Even their competition is doing the same thing -- Reunion.com And it also demonstrates my point - the exact thing people are worried about with subscriptions is that sites doing subscriptions will gouge their users.
Reunion.com is $1.33 per month. Yes, that's cheaper but it's so vastly more than what most people are willing to pay as to give subscriptions a bad name.
While I am not privy to the financials of any of these sites, it's been my experience that many sites that put up subscriptions do so in a fairly arbitrary manner. That is, they don't actually do any analysis of how many subscriptions they need at what price.
In my upcoming Avault article, I talk about this at length. Essentially sites have to first figure out what their cost per 1000 users. For WinCustomize, it's around $12 per 1000 users presently. Therefore if we wanted WinCustomize to be profitable on its own, we need to bring in say $15 per 1000 users (currently it does around $6 but we can write off $3 to $4 per 1000 users as marketing expenses).
The next thing they need to do is look at how many core users there are and how many of them they expect to sell to. Classmates.com gets probably around 15 million unique visitors per month. 5% (core users) of that would be 750,000.
If their costs are higher than $12 per 1000 users then they're doing something wrong (our bandwidth use per user is where much of our cost comes from, these sites don't have those issues but instead get their costs due to heavy promotional costs, extra servers and IT costs).
Right now, I suspect they are NOT getting $12 back per 1000 users. I'd love to get the inside scoop on their financials (are they publicly traded?). They have some compelling features but the price is too much. They're way off on the price curve (so is reunion.com). Let a user pay $25 for a LIFETIME subscription and then they have a much more attractive deal. I wonder if they've even done any re-subscription analysis. I imagine hardly anyone re-subscribes (or maybe once every 10 years). The point being, they would likely vastly increase their sales if they had a flat rate (or a much more reasonable per month rate).
Hence where the problem comes in - when these guys inevitably bomb out, the media is going to cover it as an example of why website subscriptions can't work rather than it being a failure of pricing analysis.