Our content brings visitors, but not buyers.

B2B companies publishing blog posts and articles that get traffic from search, while sales never hears a buyer mention them.

Usually seen inCompanies with a busy blog and a quiet pipeline

First testWhich pages buyers actually read

PracticeGTM

What you're probably seeing

If this sounds familiar, you're in the right place.

Traffic grows, but sales never hears a buyer mention an article.
The most-read pages answer hobby or student questions.
Articles read alike, because they were written to rank, not to help a buyer decide.
Several articles answer the same question and compete with each other.
Buyers ask AI assistants, and your company isn't in the answers.

The usual ways out

What teams usually try. Each works somewhere. None works everywhere.

The way	Works when	The catch
Publish more articles	Each one answers a question buyers really ask	Volume without that makes the problem bigger
Hire an SEO agency	They understand your buyers, not just keywords	Often optimises for traffic, which is the problem
AI-written articles at scale	Speed matters more than trust	Thin and alike, and risky under search engines’ spam rules
Rewrite the best pages for buyers	Some pages already rank and are on topic	Slower, and some traffic goes
Topic hubs with linked chapters	You can cover a subject properly	Needs an expert to write and sign them
Pages built from real sales questions	Sales calls repeat the same questions	Needs someone to collect them
Remove or merge pages that never convert	Many pages bring no buyers and no links	Some traffic goes, and smaller sites carry more risk

What other companies did. Their published results, not ours.

We read each company's own write-up and asked one question: would our diagnostic's tests have led to the same call? We judge the approach, not the result.

Company	What they did	What they published	Source
HubSpotLarge B2B CRM vendor. A LinkedIn claim in early 2025 said its blog traffic had fallen 80%; HubSpot's SVP of Marketing said that figure was wrong but confirmed traffic was declining.	Removed over 30,000 blog pages that were not contributing to conversions, uses a fit-and-intent model to match content to its audience, and shifted effort to YouTube, podcasts, a creator program and social media.	Not stated in this source (no traffic or lead numbers)They noted: HubSpot says it began preparing for search changes about five years earlier and does not chase traffic for its own sake; channels are judged on lifetime value and acquisition cost.	HubSpot diversifies marketing channels amid declining blog trafficFeb 2025
QuickBooks (Intuit)Small-business accounting product; its Resource Center blog had several thousand posts.	Deleted about 2,000 posts (over 40% of the Resource Center) that had fewer than 100 visits in six months and no conversions, and redirected their URLs to related pages.	Traffic up about 20% within a few weeks and up 44% by peak tax season (January to May). Signups up 72% (no timeframe given).They noted: Deleting pages does not guarantee more traffic. QuickBooks had strong domain authority and many channels, so smaller sites carry more risk. Try topic hubs and refreshing old posts first.	How QuickBooks Nearly Doubled Traffic by Deleting Half Its ContentJul 2018
VercelDeveloper cloud platform with self-serve signups.	Shifted content toward questions developers were asking in forums that were not well answered, with original data and code. Served static HTML so AI crawlers can read pages, and tracked AI referrer traffic and citations.	ChatGPT referred about 10% of new signups, up from 4.8% the month before and 1% six months earlier (date of the figure not stated).They noted: There is no reliable way to confirm whether content appears in AI answers, and attribution is not clean. They present their methods as patterns being tested, not proven results.	How we're adapting SEO for LLMs and AI searchJun 2025
WebflowWebsite builder. Before the work, 4% of its signups came from AI assistants (the page gives no date for this figure).	With its SEO partner (Graphite), built a plan for AI answers between January and July 2025, and tested some changes against control pages over about five weeks: answered specific long-tail questions, added direct-answer copy to about 1,500 template pages, rewrote headers as questions, and posted helpful replies on Reddit.	Share of voice in AI answers rose from 33% to 64%. AI-assistant signups rose from 4% to 8% of all signups. AI traffic converted to signup at 24% against 4% for non-brand search. Template-page AI signups went from 168 in Q1 to 983 in Q2.They noted: Control pages lost traffic during the tests, so control-adjusted gains look larger than the raw changes. Reddit made up about 10% of ChatGPT citations.	Webflow 2x'd Sign-Ups from LLMsJan 2026 · vendor story

Our read

Same call. The companies that acted judged pages by conversions or signups, not traffic alone, and Webflow measured its share of AI answers directly, which matches our tests.

HubSpot removed over 30,000 pages that did not lead to conversions, and QuickBooks deleted posts with under 100 visits in six months and no conversions; both judged pages at least partly by buying behaviour, as the top-pages test does. Vercel and Webflow measured signups from AI assistants, and Webflow tracked its share of AI answers, matching the AI-mention test. None of the four says it compared published pages with the questions buyers ask on sales calls, or checked for pages competing for the same search, which we would still test. For Vercel and Webflow the measure is self-serve signups, not sales conversations.

What we would test first: List your top pages by traffic and check whether a buyer would actually read them.

What the diagnostic would test

The tests that rule the wrong ways out.

On your own systems and data, read-only first. The number of tests sets the diagnostic's price, quoted before it starts.

1
Your top pages by traffic, and whether a buyer would read them.
2
The questions buyers ask on sales calls, against what you've published.
3
Pages that compete for the same search.
4
Whether AI assistants mention you for your category's questions.
5
Which pages a booked lead read before the call.

This week, without us

Three things you can do yourself.

1
Take your ten most-visited pages and mark each: would a buyer read this?
2
Write down the questions from your last ten sales calls, and check if you've answered each one in writing.
3
Ask an AI assistant the question your buyers ask most, and see who it names.

Sound like you? Bring it to a 30-minute call. We'll tell you which tests your case needs, and what they cost.

Back to every case

This page describes what teams in this situation commonly face. It is not a client story.