Beware the fake subscription app

I miss software. I miss buying a tiny CD swimming in a big box of air for $49.99, installing the program on my PC, and not worrying about paying for the product again.

Today, thanks to "The Cloud," consumers are being sold subscriptions to products that should really be one-shot purchases.

Many services deserve to be subscription-based. But not all. And while you can finagle a justification to charge a few dollars a month for almost anything, that doesn't mean you should.

Here's are some guidelines to see if the app or service you're thinking of subscribing to deserves your recurring fee, or if you're being ripped off:

1. It is a real subscription service if it is based on a constant flow of new content or data. You buy a book, even if you may read it a hundred times. But you subscribe to a newspaper. The expense to produce content on an ongoing basis is why you keep paying for it. Likewise, some apps, like antivirus applications, require constant updates, and it makes sense to pay for them like you would a periodical.

Note that code updates and bug fixes are not reasons to charge a subscription.

2. It is also a subscription service if there's a high bandwidth or network cost to running it. If you're storing hundreds of gigabytes of data on someone else's hard disks, or transferring gigs a day of video, then you should not be surprised if you're asked to pay for the service. Bandwidth costs money, and so does managing storage. A hard disk in the cloud is not a one-time cost for an online storage company, by the way; disks fail, and the bandwidth to move backups between data centers (for redundancy) is not free.

However, storing token amounts of data is pretty close to free and should not be used as a reason to charge a subscription.

3. It might be worth a subscription fee if the app can be accessed from anywhere, from any device, and all your data is always there. Many products that synchronize data between devices do so by storing the data in the cloud, so they meet the second test for a subscription service. But synchronizing or data sharing doesn't necessarily mean the vendor has to support a cloud infrastructure. Crashplan is a backup service that can send data from one PC to another PC; Microsoft's Windows Live Mesh synchronizes folders between computers directly (unlike Dropbox, which stores your data in the cloud and syncs it down to your devices from there). Crashplan's computer-to-computer backup service and Live Mesh are both free.

4. It's not really worth a subscription fee if the best justification for it is that the product provides "recurring value" or, as I recently saw, "a solid ROI." That should be true for most goods you buy, shouldn't it? You don't pay a subscription fee for your car if you continue to drive it once you make the last loan payment. And just because you can continue to use software doesn't mean it should carry a recurring fee (unless you're talking about an enterprise application that gets you a technician on-call 24/7 to keep your installation running; but we're considering consumer apps here). Even Quicken, versions of which only work for three years before the company "sunsets" them, isn't technically sold on a subscription basis: you can continue to use the code forever. It's just the server link that goes belly-up.

I get the feeling that some services are sold as subscriptions simply because their primary interface is Web-based, or because data is stored in the cloud and not your local machine. In other words, they don't really have to be Web apps.

This article is based, in fact, on a pitch I got for a Web service that carried a $9.95-a-month charge. It looked like a useful product, and it had an ongoing value. But it did not carry much of an ongoing expense for the vendor. Ten years ago it would have been a box on a shelf at CompUSA, overpriced at $19.99. Now it's six times that for a single year of use, and if you stop paying, you lose all access to the product. The Web lets vendors charge a chutzpah tax for what is, essentially, boring old software.

A fair fee for a $20 software app that's become a Web service? Let's say $2.49 a month. Amortized over 24 months of use, that's still a lot more than the standalone app, but if the service has lasting value, over the years the user will save money instead of having to buy major upgrades every few years; and the longer customers stay with the service, the more reliable the vendor's income projections will get, making the business, and possibly the product, better.

You really don't have to fall for the fake subscription pitch. So many great apps and Web services are free or freemium (free with good functionality until you decide to pay for additional features that, hopefully, pass tests 1 and 2 above). Hundreds of thousands of mobile apps are sold outright for just few dollars. Buy these apps. Programmers do deserve to get paid for their work, but fair is fair.