HubSpot Loop vs Flywheel: B2B Marketing With Fewer Clicks

HubSpot’s Flywheel and its Loop answer different questions. The Flywheel, which HubSpot introduced at INBOUND 2018, is a model of how a company grows: you attract, engage and delight customers, and their momentum brings in the next ones. The Loop, introduced at INBOUND in September 2025, is a four-stage playbook for marketing when buyers get answers from AI: Express, Tailor, Amplify and Evolve. HubSpot has not retired the Flywheel. Its Flywheel page is still live, and when HubSpot launched the Loop it set it against the funnel and the old inbound playbook, not against the Flywheel.

What changed between the two is search. Two of HubSpot’s marketing leaders, Kipp Bodnar and Kieran Flanagan, wrote in Fortune in September 2026 that “Our blog lost 5 million visits in a thirty-day period,” and called the inbound era they had built over. The company that taught B2B marketers to publish, rank and convert says that model no longer carries a business on its own. This guide covers how the two frameworks fit together, where the clicks went, what to measure instead, and how a B2B team can change its plan.

The Flywheel: a growth model from 2018

Brian Halligan, HubSpot’s co-founder, presented the Flywheel at INBOUND 2018 and explained it in Harvard Business Review: “The traditional metaphor for sales is the ‘funnel.’ But today, a better metaphor is the ‘flywheel.’” A funnel ends at the sale, and customers are its output. A flywheel puts customers at the center and treats them as a source of energy. Happy customers refer others, write reviews and buy again, and every point of friction in sales, service or onboarding slows the wheel.

The Flywheel has three phases, attract, engage and delight, and it covers the whole company: marketing, sales and service each add force. HubSpot’s page still says funnels “aren’t going away” and are “one piece of a larger flywheel.” Nothing in the model depends on how people search, which is one reason it hasn’t needed replacing.

The Loop: a marketing playbook from 2025

HubSpot announced the Loop on September 3, 2025, the first day of INBOUND in San Francisco, where CEO Yamini Rangan presented it in the opening keynote. Bodnar, HubSpot’s CMO, gave the reason in the release: “The old funnel assumed customers would come to you.” HubSpot’s newsroom post defines the four stages:

• Express: “Define your taste, tone, and point of view before bringing in AI.”
• Tailor: “Use AI to make messaging personal, contextual, and relevant at scale.”
• Amplify: “Diversify content across channels to meet buyers where they are.”
• Evolve: “Iterate quickly and effectively with AI.”

The Loop is narrower than the Flywheel. It describes how marketing makes and distributes content, not how a company keeps customers. It is also a cycle with no end point: what you learn in Evolve feeds the next round of Express. On whether it replaces inbound, HubSpot’s Loop page says: “The core components of inbound marketing are still essential, but they aren’t enough anymore.”

HubSpot has since built products around it. Its Spring 2026 Spotlight trained the Breeze Assistant on Loop Marketing and launched HubSpot AEO, a tool for tracking how a business appears in answer engines such as ChatGPT, Gemini and Perplexity.

Loop vs Flywheel, side by side

• Scope: the Flywheel covers the whole customer relationship across marketing, sales and service. The Loop covers how marketing creates and distributes content.
• The problem it answers: the Flywheel replaced the funnel, which treated customers as an end point. The Loop responds to buyers finding answers without visiting your site.
• Stages: attract, engage and delight, against express, tailor, amplify and evolve.
• Role of AI: none in the 2018 model. In the Loop it is central, helping create, personalize and test content.
• Status: both are on HubSpot’s site, and we found no HubSpot page that compares them directly.

A team can use both. The Flywheel explains why retention and referrals matter more when new customers get harder to reach. The Loop explains how to make marketing that gets found and quoted when fewer people click through to read it.

Where the clicks went

The number HubSpot cited at launch, that “60% of searches never leave Google,” comes from a 2024 SparkToro study based on clickstream data from Datos, which HubSpot footnoted in its release. It found that 58.5% of Google searches in the US and 59.7% in the EU ended without a click on any result. Its data ends in May 2024, the month AI Overviews launched in the US. SparkToro’s 2026 follow-up, using Similarweb data from January to April 2026, found that 68% of US Google searches ended without a click, though the two panels aren’t directly comparable.

AI summaries have pushed further in the same direction. In a Pew Research Center study of 900 US adults’ browsing in March 2025, people clicked a traditional search result in 8% of visits when an AI summary appeared and in 15% when none did. They clicked a link inside the summary in 1% of visits. Ahrefs found the top-ranking page had a 58% lower average click-through rate when an AI Overview appeared, comparing December 2023 with December 2025. Seer Interactive measured organic click-through on queries with an AI Overview falling from 1.76% to 0.61% between June 2024 and September 2025, with paid click-through on those queries falling too.

Google disputes the scale. Liz Reid, head of Google Search, wrote in August 2025 that total organic clicks to websites had been “relatively stable year-over-year” and that some third-party reports used flawed methods. Google published no figures. Both can be partly true: total clicks can hold steady while informational queries, the kind B2B blogs were built to answer, lose most of theirs.

Seer also found that brands cited inside an AI Overview earned about 35% more organic clicks than brands that weren’t, though it says it cannot prove the citation causes the difference. Being a source the answer draws on is the closest thing to a ranking in this setting, and our piece on why AI search names your competitor and not you explains how those sources get picked.

Where B2B buyers research instead

Buyers haven’t stopped researching; they have moved part of it into chat. In G2’s survey of 1,076 software buyers in March 2026, 51% said they start their research with an AI chatbot more often than with Google, up from 29% in G2’s 2025 report, and 71% used a chatbot at some point. Forrester’s State of Business Buying 2026 says generative AI searches are “the starting point for B2B buyers.” Both measure what buyers say they do, and G2 sells to software vendors, but they point the same way.

Chat doesn’t replace the rest of the journey. In 6sense’s 2025 survey of more than 4,000 buyers, 94% used large language models during a purchase, and they still averaged 16 interactions with the vendor they chose. Buyers also pass links and recommendations through private messages, Slack groups and email. The figures often quoted for how much sharing happens privately trace back mostly to a 2016 vendor study that measured its own sharing tools, so treat any precise percentage with caution. The pattern holds without one: much of the research that shapes a shortlist happens where your analytics can’t see it.

Forecasts of how much spending will move through AI search vary widely, and our look at where the $750 billion AI search number comes from shows how thin some of them are. Plan on what you can measure in your own pipeline.

What to measure when clicks fall

Start with what the search engines report. Google counts AI Mode in the totals of the Search Console Performance report, inside the “Web” search type, with no way to separate it out in that report. Its newer Search generative AI performance reports, which reached all sites worldwide on August 31, 2026, show impressions in AI Overviews and AI Mode by page, country and device, with a separate report for AI features in Discover, but no clicks or queries. Bing’s AI Performance report, in public preview since February 2026, shows how often Copilot and Bing’s AI summaries cite your pages, also without clicks.

Next, separate demand you created from demand you served. Search Console’s branded queries filter splits branded from non-branded traffic. Rising branded search alongside falling non-branded clicks usually means people learned about you somewhere else and came looking. One practitioner argues in MarTech that branded search is a weaker signal than it was, because buyers decide before they search, so read it alongside other measures.

Then ask buyers. A required open-text “How did you hear about us?” field on demo and contact forms catches podcasts, communities and AI answers that tracking misses. Refine Labs found that across 620 of its own conversions, software credited 78% to web search while customers named web search only 12% of the time. Recast warns that people misremember too, and that relying only on survey answers is as risky as relying only on last-touch tracking. Use both and look at where they disagree.

Finally, measure pipeline, not page views. Organic sessions will keep falling on many informational pages, and that alone is not a failure. Track whether qualified opportunities, win rates and deal sizes hold up. Our guide to SEO benchmarking sets out a monthly and quarterly routine for these numbers, and an online presence analysis gives you a starting point across your website, search and social media.

How a B2B team changes its plan

Budgets won’t pay for the change by growing. Gartner’s 2026 CMO Spend Survey put marketing budgets at 7.8% of company revenue, up from 7.7% in 2025, with 15.3% of marketing budgets going to AI and only 30% of marketing leaders reporting mature AI readiness. Most teams will fund the shift by moving money and time, not adding them. The Loop’s stages give a workable order.

Express: publish what a summary can’t replace

Answers to generic questions lose their clicks first, because an AI Overview or a chatbot can give them without you. Put the effort into what only you can publish: your own data, customer results with numbers, pricing logic, clear opinions on trade-offs, and named experts. These are the pages AI answers quote and the reasons a buyer still clicks. Content marketing built this way earns citations and referrals as well as rankings.

Tailor: write for the questions buyers ask

Sales calls, support tickets and that “How did you hear about us?” field show the questions buyers bring to a chatbot. Answer each one directly near the top of a page, in plain language, with the evidence underneath. Use AI to adapt one strong piece for different industries or roles, and keep a person responsible for accuracy.

Amplify: be present where answers come from

AI assistants assemble answers from across the web, including review sites, YouTube, forums, trade publications and LinkedIn. Earn mentions and reviews there, keep product and pricing details consistent across them, and give your experts a presence buyers can find. Your own site still matters, because AI tools retrieve pages that rank, so search engine optimization stays part of the work. On iPhones the assistant itself is changing, which our look at Siri and Google’s Gemini covers.

Evolve: test on a shorter cycle

Check each month which pages gain AI impressions or citations and which lose clicks, and update or merge pages that earn neither. Use paid search to test messages, keeping in mind that Google’s automation has changed what you control, as our guide to Google Ads automation controls explains. Drop tactics on evidence, not habit, and feed what you learn into the next round of Express.

Frequently Asked Questions

What is the difference between HubSpot’s Flywheel and the Loop?

The Flywheel, introduced in 2018, is a company-wide growth model with three phases, attract, engage and delight, that puts customers at the center and treats their momentum as the source of growth. The Loop, introduced in September 2025, is a marketing playbook for the AI era with four stages: Express, Tailor, Amplify and Evolve. The Flywheel covers marketing, sales and service. The Loop covers how marketing creates, adapts, distributes and improves content.

Did HubSpot replace the Flywheel with the Loop?

No. HubSpot launched the Loop as a replacement for the linear funnel and an update to its inbound playbook, and its launch materials don’t mention the Flywheel. HubSpot’s Flywheel page is still live with no retirement notice. The company’s strongest statement came in a September 2026 Fortune piece by two of its marketing leaders, published as a book excerpt, which said the inbound era they built is over. That was about inbound marketing, not the Flywheel.

What are the four stages of Loop marketing?

HubSpot defines them as Express, “Define your taste, tone, and point of view before bringing in AI”; Tailor, “Use AI to make messaging personal, contextual, and relevant at scale”; Amplify, “Diversify content across channels to meet buyers where they are”; and Evolve, “Iterate quickly and effectively with AI.” The stages repeat as a cycle, so what you learn in Evolve shapes the next round of Express.

What is a zero-click search?

A zero-click search is one that ends without the searcher clicking any result, usually because the answer appears on the results page. A 2024 SparkToro study with Datos data found 58.5% of Google searches in the US and 59.7% in the EU ended that way, and SparkToro’s 2026 follow-up, using a different data panel, put the US figure at 68%. AI summaries add to the effect: Pew found people clicked a traditional result in 8% of visits with an AI summary, against 15% without one.

How do you measure marketing when search sends fewer clicks?

Combine four sources. Use Search Console and Bing Webmaster Tools for impressions and citations in AI features. Use Search Console’s branded queries filter to track branded and non-branded demand separately. Add a required open-text “How did you hear about us?” field to forms. Then judge results by pipeline, win rates and deal size rather than sessions, because informational pages can lose traffic while still influencing buyers.