Why you're getting this: Why you're getting this: You've interacted with Haven or AdTribe before, and this is my weekly-ish update on my journey to building an AI-native agency. Zero pressure to stick around, just click unsubscribe if you don't want to get it.
Alright, let’s get into it.
NYC trip
Got back from New York last week. And like every time, my first thought: "Should I move here?"
I arrived the day the Knicks won the championship. The whole city went bananas. Had to walk 5 blocks dragging my carry-on because all the roads were closed. First I was annoyed. Then I realized these people probably haven't felt this way in decades. Who am I to complain.
By the next morning, people were already selling $45 championship t-shirts printed overnight. The city doesn't sleep, but it also doesn't waste time.
Besides the Knicks chaos, the FIFA crowds, and some shopping I'm not proud of, it was a productive trip:
Dinner with founders organized by Romans and Kristina
Coworking and a round of golf with one of our agency partners who's become a close friend
Way too much coffee and negronis with people I've been wanting to sit down with for months
My bigger takeaway from the trip: when everything online eventually gets done by AI, in-person relationships will be the only moat left. This city is where those happen. Every visit I have more meetings than I can count. Existing clients, new intros, people I've only known through a screen finally sitting across from me.
If you're ambitious and want to build something real, I don't think there's a better place in the world.
Apify + Claude
We just rolled out an upgrade to how we do competitor research.
Most agencies read 10-20 reviews and call it a day. Or dump a product into ChatGPT and ask for "pain points."
We now scrape every review we can find. Amazon. Trustpilot. Walmart. Target. Reddit. All of it. Then Claude analyzes and categorizes thousands of reviews in minutes.
What you get back:
Real competitor reviews with source links
An angle map (competitor complaint → your client's counter-angle → ad hook)
Quick example: we ran this for a water filter client competing against Brita. The number one complaint theme was mold. Everywhere, every platform.
Each complaint becomes its own ad concept, backed by real customer language.
We're rolling this out to all partner accounts over the next few weeks. If you're curious how it'd look for a specific brand, just reply.
AI-native media buyers
Last week I got on a call with one of the biggest performance agencies in the world. Anyone who's been running ads for a few years knows who they are. When I was starting out in media buying, I used to listen to their podcast religiously.
Their head of HR found me on LinkedIn. They're scaling in Europe and need media buyers.
But not just media buyers. AI-native media buyers.
We talked for a while and I realized: that person basically doesn't exist yet. The market is so far behind on this. You can find great media buyers. You can find people who use AI tools here and there. Finding someone who's genuinely built their entire workflow around AI is a different thing.
What AI-native media buyer actually means in practice (in my opinion):
Reports run from Claude, not pulled manually every morning
Creative research and briefs done with AI, not from scratch
Custom skills built to automate repetitive account work
Alerts set up to catch performance drops before they spiral
Time spent on strategy, not navigating the account
We started pushing this hard at the beginning of the year. Most of our media buyers are now deep in Claude Code. We might be ahead of the market. Or maybe we're just early to something everyone will have to figure out eventually.
Either way, if you're a media buyer reading this: the ones who figure this out now are going to be worth a lot more in 12-18 months.
Talk soon,
Dmitry