StatCounter critical (again) of Microsoft's IE browser claims

Back in March, Microsoft announced that it would be using Net Applications as its main source for gauging the worldwide use of web browsers, including its own Internet Explorer. It was also critical of how another company, StatCounter, generates its results. We contacted StatCounter at the time and received a response, which said that Microsoft's arguments in favor of Net Applications was, in its words, "somewhat one-sided."

Since that time, StatCounter has claimed that its results show that Google's Chrome browser has now overtaken IE in the browser wars worldwide. This week, the StatCounter team posted up a massive blog post that went into great detail about how its statstics are, in its opinion, more accurate than those from Net Applications. The blog post is written as an open letter to Microsoft's Roger Capriotti, who wrote the original article about Microsoft's support of Net Applications.

In summary, StatCounter contents that its use of page views to measure web browser traffic is more accurate than Net Applications' method of measuring unique visitors. It states, "Our methodology takes account of how frequently browsers are used and also tracks multi-browser usage by individuals. It is just plain wrong to claim that 'browser usage' is measured using unique visitors."

StatCounter also says it takes its information from three million websites, while Net Applications only uses 40,000 sites. The blog post says, "Furthermore, Net Applications only offer paid web analytics services which may introduce a bias towards profit-making, ecommerce-type sites in their sample pool. Our free service with optional paid upgrades gives us better variety and coverage of all site types."

The blog also takes up the subject of Net Applications' "geoweighting” method of tracking web traffic and adding in the Internet users of specific nations around the world via CIA data. StatCounter says using that information leads to inaccurate results. One example is how geoweighting is uses to track web traffic from Chinese users. StatCounter states, "The CIA Internet User data does NOT distinguish between people who access the internet several times per day and those who only go online once every few months. This means that the data vastly overstates the influence of China ..."

StatCounter's final message to Capriotti states:

Remember Roger - we have absolutely no problem with you or with Microsoft or IE... our problem is with misleading and biased information being published under the guise of a fair and balanced analysis. We have no preference for one browser over another. We are independent and report what we see. At the end of the day Roger, whether you like it or not, we're telling you this: On a sample exceeding 15 billion page views per months to over 3 million global websites IE is not doing very well... Don't shoot the messenger though Roger! We just record the stats. 

Geekwire got a response from Microsoft about StatCounter's blog post. It states:

We stand by our assessment that Net Applications provides a more accurate analysis of the browser landscape. To give just one example, we simply cannot rely on a data source that suggests there are only 23 million Chinese citizens online. (StatCounter global market share for China shows 1.17% of the world internet traffic when reliable data from The World Bank puts that number at 460 million. Other sources like CNNIC puts it at 513 million at the end of 2011, and Internet World Stats puts it at 420 million in 2010). It is also unfortunate that StatCounter decided to respond to our critique of their data by staging a personal attack on one of our employees. We believe in a healthy dialogue and personal attacks in a professional environment are unwarranted.

Source: StatCounter blog

Report a problem with article
Previous Story

Mass Effect 3: Extended Cut to be released June 26th

Next Story

Microsoft is one step closer to opening UK retail location

45 Comments

Commenting is disabled on this article.

I am a web developer and I am working with Microsoft technologies and the only browser make us trouble every single time is IE (IE 7 and 8 mostly). Thanks god we dumped IE 6 Support for a while since that was the worst browser of entire universe for the time. I think IE should embrace the web standards. I know they made a good effort for IE 10 but its not enough as they cannot catch the pace as Chrome does it very well for instance. that's why Chrome is the king of Browsers in both HTML5 Support and Hardware acceleration for Javascript and web Graphics. refer to HTML5test.com and Microsoft's IE test drive and compare browsers for yourself

In the end it doesn't matter. They all serve the same function and perform it admirably. This, like many other things, comes down to personal preference. I went from IE, to Netscape, to Firefox, to Chrome, and now back to IE (10). I use what I believe to be the best for me to use.

I actually do have FF currently installed as there are still some sites IE 10 doesn't work with. There's very few but the two I use to pay bills are among that few. Hah.

Anyway, browsers are not sports teams. I hate the Heat, by the way.

Wyn6 said,
In the end it doesn't matter. They all serve the same function and perform it admirably. This, like many other things, comes down to personal preference. I went from IE, to Netscape, to Firefox, to Chrome, and now back to IE (10). I use what I believe to be the best for me to use.

I actually do have FF currently installed as there are still some sites IE 10 doesn't work with. There's very few but the two I use to pay bills are among that few. Hah.

Anyway, browsers are not sports teams. I hate the Heat, by the way.

AAAAAAAAAAAAMEN!!!!!!

Looking at StatCounter browser version for last 3 month
It is very misleading, as it combines all versions of Chrome (31.5%) yet it splits up other browsers to show Chrome on top .
using their figures and combining the browsers it should be:
IE 33.69%
Chrome 31.5%
Firefox 24.92%
Safari (inc Ipad) 6.65%
Opera 1.66%
Others 2.27%
so using there figures in a fair way , IE is top
As far as the top 5 browsers over last 12 months IE is still top

In summary, StatCounter contents that its use of page views to measure web browser traffic is more accurate than Net Applications' method of measuring unique visitors. It states, "Our methodology takes account of how frequently browsers are used and also tracks multi-browser usage by individuals. It is just plain wrong to claim that 'browser usage' is measured using unique visitors."

If I was to sit here and hit refresh 5 billion times on statcounter.com using Chromium (my default browser) it would vastly inflate Chrome's stats. So how exactly is using page views more accurate than unique visitors?

As for the thing about people who use multiple browsers, they're a minority. I sometimes have to open IE because a website won't render correctly in Chromium, but thats about it. I don't open IE for facebook, Chromium for Neowin, Firefox for Netflix, Opera for email and I highly doubt most people do either.

The other thing is Chrome and Chromium (and even Opera I think) pre-render websites, inflating their numbers.

And you think NetApplications' 40k sites are reflective of total browser usage stats? I don't think so. Hell even StatCounter is only three million, better, but still lacking.

What we really need is a more reliable way to glean these kinds of statistics.

simplezz said,
And you think NetApplications' 40k sites are reflective of total browser usage stats? I don't think so. Hell even StatCounter is only three million, better, but still lacking.

What we really need is a more reliable way to glean these kinds of statistics.

Depends on which 40K sites, I believe top 40K sites will cover like 95%+ web traffic, also even 3 million won't cover exact 100% so it doesn't really matter. It's like measuring linux usage with distros top 15-20 distros cover ~99% of linux market, whereas there are 100+ Distros.

Also personally I think when we talk about user marketshare it should be calculated on no. of people (unique) using that product, however when you have to compute web traffic share (only http/https traffic) then total page views is better option.

simplezz said,
And you think NetApplications' 40k sites are reflective of total browser usage stats? I don't think so. Hell even StatCounter is only three million, better, but still lacking.

What we really need is a more reliable way to glean these kinds of statistics.


why isnt google releasing these stats? they keep track off hundreds of millions of websites with analytics, should be easy.
if chrome would actually beat IE, they'd let us know! dont worry
anyone who has enough data to overthrow IE for the first time in what, almost 2 decades? will shout it from the rooftops

So Microsoft is relying on NetApplications' inflated numbers from China to boost IE and keep it on top. It would be interesting to see what the numbers are if China is removed altogether.

simplezz said,
So Microsoft is relying on NetApplications' inflated numbers from China to boost IE and keep it on top. It would be interesting to see what the numbers are if China is removed altogether.

Yes. Remove a country with nearly 20% of the world's population. Yes ...

Imo, nothing wrong with the statistics. Both make sense! IE has more users using it, as reflected by the NetApplications data. But Chrome's users are likely the more savvy users (who at least know how to install another browser) and they are thus likely to surf the web more, thus giving more web views!

If you ask me, the confusion is what constitutes browser usage: if I surf the web for 2 hours a day in Chrome, should I represent a larger share than someone who just checks their webmail and Facebook for 10min a day? Or the same? Statcounter says the former, NetApplications says the latter.

Kai Y said,
IE has more users using it, as reflected by the NetApplications data.

That's simply not true. NetApplications artificially inflates IE numbers in regions like China by applying esoteric filters and formulas.

IE has lost the top spot and should just accept second place until Firefox overtakes it sometime in the future.

simplezz said,

That's simply not true. NetApplications artificially inflates IE numbers in regions like China by applying esoteric filters and formulas.

IE has lost the top spot and should just accept second place until Firefox overtakes it sometime in the future.


Considering IE9 dominates on Windows 7, and this trend will continue into Windows 8, I'd say it's fair to say that even if they aren't on top now, they will be again very soon.

siah1214 said,

Considering IE9 dominates on Windows 7, and this trend will continue into Windows 8, I'd say it's fair to say that even if they aren't on top now, they will be again very soon.

And who says IE9 dominates Windows 7? It's installed automatically with the ΟS to be sure, but where are the stats to prove your claim?

siah1214 said,

Considering IE9 dominates on Windows 7, and this trend will continue into Windows 8, I'd say it's fair to say that even if they aren't on top now, they will be again very soon.

IE gets launched ONCE on any machine I set up for myself or my family, and that is just to download Chrome.

simplezz said,

And who says IE9 dominates Windows 7? It's installed automatically with the ΟS to be sure, but where are the stats to prove your claim?


Before you troll, at least be knowledgeable. IE9 is not included with Windows 7.

roadwarrior said,

IE gets launched ONCE on any machine I set up for myself or my family, and that is just to download Chrome.

Let's hope you leave them uninformed about spyware.

simplezz said,

And who says IE9 dominates Windows 7? It's installed automatically with the ΟS to be sure, but where are the stats to prove your claim?


its not set to default browser. it just updates the IE already installed, which is IE8 (or install CD's with SP1, which i barely come across, heh)
point is, people who didnt use IE8, will never notice its updated to IE9

siah1214 said,

Here's the stats, next time a quick Bing search will prove my claim, however, I guess you're just lazy:
http://windowsteamblog.com/ie/...on-windows-7-worldwide.aspx

Surprise surprise! NetApplications lol. And no, I wouldn't use Bing if you paid me to.

The whole point of this article is that NetApplications doesn't produce reliable stats. Nice try though.

And to others who are saying IE 9 isn't installed automatically - Yes it is, by a slient update.

simplezz said,

Surprise surprise! NetApplications lol. And no, I wouldn't use Bing if you paid me to.

The whole point of this article is that NetApplications doesn't produce reliable stats. Nice try though.

And to others who are saying IE 9 isn't installed automatically - Yes it is, by a slient update.


You only dislike them because they prove you wrong. That's the very definition of ignorance.

BajiRav said,

Before you troll, at least be knowledgeable. IE9 is not included with Windows 7.

Ah, it depends. Some OEMs include it for you... I've seen Toshibas with newer images have it included, as well as Lenovos. Any systems we sell where I work also get IE9 as well (as well as Firefox), so that if the end user does decide that they would rather use Internet Explorer, they have a version that behaves itself and keeps them out of trouble for the most part.

Chrome's icon scares me. It looks like Google's eye watching over me. That's why I don't use it. Scary icon.

IE's icon scares me. It reminds me of the days when Microsoft prevented users from having a choice. That's why I don't use it. Scary icon.

recursive said,
IE's icon scares me. It reminds me of the days when Microsoft prevented users from having a choice. That's why I don't use it. Scary icon.

hmmm, took a look at ur icon. I call troll.

ctrl_alt_delete said,

hmmm, took a look at ur icon. I call troll.

Anybody who disagrees with you is a troll as far as you're concerned.

recursive said,
IE's icon scares me. It reminds me of the days when Microsoft prevented users from having a choice. That's why I don't use it. Scary icon.

when was this? I must have missed that day.

BajiRav said,

when was this? I must have missed that day.

even in europe lots of computers had both netscape and IE installed (there was a 3rd one i cant remember, was that opera already?)
i used netscape simply cause it looked more awesome to use
altho i remember IE (4, 5? maybe 3)was 2-3 times faster to start, load pages... anything.

Enron said,
Chrome's icon scares me. It looks like Google's eye watching over me. That's why I don't use it. Scary icon.

You still use an icon to launch your browser instead of a hotkey?

The StatCounter post, at least as described here, does not reflect the fact that Chrome prefetches pages as you type in the URL, thus giving you hits for pages that you are not actually visiting. This skews their numbers, which I believe Microsoft took fault with.

I personally use Chrome because it's the fastest, and at least until IE 10 comes out, it is much better than IE in terms of HTML support (IE 9 does not support placeholder text for instance, which was a huge shock and it's definitely a no-brainer that they should have supported before a lot of other features). However, I don't really buy it that Chrome is the top browser because I know too many people not using it compared to those that are.

Denis W said,
They did directly address the prefetching point, and according to their statistics the prefetches had little impact.
Good to know. I can't help but find that dubious though considering that they count each visit.

At the end of the day it doesn't matter at all as long as IE6-8, Fx3 & old Chrome versions are going down, whilst IE9, Fx4+ & Chrome 9000 are going up.

funkydude said,
At the end of the day it doesn't matter at all as long as IE6-8, Fx3 & old Chrome versions are going down, whilst IE9, Fx4+ & Chrome 9000 are going up.

indeed, they're quite evenly matched. all 5 of them (also counting opera/safari). lets keep it like this my sites look identical in all. And use them mixed mostly. like opera, but its such a drag if its idling for more then 20minutes. barely use chrome (or rather ironware) as to often the whole browser crashed. FF hasnt done it in ages to me, and IE9 at worst has a hanging tab.

I believe that Chrome is the majority browser. I've had it as my primary browser since nearly day 1 in 2009 and never looked back. IE9 is terrible by comparison. The Chrome extension library is immense and Chrome's mechanism for adding & removing extensions is very smooth.

mutualcore said,
I believe that Chrome is the majority browser. I've had it as my primary browser since nearly day 1 in 2009 and never looked back. IE9 is terrible by comparison. The Chrome extension library is immense and Chrome's mechanism for adding & removing extensions is very smooth.

yeah the IE9 addon mechanism is awfull, hasnt required a reboot of the browser since... IE6 i think?
for those rare addons i need (only logmein actually) the rest would ever need is build in .
Adaptive adblocker, full popupblocker which FF lacks, fast, shows what addon is possibly showing down, ability to set per site settings for almost everything. what else is needed or even handy for a browser?

but who cares really as long as W3C governs the standard google can't do much to use it's monopoly to influence anything.

neonspark said,
but who cares really as long as W3C governs the standard google can't do much to use it's monopoly to influence anything.

thats not how things work. The w3c only make recommendations. Browser vendors can choose to ignore them, implement their own standards or not update their browser to current guidelines.

the better twin said,
Browser vendors can choose to ignore them, implement their own standards or not update their browser to current guidelines.

Browser developers can implement or cannot implement what they want, and all browser developers do this -- but when Microsoft does it, there is a campaign against them for it.

neonspark said,
but who cares really as long as W3C governs the standard google can't do much to use it's monopoly to influence anything.

Where in the article is there a mention of Google having a monopoly?

neonspark said,
but who cares really as long as W3C governs the standard google can't do much to use it's monopoly to influence anything.

Ya right they implemented WebM/V8 very well and didn't let h.246 rule the mobile and HTML5 videos

nohone said,

Browser developers can implement or cannot implement what they want, and all browser developers do this -- but when Microsoft does it, there is a campaign against them for it.


remember the MARQUESE or something vs BLINK days? IE vs netscape and they refused to support eachothers custom HTML code for a long time