Yamini Rangan on HubSpot, AI and Competing in the Midmarket

HubSpot CEO Yamini Rangan joins Harry Stebbings to discuss leading a public company, serving midmarket customers and building AI into its products. Her observations on HubSpot's performance and competitors come from a CEO with a direct commercial interest. The interview does not independently verify its customer, ticket or financial claims.

Rangan describes succeeding a founder while founders remain involved as both a privilege and work. She says she shared their conviction in small and medium businesses, a platform and company culture, but the relationship requires agreement on communication cadence, strategic scope and what happens when a decision goes wrong.

Asked how a public-company CEO maintains conviction under market scrutiny, she says leaders must start with what they believe about the business and explain decisions to investors. As an example, HubSpot changed its roadmap after generative AI appeared and chose to embed AI throughout its platform rather than treat every feature as a separate premium product.

Stebbings questions whether outcome-based pricing is practical when people and AI both contribute, and when customer outcomes vary widely. Rangan expects a hybrid of seat and usage pricing and sees credit-based usage as easier to validate across a platform that can answer tickets, arrange meetings or produce content. This is her pricing view, not a guarantee of future terms.

Rangan says scaling a distributed company requires more process than a young startup. Her framework connects strategy to annual priorities and accountable outcomes. At HubSpot she describes three broad aims: focus on customers, make the product easy, fast and unified, and maintain a growth mindset.

Stebbings challenges whether anyone would oppose a growth mindset. Rangan says the alternative is comfort with familiar methods and expertise: adopting AI, for example, requires people to learn new tools and practices. She sees curiosity and adaptation as habits that need active encouragement, not assumptions leaders can make about everyone.

The move to CEO brought a wider set of stakeholders, including founders, employees, the board and investors. Rangan says she had to accept that no decision will satisfy everyone. HubSpot's stated decision hierarchy is customer over company, team and self; it is a management principle, not evidence that every tradeoff is resolved in the customer's favor.

She says the CEO's work includes aligning functions rather than merely allocating resources. Making a product easy, fast and unified involves product teams, marketing, sales and customer success. Leaders need clear goals, ownership and a way to respond when a project runs into difficulty.

Asked about wartime and peacetime leadership, Rangan prefers to discuss operating modes, noting that actual wars are taking place. In difficult moments she may prioritize speed; at other times she emphasizes building consensus. A CEO must move between these modes, sometimes within a year or quarter.

Rangan says a recurring piece of feedback over her career has been to slow down. As a new sales manager she learned that moving quickly without context could leave the team behind, and she keeps a reminder to ask questions. Explaining why a change matters can take longer at first but help the organization go further together.

When Rangan joined in 2020, she says HubSpot used its own software to run marketing, sales and service at a company of about 4,500 employees. That gave her confidence the product could serve larger customers. She defines HubSpot's target as companies with roughly two to 2,000 employees, not the largest enterprises.

Stebbings asks whether moving upmarket demands an entirely different sales organization. Rangan favors an incremental approach: serve 200-person companies, then 500, then 1,000 and 2,000, improving product, partners and sales skills at each step. She cautions that simply importing an enterprise sales motion can disrupt the existing system and culture.

The biggest change, she says, comes when a single buyer gives way to a committee involving IT, procurement, security and perhaps the board. Sales teams must learn to navigate that process while keeping it useful for the customer. Her description reflects HubSpot's experience and does not establish a universal path for every company.

Stebbings asks about competition with Salesforce. Rangan says business applications are not generally winner-take-all and describes HubSpot's focus on companies with two to 2,000 employees versus Salesforce's strong enterprise presence. She acknowledges overlap at HubSpot's upper end. These are the CEO's market and positioning claims, not an independent comparison of the products.

She praises Salesforce's marketing and narrative, saying strong competitors make each other better. For HubSpot, she wants customers to deploy a CRM and find it intuitive enough to begin using quickly. If forced to choose where to put an incremental dollar, she says she would invest in the product first, making sales and marketing easier.

Rangan calls AI a potential equalizer for small and medium businesses because they face constraints in headcount, capital and expertise. She suggests AI support might delay a hire, personalized marketing might require fewer resources, and automated demonstrations could help sales. These are proposed benefits, not proof of savings or improved outcomes for every business.

Stebbings challenges the hype, saying some specialized AI products have lost customers because they do not work well enough. Rangan agrees there is hype and says HubSpot looks for repeated usage and value before scaling and pricing features. Her description is a company strategy rather than independent evidence of retention.

She says HubSpot's customer agent had 1,500 customers and that 42% to 45% of tickets were deflected by AI within weeks. She also says HubSpot internally resolved 35% of tickets with AI in year one and expected to reach 50%. The interview does not define the cohorts, ticket outcomes, time periods or verification methods fully; the 50% is an expectation.

Rangan says the company has also used personalized marketing emails to improve conversion, without giving a quantified comparison. She distinguishes a novel feature used once from a necessary feature embedded in daily work. Her approach is to start with a few use cases, gather feedback and iterate until patterns of repeat usage emerge.

Stebbings asks whether model costs could erode margins when HubSpot passes spending to model providers. Rangan says inference costs have fallen and expects further declines; at the time of the interview she does not see margin degradation and says adoption is still early. Her cost and margin outlook is management's assessment, not a guarantee.

Asked whether startups or incumbents have the advantage, Rangan says speed and innovation matter more than size. She describes a cultural shift at HubSpot toward faster cross-functional decisions, clear accountability and greater AI fluency among employees. Her assessment of the company's agility is a management claim.

Rangan says she uses Claude and other models in her own work. Stebbings recounts a model producing plausible but apparently invented claims about her, underscoring his concern about sourcing. She agrees citations matter in some use cases; the anecdote is not a systematic comparison of model accuracy.

For earnings scripts, Rangan says she gives a Claude project the previous 16 quarters of scripts and asks it to help shape and refine the next narrative. She compares outputs from multiple models and describes a writing session of around three to four hours. She treats the tools as thinking partners while remaining responsible for what she communicates.

Stebbings asks what happens to HubSpot's search-driven traffic as search engines and AI tools increasingly provide answers directly. Rangan expects that shift to continue and says HubSpot had already diversified its content through podcasts and newsletters. She reports millions of podcast listeners, without giving a defined measurement period.

Asked about returns on media acquisitions, she describes them as top-of-funnel distribution rather than a simple one-touch path from listening to a purchase. HubSpot looks at awareness and multiple touches across channels, she says, and tries to meet customers on platforms such as YouTube, Instagram, TikTok and LinkedIn. She does not provide acquisition-level ROI.

In quick fire, Rangan says pricing should attract customers through value and market-share growth rather than maximize each immediate dollar. She says HubSpot has lowered pricing to increase perceived value. This is the company's pricing philosophy, not a comprehensive history of all product prices.

Asked to choose a public stock other than HubSpot for ten years, she chooses Microsoft and praises Satya Nadella's cloud and AI bets and cultural leadership. She says she would like Anthropic CEO Dario Amodei on her board. These are personal preferences, not investment advice or an announcement of an appointment.

Rangan says she changed her mind about her son attending university away from California after seeing him feel at home on the East Coast. Parenting two teenagers, she says, has made her more patient as a leader. She also says HubSpot recently changed from unlimited vacation to flexible time off, without yet assessing the result.

She cites an estimate that gender parity could take another 134 years at the current rate and says this concerns her. The interview does not identify the source, geography, dimensions or methodology in enough detail to independently assess the estimate. Stebbings worries inequality may lose attention during difficult times.

Rangan calls public-company leadership a volatile but rewarding journey and says meditation, yoga and family help keep her grounded. She notes that at a large company there is almost always something that needs fixing, which can make it tempting to focus only on what is wrong.

She closes by challenging the assumption that software companies must move continually upmarket to become large. Rangan wants HubSpot to define a midmarket software category alongside companies focused on enterprises or small businesses. That is her strategic ambition, not evidence that the category or target will develop as planned.