AI reshapes marketing: traffic plummets, humans curate
The gist
AI has shattered marketing’s old playbook, slashing web traffic and turning marketers from content machines into brand curators.
What to know
- HubSpot lost 140 million visits in a year—traffic is down 50% since its peak—forcing a complete overhaul of their strategy.
- By late 2026, AI was running up to 95% of marketing execution, but humans still call the shots on judgment, creative direction, and brand fit.
- Anthropic estimates 65% of marketing work is replaceable by AI, but marketers are now upstream: interpreting research, curating assets, and keeping AI-generated slop in check.
HubSpot’s AI Reckoning
AI didn’t just disrupt HubSpot’s traffic—it forced the company to rebuild its entire marketing playbook and admit the old system was obsolete.
By June 2026, HubSpot was already saying AI had broken the old marketing machine. On Marketing Against the Grain, the company described a complete change in how it ran HubSpot’s marketing team for a post-AI world, and by late August Kipp Bodner made the disruption concrete on Duct Tape Marketing: HubSpot had lost 140 million visits over the course of a year. He said that between the rise of AI and Google’s algorithm updates, traffic dropped off quickly, forcing HubSpot to admit the prior workflow no longer held.
Between August and October, HubSpot and adjacent voices recast AI as an operating-system change, not a tool upgrade. Bodner said traffic was down 50% from its high and that much of it was garbage traffic, while noting that AI could run a remarkable research project on an ideal customer profile in 10 minutes instead of 10 days and that deep research reports for brand awareness came back within 10% of internal work. By September, HubSpot had launched new AI products and positioned Loop as the codification of its experiments and of how AI had changed what a marketer could do, a framework that had appeared in an internal memo about 18 months earlier.
Humans Define, AI Delivers
As AI floods marketing with content, human judgment now decides what makes the cut and keeps brands from drowning in algorithmic noise.
By mid-2026, the split in marketing labor was clear: AI could generate abundance, but humans had to decide what survived. The AI Marketing Companion said AI can give you 10 copies of something, and you will know which one is right for your brand and the moment, while Canva’s State of Marketing and AI 2026 found 78% of consumers would rather see ads made by people, even if AI could make them better.
That shift moved marketers upstream into strategy, governance, and measurement. Nir Weingarten’s 10 rule said AI has a lot of muscle but does not know what to do, leaving the last stretch to human curation, while The Agile Brand said strategists will still be there in five years to manage unintended consequences; Mark Ritson cited Anthropic’s estimate that about 65% of marketing work is replaceable, yet said we still need marketers to test 400 synthetic-data plans, interpret faster research, and prevent AI slop through centralized context and review, as a CMO survey that cost you 2 million bucks and 6 months to do was returned tomorrow morning.
