HubSpot's blog peaked at more than 20 million organic visits a month, and as AI answers spread it lost about 80 percent of that organic traffic, yet revenue kept climbing, Search Engine Journal reported on October 29, in an interview with chief marketing officer Kipp Bodnar. The story, drawn from Bodnar's new book Loop, co-written with Kieran Flanagan, is less about a traffic collapse than about what HubSpot measured instead when traffic stopped being a useful scoreboard.
Bodnar's account is candid about the scale of the loss. When Google rolled out AI Overviews, HubSpot lost nearly 5 million visits in a 30-day period, and the blog's overall organic traffic fell by about 80 percent from its peak. That is a collapse by any normal measure, and for a company that built its reputation on inbound marketing, it was a public embarrassment as well as a business problem. Other companies posted data suggesting the loss, and the story spread across social media before HubSpot told its own version.
Bodnar's response is that much of the lost traffic was never going to turn into revenue. The blog had spent years publishing broad informational content that ranked well and attracted visitors, but many of those visitors were not buyers, and many of the topics were only loosely related to HubSpot's products. Losing that traffic hurt a vanity metric, but it did not remove the customers. That distinction, between traffic and revenue, is the heart of the story, and it is one many publishers are still learning to make.
Revenue grew nearly 20 percent in 2025, according to Bodnar, because HubSpot had spent years building channels that do not depend on search. Today, he said, about 90 percent of HubSpot's content leads come from channels that barely existed in its mix five years ago, including YouTube, podcasts, newsletters, and social, while the blog accounts for about 10 percent. The company rode out the traffic drop because it had built audiences Google could not take away, well before it needed them.
The AI answer traffic that did remain also behaved differently. Bodnar said visitors arriving from AI answers converted at roughly three times the rate of traditional search visitors, and that qualified leads from AI answers grew 1,850 percent between the first quarter of 2025 and the first quarter of 2026. That suggests AI traffic, while smaller, is more qualified: a user who reaches a site through an AI answer has already had a question answered and is closer to a decision. Volume fell, but value per visit rose.
The first lesson is to diversify before you need to. HubSpot's non-search channels took five years to build, and they were built while search traffic was still strong. A publisher that waits until traffic falls to start a newsletter, a video channel, or a community will be building under pressure, with less time and less audience to transfer. Diversification is insurance, and like all insurance, it is cheapest when nothing is yet wrong.
The second lesson is to measure what matters. HubSpot tracked traffic as its scoreboard for 15 years, Bodnar said, and that scoreboard said the company was failing even as revenue grew. Publishers that still judge content by page views alone will make the same error, cutting valuable content because its traffic fell, or preserving worthless content because its traffic held. Leads, conversions, citations, and audience growth are harder to measure than page views, but they are the measures that connect content to a business.
For SEOs, the HubSpot story is a warning against defending traffic as an end in itself. If a client's informational traffic falls because AI answers satisfy those queries, the right response may not be to win the traffic back. It may be to shift effort toward content that AI cannot fully replace, toward commercial and transactional pages, and toward channels that build a direct audience. An SEO who only reports traffic will look like a failure in an AI search world, even when the business is succeeding.
For link builders, the lesson is similar. A link from a high-traffic informational page that no longer receives that traffic is worth less than it was, while a link from a page that is cited in AI answers, or from a channel with a loyal audience, may be worth more. Evaluate link targets by their current role, not by their historical traffic. The HubSpot case shows that traffic history is a poor guide to present value, and link builders who ignore that will keep buying yesterday's visibility at today's prices.
Sources: Search Engine Journal, October 29, 2026.
Linkbaits helps you create content that earns citations and backlinks. Start with the free tools and build your first linkable asset this week.
Start free trial