HubSpot\u00a0prove its\u00a0AI agents are successfully moving\u00a0into live customer-service workflows, with the company pitching its CRM as the control layer for deployment. \u00a0<\/p>\n
Its\u00a0Customer Agent results, rapid Data Agent uptake, and broader agent activity suggest businesses are beginning to embed AI across the customer journey. \u00a0<\/p>\n
For CX leaders, the priority is to pursue this integration without losing sight of reliability, governance and proof that automation improves the customer outcome.\u00a0<\/p>\n
Yamini Rangan, CEO of HubSpot, said businesses are seeking a unified, governed approach to AI as they deploy agents across customer-facing services.\u00a0<\/p>\n
\u201cCustomers don\u2019t want a chaotic agent sprawl,\u201d she said.\u00a0<\/p>\n
\u201cThey want a controlled, cohesive way to build demand, win deals, and delight their customers.\u201d<\/p>\n
Customer Agent Enhances Customer Service<\/p>\n
HubSpot\u2019s AI products have demonstrated\u00a0measurable service outcomes in Q2, providing a clear indicator that AI agents are performing successfully\u00a0in live customer service environments.\u00a0<\/p>\n
In fact, \u201cCustomer Agent now resolves 72% of support tickets without human escalation,\u201d Rangan announced.\u00a0<\/p>\n
While many vendors continue to focus on productivity gains or the number of AI conversations handled, HubSpot is assessing whether customers actually receive complete answers without requiring a human agent.\u00a0<\/p>\n
In one customer example, Sesame HR was previously responding to only 70% of incoming support tickets.\u00a0<\/p>\n
After deployment however, \u201cThey cover all tickets received with 60% fully resolved without human escalation,\u201d she explained.\u00a0<\/p>\n
By implementing AI to meaningfully resolve customer issues at scale, this enabled Sesame HR to keep pace with demand without straining\u00a0its workforce capacity, encouraging the software provider to purchase more than one million AI credits to expand into additional\u00a0HubSpot agents.\u00a0<\/p>\n
As a result\u00a0of its ability to deliver reliable outcomes, customer adoption has continued to grow, with Customer Agent now having surpassed 10,000 customers.\u00a0<\/p>\n
Rangan notes:\u00a0<\/p>\n
\u201cThe outcomes, the quality of outcomes is what will enable more small, medium businesses to adopt AI.\u201d<\/p>\n
Data Agent Increases Customer Count By 80%<\/p>\n
HubSpot\u2019s Data Agent reveals\u00a0that\u00a0many organizations are choosing to begin their AI journey\u00a0by\u00a0applying\u00a0it\u00a0to their internal CRM data, using it to identify target accounts, build audiences, and\u00a0uncover total addressable markets. \u00a0<\/p>\n
\u201cData Agent has over 16,000 customers activated, up 80% quarter-over-quarter,\u201d\u00a0Rangan revealed.\u00a0<\/p>\n
As a result, this agent has become one of the\u00a0company\u2019s fastest-growing AI products,\u00a0suggesting that organizations see lower-risk, internal AI use cases as the natural starting point for adoption.\u00a0<\/p>\n
\u201cCustomers start with internal facing agent, which means that Data Agent, the reason you see that very high adoption of Data Agent is because they\u2019re comfortable using an internal use case,\u201d she said. \u00a0<\/p>\n
By improving the usefulness of CRM data first, businesses can establish confidence in AI-generated outputs before expanding those capabilities into customer-facing\u00a0services.\u00a0<\/p>\n
Furthermore,\u00a0\u201cCustomers aren\u2019t adopting a single AI use case,\u201d Rangan explained.\u00a0<\/p>\n
\u201cThey\u2019re using HubSpot agents across the entire customer journey.\u201d\u00a0<\/p>\n
With\u00a0AI credit consumption\u00a0now evenly distributed across\u00a0HubSpot\u2019s agents\u00a0and buyer intent capabilities,\u00a0this means\u00a0adoption is becoming more balanced across departments.\u00a0<\/p>\n
In fact, more than 55% of Pro Plus customers now use either AI agents or Breeze Assistant, while monthly agentic actions across its customer base have\u00a0more than tripled\u00a0this year. \u00a0<\/p>\n
HubSpot\u2019s Breeze Assistant\u2019s weekly active usage has doubled since the start of the year, and more than 2,700 customers have already activated Agent Builder and Agent Hub.\u00a0<\/p>\n
However, HubSpot\u2019s Data Agent 16,000\u00a0activation figures\u00a0do not represent\u00a0proof of success in\u00a0engagement rates, customer retention, or\u00a0revenue impact, meaning the company needs\u00a0to demonstrate that easier access to CRM intelligence translates into recurring improvements rather than\u00a0higher levels of\u00a0adoption.\u00a0<\/p>\n
Buyer Concerns: Costs and AI Reliability<\/p>\n
Despite its Q2\u00a0adoption milestones and customer success stories,\u00a0HubSpot\u2019s buyer concerns primarily focused on\u00a0whether AI agents can reliably work with a customer\u2019s own CRM data and business processes, whether AI costs remain predictable, and whether customer-facing agents\u00a0can\u00a0justify replacing human interactions. \u00a0<\/p>\n
HubSpot\u2019s responses acknowledged that these concerns remain valid, even as the company argued it is making steady progress in addressing them.\u00a0<\/p>\n
Responding to where\u00a0HubSpot\u2019s\u00a0agentic products need\u00a0to improve, Rangan admitted that the biggest challenge is ensuring they consistently produce dependable outcomes. \u00a0<\/p>\n
\u201cWhat takes one more jump is really making sure that the outcomes of any agent that interacts directly with customers is solid, and that is the Prospecting Agent and the Customer Agent and so on,\u201d she said. \u00a0<\/p>\n
Whilst\u00a0CRM context and governance provide\u00a0an important competitive advantage, customer-facing AI will only gain widespread acceptance when\u00a0it\u00a0delivers consistent results.\u00a0<\/p>\n
In regard to cost, many\u00a0buyers are demanding clearer evidence of value before increasing AI spending.\u00a0<\/p>\n
\u201cIt\u2019s not surprising that customers who are navigating a platform shift, they want clarity in terms of the outcomes that they can get, and they want predictability in terms of the cost,\u201d Rangan admitted. \u00a0<\/p>\n
This includes offering\u00a0lower entry\u00a0and outcome-based prices, trials, usage visibility, customer-defined spending thresholds, and implementation support from solution architects and partners, aiming to\u00a0reduce\u00a0the financial and operational risks of AI adoption.\u00a0<\/p>\n
However,\u00a0spending discipline is already affecting customer behavior,\u00a0as HubSpot reported customer dollar retention in the high 80\u2019s and net revenue retention went down one percentage point YoY.\u00a0<\/p>\n
Kate Bueker, CFO at HubSpot, acknowledged that customers are reassessing their technology spend at renewal, creating pressure on retention even as AI adoption expands.\u00a0<\/p>\n
\u201cAs customers come up for renewal, we\u2019re seeing some downgrade pressure as they sort of optimize their overall spend.\u201d<\/p>\n
Whilst\u00a0HubSpot has addressed many barriers to evaluation through pricing flexibility, implementation support, and governance, it has yet to demonstrate how customer-facing agents will manage accuracy, errors, and safety at scale. \u00a0<\/p>\n
Longer term, investors and customers are likely to look beyond activation and resolution metrics toward evidence\u00a0that AI agents drive lasting renewal and expansion rather than encouraging customers to reallocate or reduce software spending.\u00a0<\/p>\n
By\u00a0demonstrating measurable business outcomes over time,\u00a0HubSpot can ensure customer retention for their newer products.\u00a0<\/p>\n
HubSpot Key Earnings Results<\/p>\n