A few days ago I read an article on Information Week’s SMB channel: “Is There an Online Backup Price War?”

Funny thing is, there’s nothing wrong with the article in itself, but it does ignore the biggest issue in the room: greed.

Fact is, most online backup services are designed to make maximum profits for the backup provider, rather than balancing profits with enough performance to provide good enough value to an SMB. So naturally, SMB’s are more interested in local backup. After all, if they actually test most of these online backup services, they are going to see, and see quickly, that the restore performance is so slow that it’s not good enough for a business that loses money when they’re down.

You would think that Kevin Casey (the author of the original article) would know this: after all he just wrote another article stating that 99% of SMBs aren’t interested in online backup because it’s too slow!

WholesaleBackup saw this time and again when we tested various white-label online backup software platforms: they’re all designed to compromise performance at the benefit of maximum profits. Here’s why. In order to minimize both data transmission and data storage, the software packages will conduct differential backups where the changes in each file are computed and placed into a “patch” that can be run against the original file to re-create it. So if you need to store 30 days’ worth of a particular file, these technologies will keep one original file, plus 29 patches, or deltas. The deltas are typically 10% of the size of the original file, or even smaller, so the storage savings initially seems like a great idea.

But when it’s time for a restore, somebody has to pay the piper, and sadly, that’s the customer, not the backup service!

When you want to restore files using this type of system, first the initial file has to be re-created (patched) 29 times, and only then if that process is successful, can the file start to be downloaded. This doesn’t seem like a big deal but consider our testing scenario. Take a 10GB database file, and you will see that you have to re-create 10GB 29 times, so you have to write and compute 290GB *before* the download starts! In our tests, this could take as long as 4 days for some of the cheaper commodity backup solutions aimed at consumers, and as little as four hours for file preparation with some of the more high-end solutions.

The best we saw was 4 hours to prepare a file: that’s simply not good enough.

Our solution is simple: with declining storage costs, and with an understanding that gouging your customers, or shortchanging them on performance is not a good long-term strategy for business, we designed a system that does differentials for each file, but then re-creates each file at backup time so that when a restore is requested, the file is already there, and requires no rebuilding!  That’s why WholesaleBackup is much faster than any legacy solution for restores!

Best Practice: Do test restores before you buy, and another test restores each quarter!

But shame on the customers who depend on these services. Our backup partners regularly tell us that they are getting new customers every day who tried a restore from one of these commodity online backup services, only to find that the restore took days, or failed altogether. This also brings up another key point: unless you recombine and recreate the patched files during the backup, you can’t be sure they will recombine at all. Our testing showed that as much as 20% of the time, files were not recoverable because of bad patching, or errors in a backup halfway through the 30-day retention period. This is truly awful: you won’t know if your backups are any good until you try to restore when it’s too late!!!

You won’t know if your backups are any good until you try to restore, when it’s too late!!!

To Kevin: it doesn’t matter if there’s a price war, because these services aren’t worth any investment if they can’t provide a business-class service!

That’s why our partners who use WholesaleBackup Software for Backup Services don’t see a price war. They don’t compete with these commodity backup services because they offer a higher level of service, using a strategic pricing model for online backup pricing, which can be tested to provide a different class of service. That means a fair profit for them, and a great value for their customers, rather than a cheap service that you will terminate the first time you try a restore!