If time permits I’ll try to do a series of posts on a topic I’ll call IT Metrics Management. As such it does not differ from managing metrics of any other kind (I would think), but as many other disciplines within IT we’re a bit behind in some areas
You can find many posts and articles on the use of metrics, , Performance Management, IT Management, the majority of these being discussions on various theoretical implementations. Many discuss what to call measurements (KPI’s, CSF’s, OKR’s or just plain metrics), how to organize them in various hierarchies, how to link them to strategy etc. All important topics that needs to be addressed one way or another, but eventually you need to get your hands dirty and create actual measures to start moving your business in the right direction (hopefully).
This usually means you end up in Excel of all ‘tools’, just to ‘start’. The same does all the other people in your organization struggling to get data and make sense of it. Eventually an unknown amount of Excel spreadsheets, interlinked in various ways, make up a spreadmart of loosely couple data sources, and their inherent metrics are made up of various Excel formulas with little or no governance or change control.
Your IT business ends up being run on this spreadmart of Excel sheets, sometimes with the addition of more formal metrics coming out of Finance, HR or other functions. At best this works and any service requests, incidents, changes or problems are handle more or less on an ad-hoc basis. No processes exist for test or verification of metrics and their ‘results’, this is left up to the analyst building the Excel workbook. At worst you live in a constant state of fear of the spreadmart falling apart (or one of the key persons having built the spreadmart leaving your company).
In addition to this no formal repository of ‘receipts’ for the metrics/KPI’s exists anywhere, not for the Excel formulas, and usually not for the more formal ones either (the measures coming from Finance, HR and the like). Even in case of formal BI environments with data warehouses, cubes, formal measures created in MDX/DAX or similar technologies, you very seldom see a repository of definitions that explain the individual measurements in a business oriented language with links to their technical implementations.
In the next part I’ll start to outline an approach on how to start overcoming the challenges described above.