When Good Enough is Not Enough Part I
Thursday, May 14, 2015One of our large enterprise customers is a Fortune 500 insurance company with a global footprint. When they implemented their initial chargeback system, their objective was to consolidate IT infrastructure costs in a central billing system and individually charge each department that was using the resources. The initial plan was to start the process by charging users based on a fully-loaded cost of storage per gigabyte. The initial number was $3,200 per gigabyte.
This was a good starting point. It allowed them to at least charge back their IT costs by gigabyte. And they aren’t alone. This is a frequent way that enterprises start a charge back program. In some cases they implement show back first to let the business units have some visibility into what their usage looks like before chargeback begins while in other cases they move right to chargeback.
Once APTARE began discussing the chargeback solution with the customer, some of the weaknesses of the one price fits all approach became obvious. First, while charging back by storage was a good way to estimate the actual usage of many other resources (like backup resources and fabric), the reality was that as the size of the environment increased the “back of the envelope” guesstimates become less accurate and more risky. By charging business units flat price based on the amount of storage they requested rather than what they actually used, the company was ignoring all the non-production environment resources that are associated with their storage requests—such as backup, security and disaster recovery.
Of course the company knew that there could be a large variance in the amount of backup volume, tape, or port switches used by a host or application, but tracking this information on a granular level was simply cost prohibitive. We refer this type of manual effort as “dudes and spreadsheets” because in many cases IT professionals have to export large amounts of data and then manipulate it extensively to get the desired analysis completed. Simply put, the manual effort required to get to the host level is so massive that the enterprise does not even consider it a viable option for chargeback purposes. This is how a lot of enterprises end up charging back IT infrastructure at $3,200 per gigabyte – it’s “good enough.” But is it really? We’ll dig into that assumption in our next blog post to see if we can shed some light on the thinking behind it and why that logic may not be the best way to think about this problem for the organization.
Continue on to Part II