Search Engine Market Share by Country (August 2026)

By Bill Hartzer · Updated 17 September 2026

Google leads almost everywhere, and the exceptions are the interesting part

Worldwide search share is a nearly useless number for understanding any particular country. Google took 91.1% of worldwide search referrals in August 2026 per StatCounter Global Stats, but that average is assembled from markets running from 97.76% (India) down to 1.41% (China, on a panel discussed below). The national picture is where search actually differs.

There are, on StatCounter's August 2026 reading, only two countries of any size where Google is not first: China, where Baidu leads, and Russia, where Yandex leads. South Korea is the contested case — StatCounter puts Google first, Korean domestic measurement puts Naver first, and the two disagree by about twenty points for a reason that is methodological rather than arithmetic. Czechia and Japan are the two remaining markets where a non-Google engine holds a double-digit share.

Two classes of figure appear below and they are not interchangeable. StatCounter Global Stats, all platforms, for the August 2026 reporting month, retrieved directly on 17 September 2026, measures referral traffic — visits arriving at a site from a search engine's results page — not searches performed. Competition regulators and courts in the United States, the United Kingdom and Australia have published shares compelled from the companies themselves; where such a finding exists for a market it is given first, with its own date, because it is the better evidence. Where StatCounter's method breaks down for a specific country, it is said in that country's entry rather than buried in a footnote.

The markets where Google is not first

China — Baidu leads, but do not trust this panel

Start with the only authoritative count of how many people are searching. The China Internet Network Information Center, the official state registry, reported in its 55th Statistical Report on China's Internet Development that China had 878 million search engine users as of December 2024, representing 79.2 percent of all Chinese internet users and an increase of 51.12 million on December 2023. That absolute figure is worth more than any percentage split, because the percentage splits for this market are the least reliable on the page.

StatCounter's China table for August 2026 reads: Baidu 59.26%, Bing 18.92%, Yandex 11.41%, Haosou 6.48%, Sogou 2.25%, Google 1.41%.

That Yandex row is the tell. An 11.41% Yandex share of Chinese consumer search is not a credible description of how people in China search, and Bing at 18.92% is hard to square with any other account of the market. StatCounter's Chinese sample is small, skewed toward internationally facing and expatriate-oriented websites, and distorted by VPN traffic. These figures are reported here as what StatCounter says, attributed to StatCounter, and should not be quoted as a straight description of Chinese search. If the Yandex row is an artifact, the same artifact is depressing Baidu's row — Chinese domestic estimates run materially higher than 59.26%. All methodologies do agree that Baidu's share has fallen substantially from its early-2010s peak, so anyone still quoting "Baidu has 80% of China" is quoting a decade-old figure.

Russia — Yandex leads decisively

StatCounter, Russian Federation, August 2026: Yandex 70.35%, Google 27.79%, Bing 0.77%, Mail.ru 0.77%, DuckDuckGo 0.22%, Yahoo! 0.05%.

This is one of the cleaner national tables and it reflects a real position: Yandex runs its own crawler and index, has done since 1997, and gained share as Google's Russian operations were disrupted after 2022. Rambler, often listed as a Russian search engine, does not appear in the table at all — its search share sits below StatCounter's reporting threshold, and it has served Yandex results rather than its own since 2011.

South Korea — the sources place different engines first

StatCounter, South Korea, August 2026: Google 46.6%, Naver 43.71%, Bing 5.44%, Yandex 1.37%, CocCoc 1.36%, Daum 0.88%. The conflict this creates is set out in its own section below; it is the single largest source disagreement in current search measurement and it should not be resolved by picking a side.

There is one number for this market that does not depend on panel methodology at all. NAVER Corporation's own 2Q 2026 earnings fact sheet, an IFRS consolidated income statement published as a spreadsheet, records advertising revenue of KRW 1,447.2 billion in the second quarter of 2026, and KRW 5,436.2 billion across the four quarters of 2025. That is the core monetization of the country's dominant domestic search portal, and it is a hard figure for the one major market where Google is not the leading search monetizer. Two limits: the advertising line covers search and display advertising combined rather than search alone, and the direct download link for the fact sheet is an opaque API URL that may rotate, so the earnings landing page is the stable reference.

Czechia — the last European market with a real domestic engine

StatCounter, Czechia, August 2026: Google 79.33%, Seznam 15.63%, Bing 3.93%, DuckDuckGo 0.41%, Yandex 0.33%, Yahoo! 0.23%.

Seznam launched in 1996, predates Google, runs its own crawler and index, and led Czech search until roughly 2011. It is now a distant second and the trend over the past decade has been downward. Czech domestic panels have generally credited Seznam a little higher than StatCounter does, for the same portal-referral reason that affects Naver, though the gap is nothing like Korea's.

Japan, and a figure worth treating carefully

StatCounter, Japan, August 2026: Google 63.02%, Bing 28.29%, Yahoo! 6.96%, DuckDuckGo 0.74%, Yandex 0.4%, CocCoc 0.35%.

A 28% Bing share makes Japan by a wide margin Bing's strongest large market in StatCounter's data, and it is a figure worth handling with care rather than repeating flatly. Japan's search market has a structural quirk found almost nowhere else: Yahoo! Japan is a major consumer property in its own right and has long licensed its web results from Google, so a substantial slice of what Japanese users experience as "Yahoo" search is Google's index underneath. Referral attribution in a market with that much brand-versus-index divergence is unusually hard to read. Report the figure with its source and month attached, and do not build an argument about Bing's global competitiveness on it.

Countries where Google leads

United States — the one market with a court-found answer

The memorandum opinion of the US District Court for the District of Columbia in United States of America v. Google LLC, No. 1:20-cv-03010-APM, issued in 2024, found that by 2020 Google handled nearly 90 percent of all US general search queries and 94.9 percent of queries on mobile devices, with Bing at 5.5 percent, Yahoo at 2.2 percent, DuckDuckGo at 2.1 percent and every other rival together at 0.9 percent. Those are findings of fact resting on Comscore data and Google's own internal figures, tested at trial. They describe 2020, and the judgment is on appeal.

StatCounter, United States, August 2026: Google 86.1%, Bing 8.92%, Yahoo! 2.74%, DuckDuckGo 1.7%, Yandex 0.34%, Ecosia 0.12%. The two columns do not match and should not be expected to: the court measured general search queries in 2020 on compelled data, StatCounter measures search referrals in August 2026 on a panel. DuckDuckGo is the sharpest illustration — 2.1 percent adjudicated, 1.7% estimated — and the direction of that gap is what a referral-based measure predicts for an engine that strips its own referrers. The US remains Google's weakest large Western market and Bing's strongest, which follows from Windows and Edge penetration rather than from preference.

United Kingdom — regulator-found, not panel-estimated

No StatCounter row is given for the United Kingdom, because a better figure exists. The UK Competition and Markets Authority's 2025 final decision on the strategic market status investigation into Google's general search services found that Google Search accounts for more than 90 percent of all general search queries in the United Kingdom, and that more than 200,000 UK firms collectively spent over £10 billion on Google search advertising in 2024. The CMA's query-share measure is defined to include search-grounded AI queries, so it counts activity referral tracking cannot see.

The same decision fixes the size of the alternative. Bing, the largest traditional competitor to Google in the United Kingdom, holds less than 5 percent of both general search query share and search advertising share. On the money side the CMA found that Google generated more than 90 percent of UK search advertising revenues in every year from 2009 through 2019 — a decade of consistency, derived from data compelled from both Microsoft and Google rather than from browser-panel sampling.

Australia — share and default placement, measured together

The Australian Competition and Consumer Commission's Digital Platform Services Inquiry, Interim Report No. 9, Revisiting General Search Services, is one of very few official documents that separates search-engine share from browser default share for the same market on the same date. For August 2024 it recorded Google Search at 94 percent of Australia's general search market, with Bing at 4.7 percent, Yahoo at 0.8 percent, DuckDuckGo at 0.6 percent and Ecosia at 0.1 percent. Beside that, it found Google was the pre-set search engine on at least 91 percent of mobile browsers and 77 percent of desktop browsers in the same market. Placing those two measures side by side is what turns a correlation into a mechanism.

The report also records DuckDuckGo falling from 0.9 percent in September 2021 to 0.6 percent in August 2024 — a regulator-verified decline that runs against the usual account of privacy engines gaining ground. These are August 2024 figures and specific to Australia; check for a later interim report before citing them as the current position.

Germany

StatCounter, Germany, August 2026: Google 88.49%, Bing 5.84%, Yahoo! 1.57%, Yandex 1.56%, DuckDuckGo 1.2%, Ecosia 1.02%. Germany is the strongest market for privacy-positioned and mission-positioned engines in absolute terms — Ecosia is Berlin-based and DuckDuckGo does better here than in most of Europe — but even combined they are under three points.

France

StatCounter, France, August 2026: Google 88.76%, Bing 5.13%, Yahoo! 2.55%, Ecosia 1.4%, Qwant 0.83%, DuckDuckGo 0.79%. Qwant is French, is one of the few European engines running any crawling of its own, and appears in its home market's top six. That is close to the ceiling for a national European alternative engine.

India

StatCounter, India, August 2026: Google 97.76%, Bing 1.23%, DuckDuckGo 0.47%, Yahoo! 0.37%, Yandex 0.1%, PrivacyWall 0.03%. India is the most Google-concentrated large market on Earth, a consequence of Android's near-total share of Indian smartphones, and the sixth row is a reminder of how little is left once the first row is taken.

Brazil

StatCounter, Brazil, August 2026: Google 87.5%, Bing 10.28%, Yahoo! 1.73%, DuckDuckGo 0.22%, Yandex 0.22%, Ecosia 0.03%. Brazil is one of the few large markets where Bing holds double digits on an all-platform measure.

Vietnam

StatCounter, Vietnam, August 2026: Google 94.72%, CocCoc 4.44%, Bing 0.47%, Yahoo! 0.27%, DuckDuckGo 0.04%, Yandex 0.02%. Cốc Cốc runs its own Vietnamese-language crawler and is the clearest surviving example of a national challenger engine. Its share has fallen a long way: StatCounter-derived desktop figures cited on Wikipedia show a 2017 peak around 32.5% and 11.41% by August 2024. Do not compare those desktop numbers to the all-platform figure above — the apparent collapse is largely a change of denominator as Vietnamese internet use went mobile, not a crash.

Europe — a harmonized official measure, of something slightly different

Eurostat publishes the only cross-country European series on search behavior that is official statistics rather than a commercial panel. Its 2025 digital economy and society tables record that 71.81 percent of individuals aged 16 to 74 in the European Union used the internet to find information about goods and services in 2025, ranging from 94.53 percent in Norway and 90.60 percent in the Netherlands down to 34.34 percent in North Macedonia, with Czechia at 85.49 percent. The spread of sixty points between the top and the bottom of that range is far larger than any difference between engines within a single country.

Read the definition carefully before using it. This measures using the internet to find information about goods and services, which is a proxy for commercial search behavior and not a direct measure of search engine usage, and it says nothing about which engine was used. It belongs beside the share tables as a measure of how much searching happens in a country, never inside them.

The Korea conflict, printed in full

This is the one national figure on the site that cannot be given as a single number, because the two credible sources disagree about which engine is first.

StatCounter Global Stats, South Korea, August 2026, all platforms: Google 46.6%, Naver 43.71%, Bing 5.44%, Daum 0.88%.
InternetTrend, March 2026 monthly average, as reported by The Korea Times on 16 April 2026: Naver 63.8%, Google 28.7%. The same report records Naver's daily share peaking at 70.6% on 28 February 2026 and gives an annual trend of 58.1% in 2024 rising to 62.9% in 2025.

The gap is about 20 percentage points and the ordering is reversed. Both figures are honestly produced; they measure different things.

StatCounter counts outbound referrals. Naver is a walled garden by design: a very large share of Naver searches are answered by Naver's own Blog, Cafe, Knowledge iN and AI Briefing surfaces and never leave naver.com. Those searches emit no referral, so a referral-based measure cannot see them. StatCounter structurally undercounts precisely the behavior that makes Naver what it is.
InternetTrend measures query share from Korean web-log panel data and does capture searches that terminate inside Naver.

Do not average them. Do not silently choose one. Note as well that the measurement months differ — August 2026 against March 2026 — so the two are not strictly comparable even before the methodology is considered. A page that quotes only StatCounter will be wrong about Korea in a way Korean readers spot immediately; a page that quotes only InternetTrend will be inconsistent with every other country on this page. The correct answer is both figures, both sources, both dates, and the reason.

Why national figures are less reliable than worldwide ones

StatCounter's worldwide series draws on a very large sample. Its national series draw on whatever slice of that sample happens to sit in one country, and the quality varies enormously. Three failure modes recur.

Thin or skewed panels. The China table is the clearest example — the Yandex row at 11.41% is not believable, and where one row is contaminated the others are too. Smaller markets can show the same problem less visibly.
Portal and walled-garden effects. Any engine that answers queries inside its own properties is undercounted: Naver severely, Seznam mildly, and Yahoo! Japan in a different way again because its results come from Google.
Brand-versus-index confusion. Referral attribution records the brand a user arrived from, not whose index produced the ranking. Yahoo! rows worldwide are Bing results under a syndication deal; Yahoo! Japan rows are Google results. Two identically labeled numbers can mean different things in different countries.

Privacy engines are undercounted for a fourth reason: DuckDuckGo, Startpage and Brave Search strip or shorten the referrer on outbound clicks by design. An engine that deliberately does not tell the destination site where the visitor came from is an engine that a referrer-reading tracker will underreport. The court record supports the direction, if not a correction factor: the US court found DuckDuckGo at 2.1 percent of US general search queries in 2020, above StatCounter's current US reading. The measured share is a floor; nobody can say how far above it the true figure sits, and no number should be invented to fill the gap.

Regulator and court figures avoid the panel problem entirely, because the data is compelled rather than sampled, but they come with their own limits. They exist for only a handful of markets, they are published years after the period they describe, and confidential material inside them is released as bands rather than as point estimates. They are better evidence about the past; they are not a live series.

StatCounter publishes the country filters used throughout this page free and at monthly granularity, so any StatCounter figure above can be re-checked at source in about a minute.

When this page was last checked

Every StatCounter figure here was retrieved from gs.statcounter.com on 17 September 2026 for the August 2026 reporting month, the most recent complete month published at that date. The InternetTrend Korean figures are for March 2026, reported April 2026. The historic Cốc Cốc desktop figures are StatCounter-derived numbers cited on Wikipedia and are not comparable with the all-platform figures.

The compelled figures carry their own dates and do not move monthly: the US court findings describe 2020 and the judgment is on appeal; the CMA's final decision was published in 2025 and its advertising figures cover 2024 and the years 2009 through 2019; the ACCC interim report describes August 2024; the CNNIC report counts users as of December 2024; the Eurostat table is for 2025; the NAVER fact sheet covers the second quarter of 2026.

National shares move more than worldwide ones, because a single country's panel is smaller and noisier. If you are reading this more than a few months after September 2026, treat the tables as a dated snapshot rather than a current state, and re-check at source before citing.

Frequently asked questions

Which countries does Google not dominate?

On StatCounter's August 2026 data, only China (Baidu 59.26%) and Russia (Yandex 70.35%) have a non-Google leader. South Korea is contested: StatCounter puts Google first at 46.6% but Korean domestic measurement puts Naver first at 63.8% for March 2026. Czechia (Seznam 15.63%) and Japan (Bing 28.29%) are the only other markets where a non-Google engine holds double digits. Where regulators have measured instead of panels, Google's position is stronger still — more than 90 percent of UK general search queries per the Competition and Markets Authority in 2025, and 94 percent of the Australian market per the ACCC for August 2024.

What is Naver's market share in South Korea?

It depends entirely on who is measuring. StatCounter reported Naver at 43.71% for August 2026, behind Google's 46.6%. InternetTrend, the Korean domestic standard, reported 63.8% for March 2026, ahead of Google's 28.7%. The gap exists because Naver answers most searches inside its own properties, producing no outbound referral for StatCounter to count. Both figures should be quoted, with dates and sources. NAVER Corporation's own 2Q 2026 fact sheet reports KRW 1,447.2 billion of advertising revenue in that quarter, which is a measure no panel dispute touches.

What is Baidu's market share in China?

StatCounter reported Baidu at 59.26% of Chinese search for August 2026, but that panel is not reliable — the same table shows Yandex at 11.41% in China, which is not a credible figure. StatCounter's Chinese sample is small, skewed toward internationally facing sites and distorted by VPN traffic. Domestic Chinese estimates put Baidu higher. For the size of the market rather than its division, the China Internet Network Information Center counted 878 million search engine users as of December 2024, 79.2 percent of Chinese internet users. All sources agree Baidu's share has fallen well below its early-2010s levels.

Why does Bing have such a high share in Japan?

StatCounter recorded Bing at 28.29% in Japan for August 2026, its strongest large market by far. Japan's market is structurally unusual: Yahoo! Japan is a major consumer property that has long licensed web results from Google, so brand and underlying index diverge more than elsewhere, making referral attribution hard to read. Report the figure with its source and date rather than generalizing from it — in the one market where a regulator measured Bing directly, the UK, the Competition and Markets Authority found it below 5 percent of general search queries.

What search engine is used most in Russia?

Yandex, with 70.35% of Russian search referrals in August 2026 per StatCounter, against Google's 27.79%. Yandex runs its own crawler and index and has since 1997, and gained share after Google's Russian operations were disrupted from 2022. Rambler, frequently listed as a Russian search engine, falls below StatCounter's reporting threshold and has served Yandex's results rather than its own since 2011.

Are country-level market share figures accurate?

Less so than worldwide ones. A national figure draws only on the part of StatCounter's panel located in that country, and coverage varies enormously. Three problems recur: thin or contaminated panels, undercounting of portal engines that answer queries internally, and confusion between the brand a visitor arrived from and whose index actually ranked the results. Treat national tables as indicative. Where a court or a competition regulator has published shares compelled from the companies — the United States, the United Kingdom and Australia — that figure is the better evidence for the period it covers, though it is usually several years old by the time it appears.

Which European country has the strongest local search engine?

Czechia. Seznam held 15.63% of Czech search in August 2026 per StatCounter, against Google's 79.33%. Seznam launched in 1996, predates Google, runs its own crawler and index, and led its home market until around 2011. France's Qwant is the next closest at 0.83% domestically in August 2026. No other European engine holds a comparable position in its own country.

Do privacy search engines have more share than the figures show?

Almost certainly yes, and nobody can say how much. DuckDuckGo, Startpage and Brave Search deliberately strip or shorten the referrer on outbound clicks, which is exactly the signal a referral-based tracker reads. Their measured shares are therefore floors rather than estimates. The court record points the same way: the US District Court for the District of Columbia found DuckDuckGo at 2.1 percent of US general search queries in 2020, above StatCounter's 1.7% US reading for August 2026. The Australian regulator, measuring the same engine differently, found it falling from 0.9 percent in September 2021 to 0.6 percent in August 2024. Publish the measured figure with the caveat attached and do not invent a corrected number.

Search engine market share by country