Instagram story viewer> @daypartai> Tagged

daypartai

@daypartai

undefined

POSTS STORIES REELS TAGGED

Lots of AI marketing bros on IG and X have entire threads on “optimizing your marketing workflow and 100x your ROAS.”

Daypart’s latest newsletter addresses that and gets into the liability questions around generative creative.

430

4

Are you running media spend against bots/crawlers/agents?

2k

22

I hopped on a podcast with @daypartai and @augustinefou (man’s got some food pics on his IG if you want to peep) and learned some of the RON inventory stuff vendors throw at us low key kinda intersect us with AI generated fraud. Fullpodcast is up on @daypartai’s linktree on their bio.

118

8

Another collab with @daypartai where they went and read the actual IAB vendor registry, which nobody does.

Google registered TCF Feature 3 to start using IP addresses for ad measurement across the EEA, UK and Switzerland. Registry wording is identify devices based on information transmitted automatically. I did not know until they pointed it out that a Feature carries no separate user toggle, it just rides in on whatever purpose you already accepted. The one that needs an explicit opt in is Special Feature 2. Passive gets waved through.

Three days later IAS pulled 800 domains for Papyrus, novel reading apps opening hidden browser windows to fake clicks and scrolls. That traffic came in about 13 percent above baseline on attention. The fake inventory graded better than the real inventory on the exact metric the industry bought in order to catch fake inventory.

I have sat on the buy side of a match rate conversation and I did not ask the follow up either.

Fingerprint slide is the one to send to whoever runs your CTV buy.

Long version is their newsletter this week, When The Moat Is A Sewer. Link is on their bio.

1k

18

AAK is being talked about like it landed last week. It shipped in 2024, got expanded at WWDC 2025, and SKAN keeps doing the job it has done since 2018. Crowd anonymity tiers unchanged, postback windows unchanged.

Re-engagement measurement did get better. That was the actual gap, so credit where it is due.

Slide 4 if your small campaigns keep coming back with nothing.

Slide 6 has the number worth arguing about. Adjust and AppsFlyer sit fifteen points apart on global ATT opt-in because they are measuring different things.

Nobody’s iOS measurement problem got solved by the rename.

Sources on slide 9. Full breakdown in this week’s newsletter, link in bio.

Memes with @digital_chadvertising

290

3

Agencies have started folding their AI costs into principal media deals. We broke down the trade: where it came from, what the ANA has said about it for a decade, and a $1M simulation where both agencies hit the client’s goal and one made $256K more doing it.

The round by round breakdown, the tokenmaxxing survey, and what to ask before anyone signs the free AI trade are in this week’s Diligence Report.

566

5

Do you remember Paypal’s Honey cookie stuffing drama that surfaced back in 2024? 👁️👄👁️

Affiliate marketing and last click attribution can set you up for some pain the second a browser extension can overwrite whoever really drove the click.

1k

28

This is gross and nobody likes this. Slide 3 you should share to everyone since Meta seemingly loves to nerf this story on people’s feeds. If you are private you have nothing to worry about but if you switch back to public that setting is always reverted back to on as a default.

2k

64