In the past I’ve written about DeviceAnywhere; in my blog, and I even got quoted about DeviceAnywhere on Wired Magazine.

In this piece I’ll cover some information to help you determine if DeviceAnywhere is a good choice for you.

DeviceAnywhere provides a service that allows for remote access to real handsets in live networks. I’ve used it in the past, while providing services to other companies, but this is the first time I used it for my own company (i.e. I’m paying for the service). The cost structure for DeviceAnywhere at the time of this writing is as follows:

Packages Monthly Fee  
First Package $200  
Each Additional Package $100  
     
Device Credits    
Credit Level Monthly Fee Additional Hourly Usage Fee*
10 FREE with sign-up of first package $18.00/credit
30 $300 $17.50/credit
50 $500 $17.00/credit

Which in theory when compared to owning many handsets and many subscriptions, seems reasonable. But I have come to realize that while in concept DeviceAnywhere sounds like a great tool, it might not be as cost-effective as you might think.

The Round Up Issue
DeviceAnywhere rounds up device usage to the nearest 6 minutes. Let me provide a comparison between round up usage, given to me by the salesperson, justifying why the round up was OK:

The salesperson said:

“30 seconds on T-mobile it still charges you for a full minute”

Which I follow with the comparison:

“30 seconds on T-mobile it still charges you for a full minute”
“30 seconds on DeviceAnywhere it still charges you for full *6* minutes”

The ratio is not acceptable in my opinion. The $200/month paid to use for (all the handsets on) one carrier for 10 hours, is not really 10 hours, and you can burn that time fast.

The Effect of the Round Up Issue on Application Types
Because of the rounding to the nearest 6 minutes approach of DeviceAnywhere, the service is no good for short-lived applications, such as messaging applications. Just acquiring the handset, receiving a message, releasing the handset, which takes like 1 minute or so, you are charged six minutes. You can burn your 10 hours time very fast, as I have witnessed.

The Packaging Issue
In very rare circumstances (or never) you will ever use all the handsets within one carrier. A more reasonable packaging/cost structure is one that allows you to pick and choose a number of handsets from different carriers.

I would still recommend DeviceAnywhere, if:

  • You have plenty of money to cover the cost (or don’t care about cost) of using DeviceAnywhere. This is $200/month for “10 hours” for 1 carrier, plus $100 for each additional carrier.
  • You have additional time to manage the time required to manage or maximize the time used with DeviceAnywhere. Sounds kind of redundant but it is not.
  • Your type of application fits well for testing under DeviceAnywhere (i.e. it is long-live, in a way the rounding-up to 6 minutes issue doesn’t affect you). For example, messaging applications (or applications that are short in execution) are not good candidates for DeviceAnywhere.
  • You have no way to gain access to handsets you *must* test on, or your test period is short for the handsets needed.
  • It doesn’t make sense for your organization to buy the handsets you are testing against, or your company doesn’t like to buy handsets.

So keep in mind the above. DeviceAnywhere might not be cost effective for your startup. Time to market, and ensuring your application runs well across all devices and networks is very important, but consider testing first using your own set of handsets (set of handsets that should be representative of your target user base). Once you have completed testing using your own set of handsets, if you need to go beyond that, use DeviceAnywhere.

ceo