HubSpot Lost 80% of Blog Traffic, Saw Revenue Rise 20%: CMO Explains

HubSpot's blog lost roughly 80% of its peak traffic — more than 20 million monthly organic visits — and the company's revenue still grew nearly 20% in 2025, according to CMO Kipp Bodnar. His account of what HubSpot measures now, and what it stopped measuring, appears in Loop: Outlearn. Outmarket. Outgrow., the book he co-wrote with HubSpot's Kieran Flanagan and published Sept. 22, 2026.

The old playbook died fast. When Google rolled out AI Overviews, Bodnar said, HubSpot lost nearly 5 million visits in a 30-day period.

"We had tracked traffic as the scoreboard for 15 years, and it said we were failing," Bodnar said. Then the team examined the visitors still arriving, especially those coming from AI answers, and found they converted at roughly three times the rate of traditional search.

What Was the Lost Traffic Actually Worth?

"Much of that traffic was never going to turn into revenue," Bodnar said. Some HubSpot posts chased search volume on topics like "famous quotes" or "how to type the shrug emoji." When AI Overviews answered those questions directly, the traffic vanished, and almost nobody who mattered went with it.

Revenue held up because HubSpot had spent years building channels that don't depend on search:

• The Hustle newsletter, acquired in 2021
• A podcast network, launched in 2021
• A creator program, launched in 2022
• A YouTube network, launched in 2023
• More than 3,000 blog pages pruned for not reaching customers

About 90% of HubSpot's content leads now come from channels that barely existed in its mix five years ago, Bodnar said. The blog accounts for about 10%. HubSpot's rebuilt chat was generating more than 10,000 sales meetings a quarter by Q4 2025.

"Website traffic used to track growth," he said. "Now we look at revenue per visit, AI share of voice, and content-influenced pipeline."

What Is Loop Marketing?

Bodnar calls Loop Marketing "the operating system for marketing in the AI era." The old funnel assumed buyers moved from awareness to consideration to conversion. Awareness now scatters across Reddit, YouTube, Discord, and AI chat windows, and buyers show up already informed. The Loop runs four stages that feed each other:

• Express: define who the team is and who it serves, captured in a Taste Profile
• Tailor: personalize messages using data from real customer interactions
• Amplify: place the brand wherever people and machines look for answers
• Evolve: feed every result back so the next cycle starts smarter

"Because each cycle improves the next, the team that learns fastest wins," Bodnar said.

How Big Are the AI Email Gains?

The publisher claims HubSpot's Loop system was "outperforming the old one" within a year. Pressed for the metric, Bodnar said the baseline was HubSpot's segment-based email program, where gains from testing subject lines and button colors "were incremental." The rebuilt program uses AI-written emails personalized per recipient, with prompts that run pages long and draw on a Taste Profile, tone guidelines, and a banned-word list.

The result: an 82% improvement in conversions for the AI-powered email, and an average 50% increase across the broader email program. Bodnar also conceded that his public claim of being a "C" with generic AI and an "A" with a Taste Profile rested on self-assigned grades.

HubSpot tests work early against a Buyer Sim, an AI version of the ideal customer that is "blunt on purpose," and refreshes the Taste Profile quarterly. The first prototype of its AEO Grader failed the test — the Buyer Sim called it generic — and took three more rounds before version 4 earned the verdict: "I'd bookmark this and send it to my CMO."

How Fast Are AI-Sourced Leads Growing?

Qualified leads from AI answers grew 1,850% between Q1 2025 and Q1 2026, Bodnar said, converting at about three times the rate of traditional search, with deal cycles roughly half as long. HubSpot tracks mentions (named in an AI answer) and citations (linked) separately across its most important prompts, and watches which sources engines pull from — often Reddit, LinkedIn, YouTube, or G2.

"Attribution is imperfect," Bodnar acknowledged. The signal is correlative, but "strong enough that we treat AEO as a core part of how we market." Prompts now run 40 to 100 words, he added, versus two to four words for a traditional search query.

What About the Layoffs?

Bodnar declined to answer a question about HubSpot's Oct. 6 announcement that it would cut nearly 660 jobs, about 7% of its workforce. News accounts report CEO Yamini Rangan told employees the move is "not driven by AI-related efficiencies" and "not simply a cost-cutting exercise," and that product teams will reorganize around four outcomes: generating demand, winning deals, delighting customers, and scaling growth. The Boston Globe reported HubSpot had 8,882 full-time employees at the end of 2025, expects $65 million to $75 million in restructuring costs, and saw shares down about 46% for the year on announcement day.

Rangan said headcount has grown more slowly than revenue and will continue to do so. The role gets bigger for the people who keep it.