Hacker News Reader

Comment by Semaphor | Hacker News Reader

Just to add some anecdata, we have both Google Analytics (if you accept the GDPR tracking request…) and our own internal statistics based purely on useragents. Here are some percentages for November (this is Germany, generally far higher FF usage)

Ours:

Chrome 37.4% Firefox 24.7% Safari 21.1% Edge 7.5% Opera 2.6%

GA:

Chrome 39.3% Safari 31.5% Firefox 11.9% Edge 9.9% Samsung Internet 2.9%

For us, GA is undercounting FF by almost 13 percentage points, over 50%.

View on HN

if google analytics is undercounting firefox in a way that may make websites to drop firefox support, i suppose this is a case for antitrust?

because google is using a near monopoly on analytics to bury a competitor in another segment

It's not GA being malicious in any abnormal way. Some Firefox users are blocking some GA functionality, so GA can't count some Firefox users.

Of course this heavily depends on niche. In some European countries Firefox is at almost 20% marketshare

I mentioned that:

> this is Germany, generally far higher FF usage

But the interesting part is not how much FF has here, but how much GA undercounts it.

Yeah it's interesting. I wonder if it's done on purpose or if it has to do with the way bots/crawlers are counted

See parent. The default in FF is to shim the GA script

Also GA requires opt-in in the EU (even if some sites try to be clever and illegally make it harder to refuse than to accept) whereas presumably their first-party tracking is done in such a way consent is not necessary (e.g. sufficient anonymization that a data point can not be correlated with the user it originated from). Presumably FF users are more likely not to consent by default to begin with.

My programming blog (https://thecodist.com) sees Chrome 52%, Safari 27%, Firefox 9%, over six months covering a fair amount of the world. I use Plausible. I never found GA to be very reliable.

Reply on news.ycombinator.com