TTD OpenPath vs non-OpenPath. What actually performs better? The Programmatic 101 newsletter is back (programmatic101.beehiiv.com/…), and the first issue just went out. I wanted an actual answer so I finally sat down and looked properly...👇

If TTD ads are blocked in Safari, that's a big deal for every buyer running on it. And we're learning about it from AdExchanger, not from TTD. Not a great look.

THIS

Is OpenPath actually cheaper, or is it just a discount on worse inventory? I've been checking that against real data instead of taking either side's word for it, and depending which number you look at, you get two completely different answers. On price, OpenPath is doing a genuinely good job. Across 90 days of data, OpenPath impressions ran about 43% cheaper than the same inventory bought through named SSPs, and when I checked the actual attention environment those impressions land in, ad refresh rate, ad density, viewability, competing ads in view, it came back basically identical to what I'm paying a real premium for elsewhere. That reads less like a junk-inventory discount and more like the market correcting a price that named-SSP supply may have been charging without the environment to back it up. Then the platform numbers tell a different story. CPA ran about 31% higher on OpenPath and ROAS came in about 28% lower, and it holds up even above the fold and on desktop, the exact segments where OpenPath should look most like named-SSP supply if this were just a placement mix issue. So is the price win real and the performance gap the actual cost of getting it, or is something else entirely eating the difference? I'd rather stay careful than confident here. These are last touch, post view attributed conversions, which means the model credits whichever impression happened to sit closest to the conversion and doesn't prove that impression caused anything, so if OpenPath impressions tend to land earlier in someone's path, or the attribution model is just structurally less generous to that supply route, a real chunk of that CPA gap could be measurement rather than performance. I don't think that's the whole story though, the gap survives across two separate 90-day windows and holds even in segments where placement shouldn't matter, but I'm not calling it solved until somebody actually runs an incrementality test on it. So here's what I'd actually do with this. Don't exclude OpenPath publishers on the assumption they're low quality, the attention data doesn't support that at all, and you're probably cutting a fairly priced route for a reason that isn't true. Do stop trusting a blended CPA or ROAS number if a meaningful share of your spend runs through OpenPath, break it out before you conclude anything from it. Full breakdown is in the first issue of Programmatic 101 on behiiv (link in my profile bio and on the thumbnail), the fold and device cuts that rule out placement mix, all four attention metrics and what they actually showed, and the attribution caveat in more detail than I could fit here.

interesting

The CPM is 43% cheaper on average. Attention quality held up too, which honestly surprised me. And the performance data surprised me even more, in a direction I wasn't expecting going in. In the first issue I break down: - Why the CPM gap holds up and where it's biggest - What the attention data actually showed, and why it wasn't what I expected - The CPA and ROAS numbers, and a possible explanation for why they might not mean what they look like - What I'd personally do differently for bottom-of-funnel campaigns until someone runs a real incrementality test LINK - programmatic101.beehiiv.com/…

The thing is that AI slop can have pretty strong metrics (viewability, ad density, etc.), which makes it harder to detect. Legacy verification tools are also not very good at identifying it.

definitely a good move in the right direction, but it still doesn’t give you much control over other properties. The algorithm still decides what percentage of the budget is allocated between Search, YouTube, Gmail, and so on

Not a lot of people talking about this. Really great work by Haus team. The finding that stuck with me: noisy experiments left teams worse off than doing nothing 38% of the time.

❓what's your experience with ChatGPT ads? Especially compared to search? From what I see, so far it looks like a good consideration channel, but definitely not google search replacement/direct competitor.

really interesting read by @Adweek & @KendraEBarnett especially "standard IP targeting accuracy is so poor that just 23% of residential IP addresses reached their intended geographic target, per the study. Meanwhile, CTV identifiers, which, like mobile ad IDs are device-level, were 24% more consistent over time than IP addresses"

Ok so everyone is back. What’s the hottest topic in adtech right now?

Lets goooo adtech x!!!