Data Report: Why Organic Traffic Benchmarks Mislead You in the GEO Era

Ahrefs released median organic traffic data from more than 344,956 real websites The numbers are useful. They are also already obsolete as a planning input if you treat them the way most marketers do.

Here is the problem: benchmarks describe yesterday’s search. They tell you what a typical site in your industry earned last month from Google clicks. They do not tell you how much of that traffic is being absorbed by AI Overviews, how much will move to answer engines this quarter, or what “normal” looks like once Generative Engine Optimization (GEO) becomes table stakes.

And the gap between the two is widening fast. A separate study from Search Engine Land, analyzing 150,000 pages across 10 websites, found that nearly half of the top 100 organic pages had zero LLM traffic. The pages winning Google are not the pages winning AI search.

If you are setting 2026 organic targets, you need both lenses. This post pulls the most useful signals out of the Ahrefs dataset, layers in the Search Engine Land findings on how LLM traffic actually behaves, then shows you how to translate both into expectations that account for GEO, not just classic SEO.

What the data actually says

Three findings matter more than the rest.

1. Industry is the smallest variable that gets the most attention

The headline tables are organized by industry, and that is the comparison most teams reach for first. It is also the least useful one. The median News site pulls 102,910 monthly clicks. The median Law & Government site pulls 4,417. That looks like a 20x industry advantage, but the gap inside any single industry is much wider than the gap between industries.

You see this in the mean-to-median spread. In Computers & Electronics, the median site earns 34,679 clicks. The mean is 2,555,795. That is a 73x gap, and it tells you the typical site lives nowhere near the average. A handful of giants are pulling the curve up.

Industry sets the room you are competing in. Authority and size determine where you sit inside it. Confusing the two is how teams end up chasing benchmarks that were never theirs to hit.

2. Domain Rating is the real predictor, and the curve is exponential

Median monthly traffic by Domain Rating, from the same dataset:

Each tier earns more median traffic than the one below it, and the jumps get steeper as you climb. From DR 0-10 to DR 90-100 is roughly a 9,000x difference in monthly clicks. That is four orders of magnitude, and it dwarfs the differences across every industry combined.

The implication for goal-setting: if your site is DR 35 and you are getting roughly 500 clicks a month, you are not behind. You are exactly where the data says you should be. Pushing past that requires the things that build DR in the first place: real content depth, real links, and real brand demand.

3. Site size compounds the authority effect

Bigger sites generally earn more traffic, and the relationship remains strongly exponential.

More pages give you more surface area to rank, but only if those pages are worth ranking. Publishing 5,000 thin pages does not produce 5,000-page traffic. The sites that show up in the top tier got there because they published things people actually search for, then earned authority on top of that.

Why these benchmarks tell you less than they used to

Every number above describes Google clicks. Clicks that left Google’s search results and landed on a website. That definition still matters, but it no longer captures the full picture of how people discover brands.

Today’s search journeys increasingly span AI-powered Google experiences, YouTube, Shopping, and conversational AI platforms alongside traditional organic search. Organic traffic benchmarks remain valuable, but they measure only one part of modern search visibility.

Three forces are eating into the baseline:

Put those three together and the picture is this: the benchmark you are measuring against was built from a search ecosystem that no longer fully exists. The data is real. The frame is shrinking.

The data: AI search rewards a different content profile

A recent Search Engine Land case study by Adam Gnuse made this concrete. The study analyzed GA4 data from 10 websites and 150,000 indexed pages, isolating LLM referral sessions from ChatGPT, Claude, Perplexity, Copilot, and other conversational AI platforms. Three findings reframe what “organic traffic” means in 2026.

Finding 1: The content that earns clicks is not the content that earns citations

Educational how-to content, the bulk of most editorial calendars, was cited by LLMs only 12% of the time. Trends and analysis posts were cited 78% of the time. Data-driven year-in-review posts: 61%. The gap is not a rounding error. It is a different game.

LLMs already know how to explain things. What they need from your site is something they cannot generate: original data, proprietary research, named tools, and answer-first formats. If your traffic floor is built on comprehensive guides, that floor does not extend into AI search.

Finding 2: Organic winners are not LLM winners

In the same study, the top 10 organic pages captured 55% of organic sessions but only 29% of LLM sessions. Of the top 100 organic pages, 49 had zero LLM traffic. Zero.

Your highest-ranking page in Google may be invisible inside ChatGPT. The benchmark looks fine. The GEO position is empty.

LLM citation and organic ranking are correlated, not identical. Optimizing one no longer guarantees performance in the other.

Finding 3: Service pages and tools punch above their weight

Measured per 1,000 organic sessions, service and product pages produced 29.4 LLM sessions. Articles produced 23.4. FAQs produced 14. Homepages produced 5.6. The page types most marketers treat as bottom-of-funnel destinations are some of the most efficient AI-traffic earners on the site.

Engagement followed the same pattern. LLM users spent 146 seconds on tool and demo pages versus 101 seconds for organic visitors to the same pages. They were not skimming. They were evaluating.

And 14% of LLM-receiving pages in the study had zero organic clicks during the window. That is pure LLM-driven discovery, invisible to every benchmark built on Google Search Console data.

How to read the benchmark through a GEO lens

This does not mean throw the benchmarks out. It means use them for what they actually measure, and add a second layer for what they cannot.

Use Google click benchmarks for floor-setting, not goal-setting

Find your row by DR and site size in the Ahrefs tables. That is your realistic floor for organic Google clicks under current conditions. If you are well below it, you have a fundamentals problem: thin content, weak topical coverage, low authority, or technical issues blocking the easy wins.

If you are at or above your row, classic SEO is doing its job. The question is what to layer on top.

Add a GEO visibility layer the benchmark cannot show you

The clicks you are losing to AI Overviews and answer engines are not lost demand. They are demand being satisfied somewhere your analytics do not see. To track that demand, you need a parallel set of metrics that do not exist in Search Console:

None of these show up in the Ahrefs benchmark. All of them will matter more in 2026 than the click number does.

Set a two-track target

Pick a Google clicks target based on your DR and size row, growing at a defensible rate (typically 10-25% year over year for established sites, higher for newer ones still building authority). Then set a second target for AI surface presence: citation count, mention frequency, or referral volume, measured against a baseline you establish this quarter.

One number measures the search world that exists today. The other measures the one being built around it. You need both, and the second one will outgrow the first.

What to do this quarter

If you are a marketing leader trying to make this concrete, here is a 90-day sequence:

The benchmark is a starting point, not a destination

Organic traffic benchmarks are useful for one thing: telling you whether the search engine that existed last month thinks you are doing the work. They are not useful for telling you whether you are positioned for the search environment being built right now.

Treat the Ahrefs numbers as your floor. Treat your GEO presence as your ceiling. The brands that hit both targets in 2026 will be the ones that did not confuse one for the other.

If you want to understand where you stand today, our free GEO Scorecard gives you a clear starting point. It evaluates how your brand appears across AI search, identifies the authority and technical signals affecting your visibility, benchmarks you against competitors, and highlights the biggest opportunities to improve your AI search performance. It’s the fastest way to see how your traditional SEO metrics compare to your AI visibility and where to focus next.

Note: The benchmark figures referenced throughout this article are based on Ahrefs’ July 2026 analysis of anonymized Google Search Console data from nearly 345,000 websites.