Tackling Total Cost of Ownership

Share on LinkedIn0Tweet about this on TwitterShare on Google+0Share on Facebook0

Guest post by Sean Sell

Total Cost of Ownership (TCO) analysis is intended to determine the lifetime costs of acquiring, operating, and changing something. It is primarily used to make apples to apples comparisons in large capital investments, and, if done properly, can bring out hidden costs of ownership. The term is regaining popularity largely because of the questions about cloud computing.

TCO to Support Vendor Claims

Although the concept has been around forever, TCO has become important in IT since the 90s. Google the phrase “TCO Mainframe” and there are countless listings of tools and studies validating, interestingly, both HP’s and IBM’s claims of cost superiority in their mainframe vs. server battle. In the 90s in an effort to protect its Windows revenue, Microsoft commissioned several TCO white papers and studies (none definitively conclusive) to offset Linux claims of being “free.” Of course, the problem with these approaches is that the rooster is counting the hens. This tactic of using TCO analysis to win favor is being revived by the cloud vendors - the Cumulux Cloud Economic Justification tool is just one example.

Developing Your Own TCO Methodology

If you use external benchmarks, you know that it can be difficult to map the author’s framework to your own. It is the same with vendor TCO models. So, we recommend that you develop your own, so that you know whats in it and that it meets your specific needs. With it as your reference point, you can then map in individual vendors and do custom analysis for yourself.

A typical breakdown of the three buckets of costs that would go into a TCO model are:

Acquisition Costs Operating Costs Change Costs
  • Hardware
  • Software Licenses (one-time)
  • Network
  • Storage
  • Facilities Renovations (one-time)
  • Procurement Costs (i.e. RFP)
  • Operations/Maintenance Personnel
  • Training
  • Power (Cooling)
  • Physical Space
  • Backup/Recovery
  • Security
  • Disaster Recovery
  • Release Management
  • Audit
  • Insurance
  • Design, Development, and Testing
  • Scaling
  • Upgrading
  • “Greening”
  • Decommissioning
  • Disposal
  • Extraction

The biggest challenge with TCO modeling is the allocation of the operating cost to the asset. However, with the advances in resource management tools and bots to track usage, the problem has become more understood, automated, and less daunting.

Don’t Forget the Change Costs

Where our clients are less consistent in how they account for costs, or where they overlook them altogether, is within the change bucket. The costs needed to make an asset operational continue to be underestimated by 20-30% on average. The costs to enhance or upgrade an asset are often neglected in a TCO analysis. And the costs to eventually decommission the asset are usually left out altogether. (Maybe this is because retiring 100% of an established asset isn’t done too often?)

One of our insurance clients introduced new CRM and lead management systems but paid little attention to user adoption sensitivity. To scramble to keep the users happy, the legacy systems were phased out much more slowly than planned. As a result, the actual benefits fell short of expectations. While communications and managing change might seem like a soft element of TCO, ignoring it can create more hard costs than necessary.

So, as the cloud debate continues in your organization, or you just need some solid TCO analysis, be sure you are armed with your own model, not just the one that is provided to you by your friendly neighborhood salesperson.

cc licensed flickr photo shared by AgentAkit

Share on LinkedIn0Tweet about this on TwitterShare on Google+0Share on Facebook0
  • Mozzello

    I totally second your comment about the importance to factor in the change costs.
    I’ve seen several studies that indicate that looking at the total cost of ownership for an application during it’s entire life-cycle, 80%-90% of that cost is actually for maintenance after the first release goes to production. That means that if you spend $1M over a 5 year period, you’ll have used between $800,000 to $900,000 in maintenance. What this means is that the largest chunk of the IT budget is actually being gobbled by maintenance costs, leaving very little for delivering new applications…