From Experimental Bot to 70% Autonomous Resolution
In just twelve months, HubSpot’s Customer Agent has jumped from resolving 20% of support conversations on its own to 70%, with some customers already surpassing 85% and even 90% autonomous resolution rates. That trajectory signals a step-change in what AI customer support agents can handle in live environments. HubSpot reports that Customer Agent now serves more than 9,000 customers and accounts for over half of all AI credits used on its platform, outpacing other agents focused on prospecting and data tasks. The dominant use cases align closely with contact center automation priorities: always-on, after-hours coverage and tier-one ticket handling, so human teams can focus on more complex issues. HubSpot’s leadership emphasizes that this is not a ceiling. As underlying models improve, they expect Customer Agent to move from basic queries into higher-level support, expanding from chat into email and other channels with steadily rising autonomous resolution rates.
What 70% Autonomy Means for Contact Center Operations
An autonomous resolution rate of 70% effectively turns AI customer support agents into a digital front line, taking the bulk of repetitive queries off human queues. For contact centers, this changes staffing and workflow assumptions: instead of sizing teams around peak inbound volumes, leaders can design operations where AI handles routine questions at scale while humans intervene only for complex, emotionally sensitive, or high-risk cases. After-hours and weekend coverage, once a major cost and scheduling challenge, can be augmented by AI with minimal incremental overhead. Tier-one agents become escalation specialists and coaches for AI, refining knowledge bases and reviewing edge cases rather than answering the same basic questions all day. As AI learns from interaction data, the autonomous resolution rate can climb further, compressing handle times and reducing backlogs. The result is a structurally leaner support operation that can absorb growth without linearly adding headcount.
Amplitude’s Bet: AI Agents as Data-Driven Growth Engines
While HubSpot showcases what happens inside the contact center, Amplitude is positioning itself as the observability and instrumentation layer for AI agents and digital products more broadly. Its platform unifies analytics, experimentation, session replay, guides, surveys, and web analytics so companies can see how users behave and iterate quickly. The acquisition of Statsig assets extends Amplitude’s reach into data warehouse-centric experimentation and feature flagging, important for AI-focused companies that store data in platforms like Databricks or Snowflake. Amplitude’s CFO describes AI agents as software entities that must be measured like any other application: how they respond, where they fail, and which interactions lead to better outcomes. Foundational agents in Amplitude’s platform allow teams to query data conversationally, while more specialized agents can automate insights. This positions customer service AI not just as a cost reducer, but as a growth engine, feeding product and marketing teams with behavioral intelligence.
Reducing Human Load While Elevating Human Work
As AI customer support agents absorb a majority of tickets, companies gain an opportunity to redesign human roles rather than simply reduce headcount. HubSpot’s customers are already using Customer Agent to clear tier-one tickets so human agents can focus on complex resolutions that require negotiation, empathy, or deep product knowledge. This shift can improve both customer experience and employee satisfaction: customers get faster answers to simple questions, and human agents spend more time on cases where they can genuinely add value. In parallel, Amplitude’s strategy highlights how AI agents themselves become a rich source of behavioral data. Support leaders can mine interaction logs to refine FAQs, detect product friction, and inform roadmap decisions. Over time, human expertise moves upstream into training, quality oversight, and customer success, while contact center automation handles volume at scale—turning support from a pure cost center into a strategic feedback and loyalty engine.