Vercel replaces 96% of marketing, 93% of support with AI agents

Vercel replaces 96% of marketing, 93% of support with AI agents

Yesterday
SaaStr ARMARM Gtm_strategy

The Gist

  • Vercel's AI agents handle 96% of marketing and 93% of support workflows
  • Entire SDR team reabsorbed into other functions after agent deployment
  • CPO built hundreds of agents internally before productizing them
Key Quotes

The 7% that gets past the agent is a curated list of real product gaps and misconfigurations. The agent isn’t just clearing tickets, it’s filtering signal from noise so humans only see the problems worth solving.

Tom Occhino

Documentation is the bottleneck, not the model. The hard part of building DealOne wasn’t the AI. It was being able to write down exactly how the work should be done.

Tom Occhino
Key Insights
  • Vercel replaced 96% of marketing and 93% of support tasks with AI agents, showcasing significant operational efficiency gains.
  • Every company will need at least two agents: one for employees (internal workflows) and one for customers (support, transactions, etc.).
  • The 'build vs buy' decision shifts to buying foundational infrastructure and tools while building the differentiated layer of agents.
  • Documentation is the critical bottleneck for deploying agents, not the AI models themselves.
  • Agents flip the traditional software model by making the UI a lightweight surface while the core functionality runs autonomously.
  • The fastest ROI for AI agents comes from automating highly repetitive, low-value tasks rather than high-value problems.
Actionable Takeaways
  • Start building AI agents by automating low-value, repetitive tasks to achieve quick ROI and learn agent behavior.
  • Invest in documenting internal processes thoroughly, as this is the foundation for effective AI agent deployment.
  • Focus on buying foundational infrastructure (SDKs, gateways, sandboxes) rather than pre-built agents, and build the differentiated layer on top.
  • Prioritize reliability infrastructure (durable workflows, retries, fallbacks) for agents, as these are critical for production trust.
Data Points
  • 96% (Percentage of marketing tasks replaced by AI agents at Vercel.)
  • 93% (Percentage of support tasks replaced by AI agents at Vercel.)
  • 7% (Percentage of support issues that bypass the AI agent, representing curated high-value problems.)

RevBots.ai View:

ARM leaders eat their own dog food by automating internal workflows before selling the tools externally.

Full Story: SaaStr →