I woke up today and Claude killed my startup. We got several hundred paying clients in 2 months, was growing like crazy. One Claude/Manus feature and our close rate dropped from 70% to 20%. Claude just made our entire product category obsolete.
We built an AI agent that automates ad management for you. Like, you can give it access to all your Google and Meta accounts and it will manage it for you. Pretty cool. Customers loved it.
Claude and Manus both released connectors for Meta Ads. Claude still can't make changes in ad accounts — it can only do analysis, and it has no access to Google Ads. But give it a few months, and it will. So building here feels pointless.
Like, every marketer will have Claude Cowork, ask "launch me a million-dollar ad campaign. make no mistakes" and go go go.
But what's next?
We need to think about the overall future of the GTM stack.
Here are the big GTM categories today:
- Own data — CRMs will be fine. Claude can't store your customer data.
- Lead databases — Clay, Apollo, RB2B will be fine. Claude can't collect 1B phone numbers and emails.
- Outreach automation — dead soon. You'll be able to tell Claude "Spam all YC founders offering my SaaS" and it'll just do it.
- Outreach infra — also dead. Agents will soon buy domains and set up outreach automation on their own.
- Niche tools — server-side tracking and similar will survive.
- Creatives — mostly dead. In a few months Meta and Google will generate them inside ad accounts. There will be some players who do top-tier large volume creatives for huge advertisers, but for small businesses it'll be inside Google/Meta.
- Enterprise complex workflows — like CDPs, customer engagement for huge brands. They are safe and will be doing exceedingly well. And waiting for more cool players in the space.
What's coming for marketing and distribution
AI made building free and distribution harder than ever. Everyone ships now. Good luck getting anyone to notice. Current GTM tech is pre-agentic era — it's not up to speed with where things are going. It can't sell to agents. But I'm a huge believer in the future of this space.
When building is free, distribution becomes everything — and that means more money, attention and energy will flow here than ever.
A few predictions:
- MCP is the new App Store. Claude picks a tool via MCP. No marketplace, no comparison — the AI picks one tool and the user never sees alternatives. This is the iPhone App Store in July 2008.
- Ads inside LLMs are the next trillion-dollar channel. Someone will build the ad network that connects thousands of chatbots the way Google connected websites. That company will be generational.
- Agentic commerce is coming. AI agents don't pay extra for brand. They compare specs, read reviews, pick the best value. Every category races toward commodity pricing.
- AI-to-AI sales will break enterprise playbooks. AI agents will read docs, compare specs, analyze pricing. They don't watch demos or get charmed.
- Social: completely botted and hijacked. The best GTM people know how to work the algorithms and get billions of views for free. It's a volume game once you figure out how to create decent, engaging content. Very soon — if not already — 98% of TikTok and Instagram will be fake AI content boosted by bots to go viral.
- Brand is more important than ever. It's almost impossible to replicate with AI. Last moat for selling to humans: story, trust, taste.
- The gap between good and bad marketers will explode. Anyone can generate AI slop for ~$0. It doesn't work. Taste, viral instincts, and storytelling are your alpha.
- Marketing is still a volume game. Try every channel, scale what works, do 10,000x of it. The tools die, but the hustle doesn't. We need human hustlers who can test everything and scale it.
What we're going to do next as a team
Our current business will be fine anyway. We knew what was coming and started actively pivoting a few weeks ago. Now we also build complex workflows for huge ad agencies managing 100s of accounts — like some of our clients manage 600 accounts with 6 people. Pretty insane.
We also sell AI-native ad agency services to SMBs because we know how to sell to small gas stations in Texas type of businesses, and they love how our AI and humans genuinely care about them.
But I definitely want to think more about the future of distribution in the post-Claude era and what cool and great things we can build here. Stay tuned.
