I hope some folks may be able to help us with some questions regarding Application TCO.

1.  Does anyone have a best practice or example for setting and measuring progress against application cost reduction goals, for purposes of (a) financial management and/or (b) IT application owner accountability? 

2.  Any examples or metrics around what 'good' application cost management looks like?

Thank you!

Hi Kathy.

Great questions.  I believe what you are referring to is around our Application TCO Process... which forces more accountability of Application Spend on your App owners as well as increases collaboration between your App owners, project owners, tower owners, infra owners to help optimize the application stack.  The benefits of running an Application TCO process on a regular basis are:

1. Reduce time and effort required to calculate TCO
2. Repeatable process can be performed monthly, enabling trend analysis and tracking remediation impact
3. Single repository for data required to perform App TCO Analysis
4. Visibility into TCO highlights opportunities and directs subsequent analyses to impact TCO
5. App owners with direct access to reports equals faster decision making
6. Increases collaboration between App Owners, IT/IT Finance, and Business Units
7. Standardize process & improve efficiency.

If needed, we can have someone walk you through the App TCO process to give you a better understanding of the reports and dashboards that help enable this analysis.

"Good" application cost management metrics will be different depending on the "type" of application.  Ie... Mission Critical vs. Business Essential, etc.  Also looking for trends associated with optimizing App Spend across the infra layers... things like:

1. Does my application have the right mix of Physical / virtual ... can you move more to virtual / cloud?

2. Can you optimize storage? How much Tier 1 storage do you really need? Can you re-tier storage based on environment (DEV, STG, PROD, etc)

3. How many critical incident tickets was associated with my App this month? IS there a trend? Is there an issue with the stack because there are too many incidents overall

All of these examples are ways of ensuring your App Owners start becoming more accountable for the FULL Stack of the technology... this is the Total Cost of Ownership.

Hope this helps.

Hi Kathy,

To build on Kevin's points above, there are a number of common App TCO related goals that we've observed.  These are usually situational - i.e., there is a specific objective, condition or hypothesis that is driving the need for timely and accurate App TCO.  For example:

• Understanding an app's cost per user, or other "cost per" (e.g., cost per transaction, account, $ of revenue). Knowing the App's TCO helps you understand it's contribution to the cost and profitability of delivering a service.
• Distribution of app spend.  Kevin mentioned one common one above: Mission critical vs. Business Essential vs. Internal vs ... - these categories will be up to you but knowing the TCO of each app and it's classification can support an objective like "60% of app spend should be on Mission Critical apps."  I don't know if there are any universal or industry-specific best-practice goals, but setting one for your own organization allows you to track the change over time, and course correct if you find that the goal is too aggressive or there were unintended consequences.
• Application comparison or retirement.  If you have overlapping or redundant apps, understanding the true cost for each can drive decisions about whether to retire one or more of those apps.  This is another case where "cost per" may be more enlightening than absolute cost, and if you've modeled fixed and variable costs you can also estimate the impact of shifting users from one app to the other.  In other cases, the findings will be more straightforward, like if you have 4 different versions of your ERP software and one has a lower cost to run, lower ticket volumes, etc.

The way you phrased your question, though, is key and very telling.  Choosing metrics and targets needs to start with a goal in mind.  App TCO on its own is useful in understanding the cost drivers of an App and for tracking changes in cost over time, at different volumes of usage, etc.  A goal of "thou shalt reduce app spend by x%" may get handed down from leadership (In which case tracking total cost by app is rather useful!) but often the goal should be more directed; e.g., reducing the cost of consumer facing apps to improve profitability for a particular business line, or tracking TCO as an app is migrated from internal to cloud infrastructure. 

Again, echoing Kevin's comments, Apptio's App TCO provides this information and can expose levers for making adjustments. 

Hopefully that helps, and others will chime in here as well about the goals and metrics they've successfully used for managing app costs!

Hi Kevin and Andrew, thank you both for your responses.  I apologize that it's taken me a while to respond back.

Your responses are very helpful and have allowed me to adjust my line / way of thinking about application costs.  We are currently exploring some of the ideas you presented.  For example, understanding the cost for applications based on their environment, ensuring classification of applications based on criticality and increasing collaboration between App Owners, Infrastructure and Business, just to name a few.  We are well on our way, but I wanted to see what other interesting benefits other Apptio clients had achieved.

Your perspectives have helped in guiding us to defing the most meaningful and useful metrics for us.  We are also in the process of defining the value (business capability) for business applications so that our conversations change with our Buisness Leadrers we demonstrate the value of IT and their consumption of IT services.

Thanks very much!

Glad to be of assistance.  If possible, could you share your progress here with the rest of the TBM Connect community?  I'm sure other people/orgs are grappling with similar issues and hearing first hand descriptions of what worked, what didn't, lessons learned, etc. would be very useful.