Is Top-of-Funnel Content Worth It in 2027? The Traffic Data

Is top-of-funnel content worth it in 2027? The answer splits down the seam of your industry. Sell to B2B tech buyers, in cybersecurity or cloud infrastructure, and the introductory guides still earn their slot. Publish a generic how-to library in finance, health, law, or consumer tech and the traffic is draining out. A Semrush analysis reported by Search Engine Land, covering 30 of the largest digital publishers across nine industries and comparing August 2026 against the same month a year earlier, makes the divide hard to wave away. The question is no longer whether to write top-of-funnel content. It is which funnel, in which market.

Is top-of-funnel content worth it? Read the split, not the average

Organic search clicks have fallen 42% since AI Overviews began expanding across Google Search, per the same Semrush pull. That is the number driving budget meetings, and taken alone it reads like a verdict on the whole format. The aggregate lies by omission. Underneath it, the decline is wildly uneven.

Every domain Semrush tracked in cybersecurity and in marketing and sales software gained organic traffic year over year. Cloud infrastructure closed the window positive as a group. The bleeding was concentrated, not general. It landed on finance, healthcare, legal, and consumer tech: the glossary pages, the “what is” explainers, the beginner guides that used to compound quietly for years.

There is a second force at work. AI writing tools let publishers flood results with top-of-funnel articles at scale, so competition for each remaining blue link has risen at the exact moment the clicks thinned out. More supply, less demand. That is a bad trade for anyone shipping undifferentiated guides.

Why YMYL and B2B tech are pulling apart

The split is not random. It tracks a line Google has enforced for years. YMYL content, “your money or your life”, covers finance and medical topics where a wrong answer can harm someone, and Google holds it to a higher bar and answers it more directly in AI Overviews. When a searcher asks what an index fund is or whether a symptom is serious, the model composes a confident summary and the click never happens.

B2B tech queries behave differently. They are narrower and more technical. They rarely end at the definition. Someone comparing an auth flow or a SIEM has a follow-up, then another, and the answer engine cannot close that loop the way it closes “what is APR.” The zero-click pressure is real everywhere. In the first four months of 2026, 68% of Google searches ended without a click, up from about 60% in 2024, per Similarweb clickstream data reported by Search Engine Land. AI Overviews appear on more than 20% of searches (Semrush data), and when they do, click-through can fall by roughly 60% (Ahrefs data). That pressure hits a definition page far harder than a page built for a technical evaluation. We traced the front-end version of this in AI search visibility versus organic rankings: a page can rank first and still open below the fold.

The methodology’s limits

Read the 42% figure with its edges showing. Thirty publishers is a small sample, and Semrush chose the largest ones, so the numbers describe major brands rather than the long tail where most developer-led teams actually compete. The traffic values are Semrush estimates, not measured clicks from anyone’s analytics, and the comparison spans one year-over-year window that catches a trend but not a full cycle. Direction, not gospel.

What a TOFU content strategy for 2027 should do

Here is the reallocation, not the retreat. Killing top-of-funnel content wholesale would misread this data, especially for B2B tech, where the SEO traffic is holding or climbing. The move is to stop funding pages an AI Overview can fully satisfy, and to fund the ones it cannot.

• Write for the follow-up, not the definition. A model answers “what is a vector database” in a sentence. It struggles with “how do I migrate a 200GB pgvector index without downtime,” because that answer needs judgment, trade-offs, and first-hand detail.
• Instrument representation, not arrivals. When the click disappears, the win becomes being the source the model cites. Build entity coverage and a mention feed you own, the argument we made in full in measuring demand generation in the AI search era.

Durability decides the rest. The pages that survive as AI Overviews drain the organic traffic are the ones grounded in something a competitor cannot regenerate by prompting the same model, a case laid out in surviving Google algorithm updates. If your budget funds one library this year, make it the one built on original analysis and real experience, aimed at the questions the answer engine hands back rather than the ones it absorbs.

AstroAgent’s defaults point content at that durable layer by design, so a small B2B team can ship the pages that still earn a click instead of the ones an AI Overview swallows. That is the bet worth making in 2027.