Statcounter is a web analytics service. Our tracking code is installed on more than 1.5 million sites globally. These sites cover various activities andgeographic locations. Every month, we record billions of page views to these sites. For each page view, we analyse the browser/operating system/screen resolution used and we establish if the page view is from a mobile device. For our search engine stats, we analyze every page viewreferred by a search engine. For our social media stats, we analyze every page view referred by a social media site. We summarize all this data to get our Global Stats information.
We provide independent, unbiased stats on internet usage trends. We do not collate our stats with any other information sources.No artificial weightings are used. We removebot activity and make a small adjustment to our browser stats forprerendering in Google Chrome. Aside from those adjustments, we publish the data as we record it.
In other words we calculate our Global Stats on the basis of more than 5 billion page views per month, by people from all over the world onto our 1.5 million+ member sites.
By collating our data in this way, we track the activity of third party visitors to our member websites. We do not calculate our stats based on the activity of our members alone. This helps to minimise bias in the data and achieve a random sample.
In July 2022, our global sample consisted of 5.3 billion page views (US: 1 billion); 885 million of these weresearch engine referrals (US: 142 million); 177 million of these were social media referrals (US: 58 million).
Our search engine stats are based on what we call search engine referrals, not the number of searches performed.
This is how a search engine referral is tracked by us.
- A person opens a search engine in their web browser or clicks/taps into a browser search box.
- They enter their search and look at the search engine results page.
- They click/tap on a website from the search engine results page and load the desired website.
If that website has the statcounter code installed, that search will be tracked as a search engine referral by us.
If that websitedoes not have the statcounter code installed then no search engine referral will be tracked by us.
If that person never clicks on any website result then no search engine referral will be tracked by us.
If that person clicks/taps on multiple websites from the search engine results page, each website that has the statcounter code installed and is loaded will be tracked as another search engine referral by us.
We've researched Comet from Perplexity, Dia from The Browser Company and Chatgpt Atlas from OpenAI. All three browsers are identifying as the Chrome browser. If they identify as their own unique browser in the future, we will track their usage. For now, we can only record their usage under Chrome.
We have no way of measuring the number of queries performed in bing chat. However, we also don't measure the number of queries to regular search engines like bing or google either. Instead we tracksearch engine referrals.
i.e. If you go to a search engine and do a search for anything and you click on a website result, we'll record that click as asearch engine referral if that website had the statcounter code installed. It's the click to a website that we measure, not the actual search queries that were performed.
When you do a search using bing chat, and you click on one of the "learn more" websites we can track that as a search referral. So we are monitoring bing chat in the same way we measure the regular bing search engine.
From this data we can see from the statcounter network of webites, that the amount of traffic being sent to websites from bing chat is very, very small. Less than 1/100 of 1 percent.
So from our data we can say that bing chat is not currently translating into enough clicks to our network of websites to change the search share.
Of course you are less likely to click on a source website from bing chat than a regular search, as it is intended to give you the answer rather than have you go visiting websites to find the answer. So that needs to be factored in when using our stats for your analysis.
Yes. Laptops and desktop machines are included in the desktop platform together. We use the browser useragent to determine the platform and there is not enough information contained in the useragent to distinguish between laptops and desktops. That is why we do not have a separate laptop platform.
Please do! AtStatcounter Global Stats we license our work under aCreative Commons Attribution-Share Alike 3.0 Unported License. This means that you are welcome to use any information provided by Statcounter Global Stats, but please ensure to credit us (with a link) anywhere you use our stats/data/graphs.
We publish stats for the following:
Line and bar graphs measure different things and SHOULD therefore differ. Both types of graph offer equally correct/robust information.
A bar graph is discrete and shows total figures. A line graph is continuous and shows trends over time. The different types of graphs are appropriate in different situations. Comparing ONE randomly selected reading from the line graph (where multiple readings are available) with the single corresponding figure from the bar graph is not a correct or appropriate comparison.
Trend graphs show multiple entries over time whereas bar graphs show totals. These are standard, universally accepted conventions.
Please note that this is theStatcounter Global Stats website. We deal with enquiries in relation to Global Stats only and cannot assist with general Statcounter enquiries.
If you have a general query about Statcounter or your Statcounter account, then you should visit theHelp Section ofStatcounter where the team on the Support Desk will be happy to help you. Please make sure to submit a full and detailed query to ensure a prompt response.
Stats are updated and made available at around 13:00 GMT for the previous day. Please note, however, that our stats are subject toquality assurance testing and revision for 45 days from first publication.
Although we make every effort to ensure that all stats are accurate and comprehensive on publication, we reserve the right to correct the stats in respect of any errors/omissions. After the expiration of a 45 day period from first publication, no changes will be made to the data.
You can keep up-to-date with Statcounter Global Stats in the following ways:
In July 2022, our global sample consisted of over 5.3 billion page views. The ten countries with the largest individual sample sizes are listed below:
A full list of sample sizes per country isavailable here.
In July 2022, our global sample for mobile devices consisted of over 3.2 billion page views. The ten countries with the largest individual sample sizes are listed below:
A full list of sample sizes per country for mobile devices isavailable here.
No. Statcounter Global Stats are based on over 5 billion page views per month, by a random sample of people worldwide, to over 1.5 million global websites, covering multiple interest areas and geographic locations.
No, Statcounter does not use any kind of toolbar to collect data.To find out more about our methodology, please visit this link.
No. Our Global Stats track the activity of third parties on our member websites. This helps to minimizes bias in our data and achieve a random sample.
We do NOT base our Global Stats solely on the activity of our members. Our Global Stats are based on the unknown persons who randomly visit our member websites.
We understand that there are several other sources for market share data similar to Statcounter Global Stats. All services differ. For example:
Other service providers publish their stats, we publish ours and people can choose which service or combination of services suits their needs - there's plenty of room for all of us. If you have any specific questions regarding comparing Statcounter Global Stats with another service, please justlet us know and we'll do our best to provide you with fair and accurate information.
There are a number of differences between Statcounter Global Stats and Net Apps. These include:
Please note that we are not connected with Net Apps but due to many recent queries about the comparison of our services, we have published this FAQ. Please also bear in mind that we have never used Net Apps stats and we don't have access to their paid services however, for the purposes of this entry, we've reviewed their FAQs and we've also picked up some information from our users and from online articles. In the interests of fairness and balance, we suggest you confirm all the above with Net Apps directly (and please do let us know if we need to update this FAQ!).
No stats service (even our own!) is perfect and it's up to users to decide which service suits their needs. That's why we try to make our service/methodology/stats as transparent as we can. If you have any information to update/correct/expand the above FAQ, please dolet us know ASAP.
We do not impose artificial weightings on our stats - this is a conscious and deliberate decision. Weighting stats means that the stats are only as good as the weighting methodology used. If the weighting data is inaccurate or out of date, then it renders the data completely incorrect. Further, applying a weighting factor to inaccurate data does not turn it into meaningful information - no matter what weighting factors are applied, the geographical spread of the initial stats is very important. For these reasons, we choose not to weight our data.
We would also have a significant difficulty in choosing and applying any system of weights to our data. For example, how would we handle changes in the weights? Would we rerun and restate all our prior stats? Would we apply a smoothing factor? How often would we update the weights?
By publishing our data without any adjustment for country weighting, we allow our users to re-run the numbers using any weighting system they wish. No weighting system is imposed on them and they are free to take our data and weight it however they wish or not at all. The decision is theirs. All our work is shared under aCreative Commons Attribution-Share Alike 3.0 Unported License to facilitate our users with this.
Important!
Please note that if country weighting is used, it should only be applied to worldwide, continental or similar data where stats for several countries are aggregated. Weighting factors arenot relevant for individual country stats.
We do not weight any of our data. Our users are, of course, welcome to apply weights to the data (or not at all) if they wish. Weighting is not relevant for individual country stats.Read more about our decision not to weight our stats here.
With specific reference to theCIA Internet Users data, some of the difficulties include:
We measure internet usage trends. To accurately measure usage, we have to base our stats on page views (and not unique visitors). Let's look at an example:
Person X uses two browsers. On a particular day, they load one page in Browser A. They load 500 pages in Browser B.
If we based our stats on unique visitors then usage of Browser A and B would both be recorded as 50%. This is obviously incorrect. This doesnot fairly represent the usage of the browsers given that Browser B was used 500 times more than browser A.
Using page views as the basis of our stats means that Browser A will be recorded at less than 1% in our stats whereas Browser B will be recorded at over 99%. In our view, this gives a more accurate representation of actual browser usage.
We made the following video to help explain the situation.
Furthermore, there are many limitations, difficulties and uncertainties with regard to unique visitors.
Do you define a unique visitor based on cookies? How do you handle browsers which don't allow cookies? In this case, each page view is counted as a unique. Or do you use IP address? How do you handle dynamic IPs (such as AOL) where the IP changes with every page view? In this case, each page view is again counted as a unique.
There are also difficulties with deciding how often/when to include the stats for a particular unique visitor.
What is the return period for a unique visitor? Do you reset the counters after 6 hours, 12 hours, 24 hours? Do you count each time a unique visitor visits any site in your network? Or do you only count one unique visit to any site in the network per day? What if the same person uses multiple browsers?
Due to the above uncertainties,no service can provide stats based solely on unique visitors. Instead, the stats will be based on a mixture of page views and uniques. We made a conscious decision not to introduce that inconsistency into our figures hence our use of page views as the basis of our stats.
No. We track both javascript and non-javascript enabled browsers - this is critical in the mobile space but less so in relation to the desktop market.
There are two difficulties with trying to use the stats for a single website to calculate market share information:
The Statcounter Global Stats analysis is based on over 5 billion page views per month, by a random sample of people worldwide, to more than 1.5 million global websites, covering multiple interest areas and geographic locations.
At Statcounter we track over 5 billion page views per month on the Statcounter network of more than 1.5 million websites. We publish a summary of the activity that we track via Statcounter Global Stats.
In the case of some countries/regions (e.g. Antarctica) the sample size of page views may not be sufficiently large to be statistically robust. In this case, it is possible that there may be some bias in the data and the graphs may become skewed by a small number of users.
If you remain concerned, pleasecontact us and we will investigate and respond as soon as possible.
Absolutely. Ourdetect tool can be used to confirm that our useragent detection is accurate. If you identify any problems, please let us know ASAP.
This is a difficult question to answer - to be honest, we recommend you consider our methodology in order to make your own decision regarding how accurate our stats are for your purpose. Here is some information to review:
Pleasecontact us if you have any further questions.
We make every effort to eliminate bot activity. We identify bots and prevent them from being recorded in our stats. We also monitor multiple key metrics in order to identify any potential problems with our data. We maintain communications with many of the main technology providers and we reserve a 45 day window to revise stats if/when necessary. In the event of a change or revision being required we flag this on our graphs by way of a note.
Statcounter Global Stats came about because we decided to publicly share interesting trends that we were monitoring in-house. We literally had a light bulb moment one day when we realised that other people might be interested in these Global Stats too... so we made an interface and published all the stats for free.
Our methodology is very simple and we've purposefully kept it that way - we try to make our Global stats as clear and transparent as possible and we appreciate all user queries, comments and suggestions. If you have any questions for us, please don't hesitate to contact us via ourfeedback form.
To see the list of browsers currently tracked, please select the "Browser" statistic and choose the "Bar" graph option. Next click "Download Data" - the downloaded file will list all browsers currently tracked.
To see the list of browser versions currently tracked, please select the "Browser Version" statistic and choose the "Bar" graph option. Next click "Download Data" - the downloaded file will list all browser versions currently tracked.
This statistic groups together browser versions which are on a "Rapid Release" cycle, whereby the browser gets updated from one version to the next automatically without the user taking any action.
Browsers which use this versioning model include all versions of Google Chrome, as well as versions of Firefox 5+ and Opera 15+
Please note that the significance of a browser is heavily dependent on the time period being examined. For example, Firefox 3.5 (released 30 June 2009) will not appear on a graph of browser versions for January 2009, but may appear separately on a graph for July 2009.
The Statcounter Global Stats graphs adjust to display the most popular browsers for the particular period under analysis. So a browser will be included in the OTHER category if its use is not significant enough to warrant its inclusion as a separate line item FOR THE GIVEN TIME PERIOD.
To find out how to download a full list of all browser versions currently tracked,please visit this link.
Please note that the significance of a browser is heavily dependent on the time period being examined. For example, Firefox 3.5 (released 30 June 2009) will not appear on a graph of browser versions for January 2009, but may appear separately on a graph for July 2009.
The Statcounter Global Stats graphs adjust to display the most popular browsers for the particular period under analysis. So a browser will be included in the OTHER category if its use is not significant enough to warrant its inclusion as a separate line item FOR THE GIVEN TIME PERIOD.
To find out how to download a full list of all Browser Versions currently tracked,please visit this link.
We are always happy to improve our browser detection; to do this do this we need FULL DETAILS of the user agent string.
However, before sending us any details, please review thecurrent list of all browsers tracked. If your suggested browser does NOT appear in the list, then please send us the user agent information so that we can update our detection.
Three browsers are affected by preview-type requests - Chrome, Safari and Opera.
Chrome
Further to a significant number of user requests, we are now adjusting our browser stats to remove the effect of prerendering in Google Chrome. From 1 May 2012, prerendered pages (which are not actually viewed) are not included in our stats.
Some points to note:
Safari & Opera
The "Top Sites" feature in Safari and the "Speed dial" feature in Opera's desktop browser show preview thumbnails of frequently visited or bookmarked sites. These preview thumbnails are refreshed by the respective browsers periodically. Unfortunately, it is not possible to exclude these previews from being tracked. To get a bit technical, this is because the "X-Purpose: preview" header is only sent with the request for the base page. The header is not sent as part of requests for images, CSS or JavaScript that have to be downloaded and executed as part of the preview. With online web analytics (as provided by Statcounter) the relevant header information is not passed so these preview requests can't be detected and therefore can't be removed. Ideally Safari and Opera will change this to ensure to send the "X-Purpose: preview" header with all "Top Sites" and "Speed Dial" HTTP requests, however this is not the case at present.
Certainly not! We track IE, Maxthon and Lunascape separately - while they may use the same rendering engine these are separate and independent browsers. Bundling these third party browsers with IE would simply incorrectly inflate the IE stats.
To see the list of operating systems currently tracked, please select the "Operating System" statistic and choose the "Bar" graph option. Next click "Download Data" - the downloaded file will list all browsers currently tracked.
To see the list of search engines currently tracked, please select the "Search Engine" statistic and choose the "Bar" graph option. Next click "Download Data" - the downloaded file will list all search engines currently tracked.
We are always happy to improve our detection particularly in relation to local/regional search engines; if you would like to suggest a search engine for us to add to our detection, then pleasecontact us and provide a sample referring URL.
We define a mobile device as a pocket-sized computing device, typically having a display screen with touch input or a miniature keyboard.
We track over 5 billion page views per month across the Statcounter network of over 1.5 million websites. We analyze every page view in order to gather as much information as possible. With each page view a useragent string is sent which allows us to determine the the browser and operating system used and also to establish if the page view came from a mobile device. In this way we determine the usage of mobile devices, browsers and operating systems.
We track over 5 billion page views per month across the Statcounter network of over 1.5 million websites. We analyze every page view in order to gather as much information as possible. With each page view a useragent string is sent which allows us to determine the the browser and operating system used and also to establish if the page view came from a mobile device. In addition, we examine the referring string information to establish the search engine used (if any). In this way we determine the popularity of the various search engines on mobile devices.
To see the list of mobile browsers currently tracked, please select the "Mobile Browser" statistic and choose the "Bar" graph option. Next click "Download Data" - the downloaded file will list all mobile browsers currently tracked.
To see the list of mobile search engines currently tracked, please select the "Mobile Search" statistic and choose the "Bar" graph option. Next click "Download Data" - the downloaded file will list all mobile search engines currently tracked.
To see the list of mobile operating systems currently tracked, please select the "Mobile OS" statistic and choose the "Bar" graph option. Next click "Download Data" - the downloaded file will list all mobile operating systems currently tracked.
In April 2012, we performed a Nokia OS detection update. Prior to this, Nokia devices (including some S40 devices) had been grouped largely under Symbian OS. With assistance from Nokia and a number of individual users, this error was corrected on 23rd April 2012 and from that date all Nokia devices are now tracked as Symbian or S40 or Meego as appropriate. When all mobile operating systems are combined,Nokia currently remains the leading Mobile Device Vendor.
We apologise for any inconvenience caused by this oversight and encourage all our users touse this tool to check and/or suggest amendments to our detection. Many thanks to all who assisted us with this matter. As always, we welcome your feedback, so please don't hesitate tocontact us with any questions or comments.
We define a tablet as a portable computing device, larger than amobile device, with a touchscreen interface.
Specifically, the following devices are detected as tablets and account for at least 0.01% of tablet usage share (Dec 2013):
Apple iPad
Samsung Galaxy Tab
Samsung Galaxy Tab 2
Google Nexus 7
Samsung Galaxy Note 10.1
Amazon Kindle Fire HD 7"
Samsung Galaxy Tab 2 7.0
RIM BlackBerry PlayBook
Microsoft Surface RT
Amazon Kindle
Amazon Kindle Fire HD 8.9 Wi-Fi
Asus Transformer Pad TF300T
Acer Iconia Tab A500
Asus Transformer TF101
woPad A10
Google Nexus 10
Samsung Galaxy Tab 7.7
Sony Tablet S
Barnes & Noble Nook BNTV250
Toshiba AT100
Acer Iconia Tab A200
Motorola Xoom
Lenovo IdeaTab A2107
HP TouchPad
Barnes & Noble NOOK BNRV200
Asus Eee Pad Transformer TF201
Motorola MZ601
Motorola MZ604
Acer Iconia Tab A100
Amazon Kindle Fire HD 8.9 WAN
HTC Flyer
ViewSonic ViewPad
Acer Iconia Tab A501
Amazon Kindle 3.0
Samsung Galaxy Tab 7.0
Coolpad Unknown
Lenovo IdeaTab A1-07
Vodafone Unknown
ZTE Unknown
Motorola MZ605
Pandigital SuperNova
Archos 80 G9
Some of the Social Media sites we track include: Google+, LinkedIn, Facebook, StumbleUpon, YouTube, Twitter, reddit, Digg, MySpace, NowPublic, iWiW, orkut, Fark, Delicious, VKontakte, Hi5, Yahoo! Buzz, Vimeo, Mixx, FriendFeed, Hyves, Bebo, Tuenti, Kaboodle, Odnoklassniki...
Note that only the Top 7 Social Media sites appear on our graphs.
To see a list of the social media sites we track, please select the "Social Media" statistic, choose the "Bar" graph option and select your chosen time period. Next click "Download Data" - the downloaded file will list all social media sites tracked in the specified time period.
We are always happy to improve our detection particularly in relation to local/regional social media sites; if you would like to suggest a social media site for us to add to our detection, then pleasecontact us and provide the relevant URL.
We rank social media sites according to their traffic generation capabilities i.e. the amount of traffic they refer to other websites. We do NOT rank social media sites according to the amount of traffic they receive.To find out more about our methodology, please visit this link.
We rank social media sites according to their traffic generation capabilities i.e. the amount of traffic they refer to other websites.
Before 24 August 2011, Twitter referrals in Statcounter Global Stats are understated. This is because many third party twitter clients do not provide referral information. Referral stats were largely based, therefore, on referrals from twitter.com; referrals from many mobile/desktop twitter clients being unavailable.
While Twitter had been testing its automatic link shortening service for some time, the new t.co feature was rolled out on a wide scale in August 2011. You can learn more about the t.co URL shortenerhere.
The use of this automatic link shortening now allows more accurate analysis of twitter referrals. In other words, referring links which were previously unavailable due to the use of third party twitter clients can now be successfully tracked.
From 24 August, we have updated our detection and now include t.co links for twitter. In many regions this resulted in a significant jump for twitter on our Social Media stats graphs.
It should be noted that it is possible for t.co links to be copied from twitter and pasted on to a blog or other website. In such a case, a t.co referral would be included as a twitter referral within our Global Stats despite the referral having come from a different website. This scenario indicates that Twitter referrals may now be slightly overstated, however, we expect such overstatement to be small.
In August 2010, Digg launched Version 4 of its platform and caused arevolt among its users.
In protest, it appeared that many users began tomigrate to rival site Reddit... and this was borne out by aspecial analysis conducted by us at Statcounter Global stats. Due to significant interest in their relevant performances, we published a special stat "Digg vs Reddit" which analyzes the traffic generation capabilities of Digg and Reddit only - no other social media sites are included.
NOTE: As this stat is a comparison of Digg and Reddit only, any line graph generated using these stats must sum to 100% at each point (i.e. the trend lines will be mirror images of each other).
Please see our "Social Media" stat for information on the performance of the other main social media sites.
To view a monthly spreadsheet containing a percentage breakdown of Browser, Mobile Browser, Browser Version, Mobile Vs Desktop, OS or Mobile OS usage for each Country, go to the appropriate link:
Browser:
https://gs.statcounter.com/download/browser-country/
Browser Version:
https://gs.statcounter.com/download/browserversion-country/
Mobile Vs Desktop:
https://gs.statcounter.com/download/mobilevsdesktop-country/
Mobile Vs Desktop Vs Tablet Vs Console:
https://gs.statcounter.com/download/platform-country/
Operating System:
https://gs.statcounter.com/download/os-country/
Search Engine:
https://gs.statcounter.com/download/search-country/
Social Media:
https://gs.statcounter.com/download/social-media-country/
You can set YEAR and MONTH parameters to select a required month e.g. to select February 2012, go tohttps://gs.statcounter.com/download/browser-country/?year=2012&month=2
If no parameters are provided, the most recent month's data will be returned by default.
You can also set DEVICE parameters (desktop, mobile, tablet, console) to select a required device(s) e.g. to select Desktop and Mobile, go tohttps://gs.statcounter.com/download/browser-country/?device=desktop,mobile
If no device parameters are provided, all device will be returned by default.
Data is available from July 2008 onwards.
Data for a particular month will only become available on the 2nd day of the subsequent month.
We don't currently publish rendering engine stats, however, this suggestion has been logged and is being assessed by our Development Team.
We don't currently publish statistics on a state basis, however, this suggestion has been logged and is being assessed by our Development Team.
We don't currently publish traffic evolution stats, however, this suggestion has been logged and is being assessed by our Development Team.
We don't currently publish email client stats, however, this suggestion has been logged and is being assessed by our Development Team.
We don't currently publish flash/shockwave/unity/javascript stats, however, this suggestion has been logged and is being assessed by our Development Team.
We don't currently publish mobile statistics showing device/model/user agent, however, this suggestion has been logged and is being assessed by our Development Team.
We respect your privacy and will never share your email address with any third party.
See why over1,500,000 bloggers, web designers, marketing and SEO professionals and small business owners use Statcounter to grow their business.
Learn More© StatCounter 1999-2026. All rights reserved.