Net Applications has just released its latest report on browser usage, covering the months of June and July. The results show the impact of the launch of Firefox 3.5 on the browser market, as well as the steady decline of Internet Explorer 6 and 7 as they give way to IE8 and other browsers.
Firefox 3.5, which didn’t officially launch until June 30th, now claims 4.54% of the browser market, though most of these new users were likely previously on Firefox 3.0, which dropped from 20.03% in June to 16.21% in July. Internet Explorer 6 and 7 continue to see their user-bases fade (which is a decidedly good thing, especially in the case of IE6), and it seems that many (though not all) of these are heading to IE8. IE6 dropped from 30.15% in May to 27.21% in July, while IE7 saw a drop from 31.16% to 23.09% in the same time frame. IE8 has grown from 5.95% in May to 12.46% in July.
Adding all versions of Internet Explorer together, Microsoft’s market-share has remained fairly stable over the last few months — it dropped to 67.77% in April, then rose to 68.32% by June, and has since dipped back down to 67.68%. It’s certainly not going upwards, but it’s no longer hemorrhaging users the way it was last fall, when it dropped from 74.18% to 69.72% between September 2008 and January 2009 (other reports have shown major losses for IE more recently).
Likewise, Firefox is also holding fairly stable, dropping from 23.84% in April down to 22.75% and hovering around there since (for July it was at 22.47%). Any downturn in Firefox’s growth can likely be chalked up to the growth being seen by Safari and Chrome, both of which still represent relatively small but steadily growing user bases. Safari has risen from 3.53% to 4.07% between April and July, while Google Chrome has risen from 1.79% to 2.59% in the same time frame.
This report is also notable because it represents a change in the way NetMarketShare analyzes browser usage. Before now, Net Applications only reported its raw stats, without taking into account how disproportionally measured countries could skew results. Now the site has started to weight traffic data based on the number of Internet users in a given country (the impact of a user measured in China will be greater than one measured in the US, for example). Worth noting are some of the changes Net Applications has seen resulting from the weighting change. From the Net Applications site:
Baidu – Baidu goes to 9% of global search engine usage. Baidu is on a major growth curve, which is affecting the relative share of all other search engines.
Google – Because of Baidu’s growth, Google’s global share is actually going down. This is almost completely due to Baidu and does not reflect the rest of the world.
Apple – Since Mac share in the U.S. in significantly higher than the rest of the world, Mac and Safari share drop in the global reports.
Opera – Opera goes up to 2% in global reports. This reflects the significant share they have in Eastern Europe and Asia.
wait to see what happens when Windows 7®™ comes in Europe with Browser Freedom ™©®
FF 3.5 is literally going back in time. Tab bar is visible even when only one tab is open. Loss of vertical pixels.
And ff4.0 has bigger plans for “omni bar”, it wants to send even website address we type to google in the name of suggestions, just like chrome does. While separate search bar would have sent only search query to goog.
All these things can be turned off you know, even if it wasn’t possible to turn these off, there would be extensions to do so and even if there were no extensions, FF is open source so anybody could turn these things off and put a version online.
Good excuse, that it can be turned off. With a omni bar, you CAN NOT turn off suggestions partially, i want suggestions for search and not for website address i type.
Some google employee becomes ff ui chief and makes omni bar so that even our website info will go to google. I cant personalise ff in every other computer i use.
It will be interesting to see how Google Chrome fairs if/when they can get a developer community behind plug in’s and the polish off the rumored mass syncing capabilities that will make or break the Chrome OS.
Still amazed by Google’s 2.6% market share. On a few fronts, now, they may actually have to do some marketing?
IE8 also jumped arou 4% last month. i think that’s more remarkable, considering at this rate, it will pass IE7 (and Firefox? ) shortly.
Looks like they need to start to breakdown Others in more details. Good growth there? Chrome? Safari?
stuck with 3.0 version on hardy
Netscape increased their market share by over 50% (.41 to .67) since October, 2008? I discount the accuracy of the numbers within this report based on that alone…
I have to agree with Mo Kilma. How could Netscape possibly be *increasing* its market share?
IE 6 is still at the top?!? WTF…
Yes, yes, I know there are a TON of anemic IT departments out there, but please, let the people upgrade to something that won’t kill their PC. Have a heart IT…
The criteria for this stats changed, so Netscape may still be used in countries that are now weighting more.
Netscape identified itself as ‘Mozilla’ in the user agent string. maybe every mozilla browser that cannot be classified otherwise is marked as netscape here
NetMarketShare writes: “For example, although we have significant data from China, it is relatively small compared to the number of internet users in China. Therefore, we now weight Chinese traffic proportionally higher in our global reports. This change produces a much more accurate view of worldwide usage share statistics.”
It measures worldwide share more accurately but the new figure is less relevant to the target audience of the report. For example, if my site has zero traffic from China I don’t have to care about Baidu.
The new approach is a good first step to deal with the data variances from different countries. But there should be more meaningful ways of weighting. The least thing is that the weight should also account for usage frequency differences. Users of certain countries spend more time online than users of other countries and should therefore have a higher weight.
With the ‘Default’ browser having a tough time in Europe (Windows 7), I bet that share is going to make a higher jump… in no time
just loaded FF 3.5.2. they screwed the pooch on that release: doesn’t work properly with google mail, google reader and other sites. i’m on chrome now. fast, light, stable. may become a permanent switch.
I believe the company that produces this data is Net Applications, not NetMarketShare.
NetMarketShare is a service provided by Net Applications
Wow… IE6 is the number 1 browser again! I guess IE7 users are quicker to move to IE8 or other browsers compared to IE6 users.
Unfortunately, these type of numbers being reported makes it hard for web developers to stop supporting IE6 (unless their website’s own statistics are a lot different).
Mozilla has stepped on the gas and is doing a tremendous job with FF. Last week I had posted to my blog that they had usability issues with 3.5.1, but just yesterday they pushed 3.5.2 which took care of most of the issues. This response to user issues will keep driving their lead.
Not being a smart a$$ here, but I am learning my English… Can you hemorrhage slowly?
Dictionary.com:
5. to lose assets, esp. in large amounts.
If you lose large amounts slowly, it seems you would be trickling, not hemorrhaging.
Dunno.., just sounds like a contradiction to me.
Jason,
There is in fact a real story here, but you have missed it.
Net Applications has built a successful business based on the idea that measuring real traffic on a collection of real servers results in meaningful and valuable information. The value of Net Applications’ services rested completely on their ability to convince us that the servers they monitored were in fact representative of what was happening on the World Wide Web.
Net Applications appears to have had a sudden change of heart about what they have been telling us for years. They have decided to hide this real world data (even retroactively) and show us instead some other perspective of internet usage.
One of the most stunning pieces of the puzzle is source of this new perspective. They have chosen data published by the C.I.A. as their new lens into the internet.
I have no doubt that the C.I.A. has much data that could be very useful to those of us working with the internet. However, there is a huge difference between the data that they collect and the stuff they publish. What the C.I.A. publishes is, by nature and necessity, propaganda.
So ask the question, “Why did Net Applications suddenly switch from giving us real-world measurements to regurgitating C.I.A. data?” When you find the answer to that you will have the real story.
Well it certainly is an awkward description that an English Major would probably not use.
Probably threw it in for “headline drama”. “Hemorrhaging” is often used in financial reporting.
An odd way to describe relative market share over a 6 month reporting period.
“IE decline slows” would be more accurate but I would have written “Declining IE market share showing signs of recovery while FF 3.5 jumps to 4.5%”.
But that’s not as punchy and attention grabbing.
The real story here is that IE 6 is the number 1 browser!!! Is every IT department really that lazy? People need to realize how much this browser is holding back productivity.
for my (US-based) site based on google analytics, about 35% of visits are from IE, and 20% of those use IE6. The rest use something higher (IE7, IE8). So only about 7% of visitors to my site are IE6.
People still trust net applications statistics? Their data is flawed and biased. Google it.