December 28, 2004
ok, ok, ok, here's the graph
Pretty much since the day we released Firefox 1.0 people have been asking me to post a graph of the downloads and not just occasional postings of the total download number.
Well, unfortunately, and unlike with the PR, I haven't been recording the download numbers with any consistency. So here's what I've done. I've scoured all the little scraps of paper, post-its, backs of envelopes, napkins, and other garbage sitting at my desk and then I culled through all of my blog posts and any other posts and articles out there on the web that mentioned download totals and from all of this data, I was able to reconstruct a farily reasonable representation of our download trajectory since the release.
It's not perfect, but it's also not far off and it's a pretty sweet line.
Posted by asa at December 28, 2004 01:09 PM
Would like to see that compared to how many *downloads* IE has had... ha- not including ones the user didn't even know they were downloading.
Yet there's still no reliable stats... grr... lol- If you need someone to check the stats... I'm available... haha.
z, actually I think not. All of the press leading up to the ad probably helped to keep our curve as steep as it was for the month before the ad, and if you look closely, you can see that the curve turned back up around the 15th or 16th and stayed up some until about the 24th when it returned to the pre-ad slope.
I think there's been a significant impact from the NYT ad. It's pretty crazy to try to get a specific number out of it, but if you tried and were to extend the pre-ad slope through to today, I'd estimate that we'd be hundreds of thousands of downloads short of where we are now -- and that's not even taking into account all the downloads that resulted from the press about the ad campaign well before the ad ran.
It's also important to remember that every person you reach through something like a NYT ad is a different sort of person than you get from all the early adopter, "web savvy" people who found Firefox online. Reaching new audiences is important and I think we've done some of that with the Ad. I also think it was a great way to celebrate the release and to reward a lot of people who helped spread the word about Firefox 1.0 through donations and other means.
So, that would mean we paid maximum ~$1 per download, if we paid 100,000 and got at least the same amount of DL's. Not bad really.
My cursory glance at the graph shows roughly 1.5 million downloads a week (not counting the steep growth in the first week). Not bad! I really do think Firefox is on the verge of taking back the web. I think the techie crowd is an extremely important one to convert, and trends are showing them coming to us from the dark side in droves. Why are they important? Because they tell all their 'computer illiterate' friends to use Firefox, especially when they have to spend a few hours cleaning up the mess around an IE crater. Haven't we all had to clean up at least one friend's computer in recent months? And haven't we all told those people to use Firefox to avoid the mess in the future?
After a bit more analysis, I've been able to make some weekly estimates (from the same data as the chart above) and here's what I found.
| week | downloads |
| 1 | 3,570,000 |
| 2 | 2,130,000 |
| 3 | 1,700,000 |
| 4 | 1,700,000 |
| 5 | 1,450,000 |
| 6 | 1,650,000 |
It looks like the sixth week, which started the day after the ad ran, had a markedly better total than the fifth week which ended the day the ad ran.
Like I noted above, all of this has to be taken with a grain of salt because I don't have great data to start with, but I believe that the bump seen in the weekly graph is significant enough to be real. We were on a declining trend week over week and after the ad ran, we reversed that trend and put in two weeks that were better than before.
z, there's more to it than just downloads. God knows how many people now actually know of the existence of Firefox! It has been featured in many major news sites in the Netherlands, and I'm on the verge of converting my dad (he's really stubborn :P). Many people now support the cause, because they see it _is_ better. And luckily most web developers support it because they see developing for Firefox is much easier. But NYT ad - no flop. Publicity counts big time.
Asa, are there any statistics about the localized versions of Firefox?
$1 per download not bad? Well, buy CDs and give away them, it is cheaper. Luckily, it more likely to be a half or a third of that value, which is still not impressive, but might turn to be acceptable.
Downloads are one thing. But I presume FF is on cover disks etc. How about usage figures? I publish shareware and, during December, 13.9% of my visitors use Firefox. Microsoft IE averaged 75.2%.
I've definitively converted a few of my neighbors here in Barcelona to firefox. This, out of purely self interest, to avoid future winxp reinstalls and other unpleasant things.
I think the holiday season is a good time to get our parents to use firefox. Well, I got my mom in cuba to install it, the download isn't too bad for cuban modems!!! Happy new year everybody!!!
Any chance of getting the raw numbers used to plot that graph?
Firefox 1.0 = good , good , good
Had firefox since day one and have been happily converting all around me except for a couple of idiots who claim IE is better. I'll just wait till they next ask me to fix a major crash and gloat...
Lots of press coverage here in the UK both major newspapers and extensive reviews in the computer mags. Also have it on a few cover disks as well.
For the last few of months my site stats have been showing 1 - 2% of visitors using firefox but from 1st December its grown rapidly till its now hitting 15% on the main site and 12-14% on the others.
Interesting jump in the figures in week six. If your figures are correct that looks like a jump of 200,000 over the previous week. I know many marketing people who would kill for a jump like that whatever the cost.