How HubSpot Uses Claude to Achieve 40% Productivity Gains Across Teams
HubSpot is an agentic customer platform serving hundreds of thousands of companies worldwide, from startups to enterprises, with tools for marketing, sales, and customer service. The company deployed Claude and Claude Code across engineering, marketing, and customer success teams after benchmarking multiple AI tools and finding Claude required the least human intervention on real engineering tasks. The result was a 40% productivity increase across web development and content creation, and a drop in complex troubleshooting cycles from days to under an hour.
Impact
Up to 40%
Productivity increase across web development and content creation
3–5 days → under 1 hour
Time to troubleshoot customer escalation workflows
Challenge
HubSpot’s engineers, marketers, and customer success managers were losing hours to manual work—large-scale migrations, content production, and personalized outreach—that couldn’t scale with a distributed codebase and a global customer base.
Solution
HubSpot deployed Claude and Claude Code across teams, using Claude Code’s MCP support to connect directly to internal infrastructure, and Claude projects to centralize campaign and customer context for marketing and customer success.
Tools & Technologies
What Leaders Say
“When Claude Code came out with Model Context Protocol (MCP) first-class citizen support, we were excited to give it a go. It did a superior job compared to some of the other tools we were developing or trying when it comes to solving real problems that HubSpotters were tasked with on a day-to-day basis.”
“One of the things we love about Claude is that it has really good taste, and marketing’s all about taste. The Claude models have always been very strong at writing, whether it be writing code or writing prose.”
“Claude gives me the support and time to be a more strategic partner for my customers. As a result, they’re now telling me our conversations are more meaningful than ever.”
Sign up to read complete case studies, access detailed metrics, and unlock all use cases.
Full Story
HubSpot operates at a scale where internal productivity directly influences its ability to serve customers. With hundreds of thousands of customers globally and a distributed codebase spanning thousands of services, the company needed AI that could operate reliably inside complex, real-world engineering environments—not just perform well on generic benchmarks. The challenge extended beyond engineering: marketing teams faced an endless content pipeline, and customer success managers struggled to personalize at scale.
The pressure points were tangible. Engineers spent significant hours navigating large-scale migrations and routine maintenance that added no direct customer value. A CSM creating personalized follow-up for a customer could spend an hour on content she could only justify for her highest-priority accounts. On the marketing side, maintaining consistent context across campaigns while scaling output was a constant tension. The team wanted AI that could integrate with their existing infrastructure rather than require them to rebuild around it.
HubSpot evaluated multiple AI tools before selecting Claude as a preferred solution. The deciding factor for engineering was Claude Code’s native support for the Model Context Protocol (MCP), which allowed the team to connect Claude directly to HubSpot’s existing internal services. The team ran internal benchmarks using real HubSpot engineering tasks; Claude consistently reached finished solutions with less human intervention than alternatives. For marketing and customer success, Claude projects provided a shared context layer: once one person built a strong project with curated instructions, adoption spread organically across teams.
The outcomes were measurable across functions. Marketing reported a 40% productivity increase across web development and content creation. Customer success managers cut complex troubleshooting cycles from 3 to 5 days to under an hour when using Claude directly. During HubSpot’s 2025 rebranding, engineering teams used Claude Code to accelerate a frontend migration that would otherwise have taken months. New engineers shipped meaningful production code faster than comparable hires had in previous years.
Looking forward, HubSpot sees Claude models as central to how businesses will operate. The internal deployment has become a proof point for HubSpot’s own platform thesis: that AI should unify customer context and make growth easier. As the engineering lead noted, the shift is increasingly about creativity and system design rather than execution—Claude handles more of the routine, freeing engineers to focus on architecture and user experience.
Amazon Bedrock