In fact, the researcher recently published data suggesting that Chrome is actually on a gradual decline while Firefox is on a slow rise.
Whereas in January Chrome accounted for 18.94 percent, that figure fell slightly to 18.90 percent in February, the latest data shows. As for Firefox, inched forward from 20.88 percent to 20.92 percent over the same time period.
Chrome’s January figure also showed a decline from December, when it was at 19.11 percent.
Internet Explorer, meanwhile–which is still the leader–fell from 52.96 percent to 52.84 percent between January and February, Net Applications says.
A Self-Imposed Penalty
Now, these are obviously pretty small changes, so it’s dangerous to read too much into them. Then, too, there’s the fact that rival researcher StatCounter has already declared Chrome’s victory.
What’s particularly interesting, though, is that in addition to any natural shifts there may be going on in usage, there are also a few external factors affecting the data.
First is the self-imposed 60-day penalty Google placed on Chrome rankings in early January following a controversy over a marketing campaign that apparently artificially inflated its results via paid links, the use of which violates Google’s own rules.
Along with that penalty came a halt in Chrome’s upward trend, Net Applications asserts.
The other factor at work here is the prerendering feature Google recently added to Chrome. The Chrome browser is now the only major desktop contender to offer this feature, “which creates unviewed visits that should not be counted in Chrome’s usage share,” Net Applications notes.
Users don’t always go on to actually view the pages that are prerendered, in other words.
In February 2012 prerendering accounted for 4.3 percent of Chrome’s daily unique visitors, says Net Applications, which has announced that from now on it excludes those false visits from Chrome’s desktop browser share.
Holding Steady at No. 2
So is Chrome’s recent loss–however small–due entirely to Google’s ranking penalty and Net Application’s adjustment for prerendering? Could be–but it’s not entirely clear.
Meanwhile, for the time being at least, Firefox is still hanging in there.