XP Wow, Huh?
About a year ago, there were supposed to be about 400 million Windows XP users worldwide. Microsoft now says that more than 40 million copies of Windows Vista have been sold. That means that there should be about 360 million XP users and 40 million Vista users, very roughly speaking. About 11% of Windows users today should be Vista users.
According to my site’s stats, about 84% of my blog’s almost 25,000 visitors (that doesn’t include spammers or RSS feed numbers) in the past month were Windows users, so I have a pretty good sample pool to study. In Google Analytics, you can break down visitors by OS version. Over the past month, five months after the release of Vista, only 4.4% of this site’s visitors are Vista users. In contrast, 4.6% are Windows 2000 users.
Obviously, Microsoft is exaggerating things a wee bit. In fact, my stats may over-emphasize Vista users, if anything. Vista is likely more popular in the English-speaking world than elsewhere; it probably has not permeated China very deeply, for example. And as most of the visitors to this blog are English speakers, I should be seeing more than the average number of Vista users. Now, you might guess that since I write posts which are hostile to Vista, that might skew things; however, out of all Google visitors, only a few percent come from searches for Vista keywords, not enough to skew the results, and among Google visitors, the stats on who uses what versions of Windows remains consistent with overall visitor stats.
As I explained here, Microsoft has been caught cooking the numbers. Ever since early on in the Vista publicity cycle, Microsoft has been claiming amazing sales numbers–but those numbers included all the purchases of XP computers from late October, 2006, based on the idea that everyone who bought XP after that time got a coupon for a free copy of Vista. It doesn’t matter if they never used the coupon. In fact, a good number of those last-minute XP buyers were buying XP computers so they could avoid buying Vista computers. Sales of RAM were not nearly as much as were predicted due to Vista adoption, causing a glut in the market. While there was a post-Vista jump, sales weren’t as spectacular as predicted, nor as lackluster as thought before Vista was released.
More evidence of the numbers-cooking is that Microsoft is only counting their distribution numbers to retailers–not the actual sales to customers. Both this and counting XP purchases for three months prior to Vista naturally give an inflated view of VIsta adoption. The fact that Microsoft won’t release specific sales details which would back up or refute their general claims lends itself to the thought that Microsoft is not being fully honest here.
But the numbers that you see in the real world are more telling than anything else, and the numbers I’m seeing are not so hot for Vista. Apple saw a larger percentage of its users adopt the Tiger OS in the first month than Microsoft has seen its users adopt Vista in five months.
One more thing: yesterday, I attended a meeting at my school where we discussed what computer software the school uses in the labs. I was asked whether we should upgrade to Office 2007 to replace our aging Office 2000 software; since students will be using 2007 more and more often, and since the interface and file type is markedly different, I said that we should. However, they didn’t even ask about upgrading to Vista, and when I brought it up, they were obviously not inclined to do so.
But let me ask, to the 5% or so of my site’s visitors who have upgraded to Vista: what’s your experience? Is it positive, negative, or neutral? How does it compare to XP? Was it worth upgrading?
I’m sorry to say that your visitors are in no way representative of Vista usage statistics for a number of reasons. The primary one is the sample is way too small (even 25,000 is quite small when you’re comparing it to the installed base of Windows users).
Beyond that, your blog is frequented by a lot of expatriates living in Japan who are less likely to be running the latest versions of English software (because it is hard to obtain English PC software and hardware and unlikely they’d upgrade to Japanese versions if they are English speakers). In fact, your viewers from Japan are the least likely to be running English-speaking Vista because of the logistical problems of an upgrade.
Also, you have noted that your blog’s visit stats drop off greatly on the weekend and accurately speculated this is because people do most of their browsing at work. Companies are even less likely to upgrade to Vista because it’s costly for them both in terms of the licenses, training new employees, and updating proprietary software and buying new hardware sufficient to run a pig like Vista. There could be a lot of people using Vista at home but not at work who would be unlikely to be represented in your statistics.
Finally, you’re staunchly anti-Vista and pro-Mac. Why would anyone come to read you badmouthing something they’re embracing? You’re not likely to lure the type of audience that uses Vista. Your site has little to offer them.
In fact, I’m assuming that about 4,500 of your visitors use Macs since you said there are 25,000 visitors and only 20,500 are Windows users. That’d make 18% of your visitors Mac users, wouldn’t it? You can’t conclude that 18% of people use Macs because 18% of your visitors are on a Mac OS anymore than you can use your visitors to make conclusions about Vista. It’s simply not a representative sample.
I’m sorry to say that your visitors are in no way representative of Vista usage statistics for a number of reasons.Actually, I dealt with at least half of the reasons you list already, in the post. If you read fully, you’ll note that I detailed a sample of Google visitors; this in response to your points:
1. My readership is mostly expatriates
2. My pro-Mac and anti-Vista stance
Since the sample of people who Googled my site was not made up of Japan expats but from readers anywhere on the web, and since, as I noted, only a few percent of the searches dealt with Mac/Windows issues, and since the breakdown of the platform users matched the greater sample almost precisely, I believe that I was able to comfortably dismiss both objections. The people who came in via Google were not specifically expats (the more popular searches were by eyelid twitchers and Harry Potter fans, but there was a wide variety), nor were the looking for a Windows or Mac bias.
Another point you made was that a lot of people blog from work. I am not sure what evidence there is to substantiate that the new OS market is business-light and individual-heavy, especially considering that the business version of the OS was released two months earlier, and so had more time to pick up customers. In fact, don’t many large businesses have a subscription contract to get their OS, which would allow for immediate upgrades at no extra cost? And since the new hardware requirements would require most personal users to fork out as much additional cash per machine as businesses, I don’t see that individuals could afford the switch that much more easily than business users.
However, for fun, I tallied the totals only for weekend visitors. Vista users rose only half a percent, from 4.5% to 5%. Interestingly, the number of Windows 2000 users dropped by half, and was absorbed fully by the XP users; all other users remained close to the same. This would make sense as businesses were more likely to have Windows 2000 than individual users–but it still says virtually the same thing about Vista, that the numbers for it are about half what they should be, were MS telling the truth and not fudging the numbers.
Your last point is one I can’t speak to well, as I never took Statistics and therefore don’t know what kind of margin of error that I am dealing with. However, seeing as how the numbers line up so closely even when testing what should be radically different populations within the group studied, I have a bit more faith in the numbers.
As for so many visitors being Mac users, the Google sample works out to 86% Windows, 13% Mac, and 1% Linux–and I do not see that as being necessarily wrong. The methods for measuring market share are dubious at best, and there has long been contention that Macs have had 10% of the market or more. Numbers that say the Mac share is now between 6% and 7% (having risen dramatically, more than 50%, over the past year) come from an organization that measures visitors to its own site–one more likely to be frequented by Windows users, which they do not discount for in the tally. It is quite possible that Macs have about 10% of the market share, which is about what you’d have if you subtracted all Mac- and Vista-related search queries to my site from my Mac visitor tally.
Here’s an interesting thing: on Amazon’s software site, in the top 100 best selling software titles, Windows XP Home is #8. XP Pro is #22. Vista doesn’t show up until #45, with the Home Premium version. After that, the Vista Ultimate upgrade and the XP Home upgrade ring in at #68 and 69, respectively.
I think it says something when your outdated 6-year-old product is way outselling your shiny brand new one. Yes, Amazon’s list is quite unscientific, but the margin is so great that I think it’s worth noting. And I bet that MS is not making public sales figures of XP–or at least not very loudly. (Or are they still reporting all XP sales as if they were Vista sales?)
I think it frightens me more that there are still people out there using 95 and ME…
I use vista – since I just bought a dell – and mozilla doesn’t make an os.