Kurt Bullock

What is working for me on Meta right now: 1. Cutting audience network entirely has resulted in a substantial increase across the board. I laid out when I do this vs leaving on in a tweet last week. Hat tip @rokhladnik 2. Incremental Attribution - when you have a robust social presence, massive retention audience, or have a considerably longer consideration cycle. IA has proven substantially incremental in our Haus tests on the brands that meet this definition. 3. Consolidation - no more than 4-5 total campaigns running sales objective. Incremental reach struggles beyond this in my experience 4. Scaling campaign with taking winning creative and driving to a different LP to prevent overlap. Usually do one ad set per landing page and running either flex ads or post ID. 5. Creating a daily breakdown in Manus looking at the breakdown effect of your ad spend and ads to understand how each ad is playing in the funnel. I use CPMr as the main metric for this. 6. Partnership ads for each stage of funnel mixed in with main scaling campaign. 7. Launching whitelisted DPAs from creator accounts (hat tip @wearetheselect) and it's pretty awesome out of the gate. 8. Similar to 4, if you're launching ONLY PDP/Collection traffic you're CPMr is going to be high, your traffic is going to struggle, launch LPs, listicles, advertorials etc. You need a split set of traffic now 9. 5-10% of ad spend running sales based campaigns on daily unique reach to boost overall new reach using your top performing ads in a sequence. 10. Bid caps combined with a lowest cost testing framework. I like doing this to play both sides of the coin. It's actually incremental and also helps you know when to scale lowest cost based on bid cap spending.

I'm presently sitting in a cafe in my wife's hometown of Jeonju, South Korea drinking a latte and doing a bit of a gut check on Meta / Media Buying and the state of DTC in May of 2026. Right now it's brutal. There is no rhyme or reason as to why something is working / not working. I've got an inbox full of brands asking for audits and help. Historically post Mother's Day always tracks pretty low, but this year has been one for the record books. We will always say, is it economic issues, the media buyer, or is Meta (biggest share of wallet) just making massive updates? The reality, none of us truly know. Which makes things always that much harder to diagnose. I've been in advertising now for 18 years. The last calendar year has been the most I've ever been uprooted in my professional career. If 2021-2022 was the year of iOS14, 2025-2026 is the year of what I call "the AI guinea pig era." I'm not necessarily complaining, it was going to happen. But I just didn't realize how much all of this would impact sooooooo many brands. From AI slop to AI bots, AI has infiltrated every facet of your business for the good and bad. Even some of the most stable, highest spending accounts I have access to are seeing broad issues, mostly related to issues on Meta. I write this quickly to tell you once again, you're not alone. I think we can all find some peace in knowing this. That doesn't mean being complacent. There are plenty of things to try, test, and do in your businesses. However, for the remainder of May until Meta (again biggest share of wallet) launches their new AI pixel, new Meta workflow and slows down their "innovation" some I find it hard to spend MORE unless we absolutely see bright spots. Again, it's not all doom and gloom, some accounts are a bit unfazed because of strong organic signals, but for the majority it's rough. For most accounts we're seeing downward trends in the days pre-Mother's Day. The only thing that has helped stabilize has been reducing spend. That's it. Again your businesses are all different, but I'm not going to gaslight you in saying things are good, because the reality is, they aren't. Heads down though, figuring it out one ad at a time, one campaign at a time and one channel at a time. This shall too pass.

Why I'm more bullish on incremental attribution on Meta... 1. It's been the biggest way to increase new reach % incrementally in ad accounts at scale that struggle with this. 2. Revenue per visit is between 2-3x higher than standard attribution in most ad accounts I have it running. 3. New customer orders are nearly 100%. The leakage and cross over is much much lower. 4. They are less prone to performance swings. Once the system is optimized (50 conversions) it acts like a growth engine and fatigues less than standard attribution (in my ad accounts). So how do I run these? 1. In creative testing - all ads go into IA now. Why? Since Andromeda was implemented we've noticed less and less "new" traffic going to these pre-heated ads. In fact, it dropped down near 20-30% only. This obviously creates a problem, I don't know if these creatives are scalable. And the reality was, most weren't. So switching to IA helped identify more and more ads that work. BIG win here. 2. I am a massive fan of Flex Ads. A few reasons, Meta's ad retrieval systems love simplistic. Mainly because it makes it easier to digest and identify winning assets. So the more account consolidation on the ad level (where most of the work is being done) are now spread across less ads with more opportunities for Meta to find new winners. 3. IA + multiple traffic sources + flex ads = the scaling machine. We know Meta uses 50% of the creative and 50% of the LP in the ad account. Yet, most of us only optimize one of these two things. Become as diversified on your traffic sources as you are on your creative for larger wins. A few caveats: 1. Please don't switch your entire ad accounts to IA. IA works for me and a lot of my brands, but it doesn't work for all. What I've found is the following: a. If your audience has a TON of retention - test IA b. If you have an AOV above $80-90-100 - test IA c. If you have a long consideration window - test IA d. If new reach is ur probem - test IA For many ad accounts I'm still all in on 7/1/1 and 7DC via standard attribution. But for many, IA is truly the growth engine needed because it's not putting the window of time to act in a short, well, window. It's instead acting as a likely much more realisitic way that consumers on Meta are acting. I really hope other channels adopt this system next.