into itself.

So, if regulators are trying to create more competition via a Google sell-side divestiture, this approach seems sure to backfire. And Google wins. 10/10
January 7, 2025 at 5:39 PM
Then, DV360 will slowly stop buying any open web or app inventory. Non-O&O inventory is already less profitable today, and after a sell-side divestiture, it's even less so. This will obviously be bad for publishers and sell-side adtech firms. 9/
January 7, 2025 at 5:39 PM
Chrome will also give up Privacy Sandbox. They won't announce anything but will slowly stop building things. Sandbox has limited benefit to the O&O business of Google.

While Sandbox is flawed, publishers need all of tools they can get to combat signal loss, and losing Sandbox will hurt. 8/
January 7, 2025 at 5:39 PM
So if Google divests all of that, this allows the remainder of Google (the most profitable parts by far) to do several things.

First, Chrome will make the "cookie choice" popup very like Apple's ATT prompt. This is basically cookie deprecation. 7/
January 7, 2025 at 5:39 PM
The biggest thing Google gives up by selling its supply-side businesses is DV360 access to CTV inventory. But any divestiture would likely take years, giving DV360 plenty of time to build direct connections to the scaled CTV operators, which is a pretty small number. 6/
January 7, 2025 at 5:39 PM
Discover, AI Overviews, Quick View and other products that keep users inside Google and not on the web ("zero-click searches").

Google might also want to give up AdMob, which I doubt is hugely valuable to them anyway. Selling AdX, AdSense and AdMob would exit Google from the supply-side. 5/
January 7, 2025 at 5:39 PM
But a lot of AdSense inventory is junk. Google buys this inventory via tools like PMax to extract more revenue from advertisers. So if it loses AdSense, it loses this high-margin inventory. But it's already working to replace this on their O&O inventory via new products like... 4/
January 7, 2025 at 5:39 PM
Since DV360 is the primary way that big advertisers buy YouTube, Google doesn't want to lose control of that. I think AdSense is a likely candidate. The main strategic benefit Google gets from AdSense is access to scaled, cheap inventory. Some of that inventory is decent, 3/
January 7, 2025 at 5:39 PM
Google already offered to sell GAM/AdX to placate the European Union, but it was turned down. So if one of Google's goals with selling its adtech is to please regulators, it will need to offer more. But what? They're unlikely to want to sell DV360. 2/
January 7, 2025 at 5:39 PM
You snooze you lose! :D
November 27, 2024 at 6:35 PM
Love this. I should have gone by Paul John
November 27, 2024 at 6:35 PM
You going by Christopher now?
November 27, 2024 at 6:21 PM
Fair enough.. maybe I should register a domain.. Thanks!
November 20, 2024 at 2:35 PM
But how does pointing to a domain prove anything? Anyone could own paulbannister.com (I certainly don't own it).
Registrant WHOIS contact information verification | Namecheap.com
paulbannister.com
November 20, 2024 at 5:01 AM
I get that it's easy, but what's the value? I keep reading that it "authenticates" you, but who cares?
November 20, 2024 at 2:55 AM
Love this!
November 20, 2024 at 2:48 AM