Organic Search Statistics 2026: The Numbers Behind a Channel Everyone Keeps Writing Off

Organic search spent 2025 being written off. Two years of AI Overviews, answer engines and "SEO is dead" posts have left a lot of marketing directors quietly assuming the channel is in structural decline.

The data from 2026 tells a more awkward story. Organic search is still, by a wide margin, the largest single source of referred traffic on the open web, and the mechanics that decide who gets that traffic have barely moved. What has changed is the size of the prize at the top of the page and how long it takes a new page to reach it.

This piece pulls together verified figures from primary sources published in 2025 and 2026: Google's own status dashboard, anonymised Google Search Console data from hundreds of thousands of sites, Ofcom's UK panel research, and clickstream estimates. Where the numbers disagree, that disagreement is itself the finding.

Key organic search statistics for 2026

• A typical website's whole-site organic click-through rate sits between 1% and 2%, based on Ahrefs analysis of anonymised Search Console data from 422,421 websites in June 2026. [1]
• Median organic CTR ranges from 0.39% for sites with a Domain Rating of 0-10 up to 2.91% for DR 90-100, per Ahrefs. [1]
• 96.55% of pages in a 14 billion page sample get zero search traffic from Google, and a further 1.94% get between one and ten monthly visits, according to Ahrefs. [3]
• Keywords with fewer than ten monthly searches account for almost 93% of Ahrefs' US keyword database, roughly 2.3 billion terms, against just under 18,000 keywords above 100,000 searches a month. [4]
• Backlinko's analysis of 4 million search results puts position one at a 27.6% average CTR, with the top three results taking 54.4% of clicks. [9]
• Only 1.74% of newly published pages reached Google's top 10 within a year, and 72.9% of pages currently in the top 10 are more than three years old, per Ahrefs. [7]
• Google shipped two core updates and two spam updates in the first seven months of 2026, against four core updates in 2024, according to Google's Search Status Dashboard. [8]
• Google Search reached 82% of UK online adults in May 2025, down from 83% a year earlier, with an average of 1 hour 27 minutes spent per person per month, per Ofcom. [13]
• Google supplied 39.98% of all traffic to 76,000 sites tracked by Ahrefs Web Analytics, against 0.21% from ChatGPT. [5]
• The single most-searched term on Google in the US in July 2026 was "chatgpt", at 94.6 million searches a month, per Ahrefs. [6]
• Mobile accounted for 52.57% of worldwide platform usage in July 2026, against 45.93% for desktop, per StatCounter Global Stats. [14]
• Similarweb ranked bing.com the 10th most visited site in the world in June 2026 and duckduckgo.com 50th, with Brave absent from the top 50 search engines entirely. [15][16]

What share of web traffic still comes from organic search?

Organic search remains the dominant referral channel. On the largest publicly reported measurement, Google alone accounts for around four in every ten visits to tracked websites, an order of magnitude more than every AI assistant combined.

Ahrefs' Web Analytics data across 76,000 sites attributed 39.98% of traffic to Google and 0.21% to ChatGPT, a 190x gap in favour of Google [5]. The same analysis estimated Google's effective click-through rate at 29.2% against ChatGPT's 1.3%.

That gap is the number most commonly misread. It does not mean AI referrals are irrelevant, and it does not mean organic volumes are healthy. It means the two things are on completely different scales, and a 200% growth rate on a 0.21% base still leaves you with a rounding error.

The more useful reading is directional. AI assistants are growing fast from near zero whilst search referrals are flat to slightly down, so the crossover point, if it exists at all, is years out rather than quarters. Budgeting as though it were imminent is the expensive mistake.

What is a realistic organic CTR in 2026?

Far lower than most benchmark posts suggest. Across 422,421 websites, Ahrefs found the typical whole-site organic CTR falls between 1% and 2%, and that authority predicts it better than industry does. [1]

The breakdown by Domain Rating is stark.

Source: Ahrefs, anonymised Google Search Console data, June 2026 [1][2]

Site size pulls in the same direction. Sites with 0-10 indexed pages had a median CTR of 0.58% and a median of 12 monthly clicks, whilst sites with 5,000 or more indexed pages managed 1.74% and 132,889 clicks. [1][2]

Two things follow. First, whole-site CTR is not comparable to position-level CTR, and conflating them produces nonsense targets. Second, the spread between a DR 0-10 site and a DR 90-100 site is roughly 7.5x on CTR and around 9,000x on absolute clicks, which tells you the power law in organic search is on traffic volume rather than on click behaviour.

If your site sits at DR 30 and your board is benchmarking you against a 2.9% CTR, someone has quietly handed you a target built for Wikipedia.

How much traffic does each ranking position actually get?

Position one still takes somewhere between a quarter and two-fifths of clicks depending on whose data you use. The honest answer is that no single credible number exists, because SERP layout now varies more than position does.

Backlinko puts the top three results at a combined 54.4% of clicks, and found that moving up a single position raises relative CTR by 32.3% on average, rising to 74.5% for the jump from two to one. [9]

SISTRIX's caveat is the important one. Its position-one CTR ranged from 13.7% when Google Shopping appeared on the page to 46.9% when the result carried sitelinks. [10] That is a 3.4x swing driven entirely by page furniture, not by ranking.

seoClarity's year-on-year comparison of roughly 750 million impressions and 12 million keywords found position one CTR down just 0.66 basis points on desktop and 0.13 on mobile between April 2024 and April 2025, whilst position three fell 2.41 and 1.31 basis points respectively. [11] The erosion is real but it is concentrated below the fold, not at the top.

Treat any single CTR curve as a planning heuristic, never as a forecast. The variance within a position is now larger than the difference between adjacent positions.

How much of search is long-tail?

Almost all of it, by keyword count. Almost none of it, by volume concentration. Both statements are true and they explain most bad keyword strategies.

Ahrefs found that keywords with fewer than ten monthly searches make up nearly 93% of its US keyword database, around 2.3 billion terms, whilst just under 18,000 keywords exceed 100,000 monthly searches. [4]

The corollary is brutal. Across a 14 billion page sample, 96.55% of pages get no Google traffic at all and 1.94% get between one and ten monthly visits. [3]

Read those two figures together and the picture clarifies. The long tail is not a soft option for sites that cannot compete on head terms, because most long-tail pages get nothing either. What the long tail actually offers is cheap coverage of intent you can only reach by having a page that exists, and that only pays when the page sits on a domain with enough authority to be shown at all.

What are people actually searching for?

Brands, overwhelmingly. Ahrefs' July 2026 ranking of the most-searched US terms is a list of company names, and the top entry is a competitor to the search engine itself.

The top five US searches were "chatgpt" (94.6m per month), "youtube" (86.1m), "amazon" (85.8m), "gmail" (58.0m) and "wordle" (45.9m). [6]

Eight of the top ten are brand names or branded products, counting "chat gpt" and "google translate" as separate entries. [6]

This is the statistic that should reframe how you think about the channel. The highest-volume searches on Google are navigational, which means the majority of visible search demand is people using the search box as an address bar. Those queries are effectively unwinnable for anyone other than the brand named in them.

It also means brand building and organic search are not separate line items. If your brand is not a search term, you are competing only for the residual non-navigational demand, which is a much smaller pool than raw keyword tools imply.

How volatile is Google now, and how often does it update?

Less often than in 2024, and the updates finish faster. Google's own Search Status Dashboard is the only fully reliable record here, and it shows a clear deceleration.

Source: Google Search Status Dashboard, ranking updates history [8]

The March 2024 core update ran for 45 days. The May 2026 core update ran for 11 days and 21 hours. [8]

Shorter rollouts are not the same as calmer results, and it would be wrong to read this as Google settling down. A plausible alternative reading is that continuous ranking systems now absorb changes that once needed a named, announced update, which would make the dashboard a worse proxy for volatility than it used to be.

What has demonstrably hardened is entrenchment. Ahrefs found that only 1.74% of newly published pages reached the top 10 within a year, against 5.7% in its 2017 predecessor study, and that 72.9% of pages currently in the top 10 are more than three years old. [7] Where pages did break into the top 10, 40.82% did so within the first month.

If a page has not moved in six months, the data says it probably will not without substantial revision. That is a content-refresh argument, not a patience argument.

Is search still mobile-first, and what does the UK look like?

Mobile leads, but by less than the received wisdom suggests. StatCounter put mobile at 52.57% of worldwide platform usage in July 2026, with desktop at 45.93% and tablets at 1.5%. [14]

Ofcom's UK panel data gives a different cut of the same behaviour. Smartphones accounted for 77% of total UK online time, whilst Google Search reached 82% of UK online adults in May 2025, down marginally from 83% a year earlier. [13]

Time spent is the quieter signal. The average UK user spent 1 hour 27 minutes on Google Search across the month of May 2025, down from 1 hour 29 minutes the year before, on around 3 billion UK web searches. [13]

A two-minute drop in monthly time spent is not a collapse. It is, however, the first sustained downward movement in a metric that spent a decade going one way, and it arrived alongside Ofcom's finding that 53% of UK people say they often see AI summaries and that AI-supported overviews appear on 30% of keyword queries. [13]

The gap between the panel numbers and the doom narrative is worth sitting with. UK reach fell one percentage point. Anyone modelling a 30% organic decline on the basis of AI Overviews is extrapolating well beyond what any measured dataset currently supports.

Does anyone use anything other than Google?

Barely, and the alternatives are not gaining in any way that shows up in traffic. Bing is a genuinely large website by absolute numbers and an almost irrelevant one by share.

Similarweb ranked bing.com the 10th most visited website in the world in June 2026, with traffic up 0.73% month on month. [15] DuckDuckGo ranked 50th globally and third in the US search engines category in May 2026, with traffic up 3.61% and 95.31% of visits arriving direct. [16]

Brave Search did not appear in Similarweb's top 50 search engines at all. [15]

DuckDuckGo's direct-traffic share is the interesting detail. A search engine that gets 95% of its visits from people typing the URL or using a set default has an extremely committed but essentially fixed user base, which is why its share has been stable rather than growing for years.

For most UK businesses the practical conclusion has not changed. Optimise for Google, verify nothing is broken in Bing Webmaster Tools because Bing's index feeds parts of ChatGPT and Copilot, and stop there.

How to read these numbers

Almost every figure here comes from one of three measurement types, and they are not interchangeable.

Search Console aggregates, such as the Ahrefs CTR and traffic benchmarks, are server-side and accurate for the sites in them, but the sample is self-selected towards sites that connect their Search Console to an SEO tool. That skews towards actively managed sites.

Clickstream and panel data, such as Similarweb's visit estimates and Ofcom's UK reach figures, model behaviour from a sample and extrapolate. They are directionally strong and precisely wrong, which is why two providers can differ by 20% on the same website.

Meta-analyses, such as the First Page Sage CTR curve, blend other people's studies. They are useful as a rough consensus and useless as evidence, because you cannot audit the weighting.

Three further cautions. Position-level CTR and whole-site CTR measure different things and should never be compared. Studies published in 2025 or 2026 may still rest on data collected in 2023, as the Ahrefs ranking-timeline study does. And Google's update dashboard records announced updates only, so a quiet year on the dashboard is not evidence of a quiet year in the SERPs.

What this means going into 2027

Authority compounds faster than tactics. A 7.5x CTR spread between low and high Domain Rating sites, on identical positions, says the cheapest available lever for most sites is not on-page work.

Brand is a search strategy. When eight of the ten highest-volume US searches are brand names, the fastest route to more organic clicks for many businesses is becoming a term people type rather than ranking for terms they already type.

New pages need a longer runway. With 1.74% of new pages reaching the top 10 within a year and 72.9% of incumbents older than three years, content plans built on a six-month payback are mispriced.

Measure decline, do not assume it. UK Google reach fell one point and average time spent fell two minutes. Those are the observed numbers. Anything larger in your own reporting is either sector-specific or a tracking artefact, and it is worth knowing which.

Watch the layout, not the ranking. A 3.4x swing in position-one CTR depending on whether a SERP carries Shopping results or sitelinks means SERP feature monitoring is now more predictive of traffic than rank tracking alone. If you want help working out how findable your business really is in organic search and AI answers, you can book a call with me.