Second in a Ten Part Series
By Chad Greenslade
I have often been asked about my lessons learned in implementing an IT Service Management (ITSM) tool. Below is the second in a ten part series examining my ITSM lessons learned. I hope that these lessons help you on your journey to ITSM nirvana.
Lesson #2: Avoid “Other”. When you’re going down the path of configuring your ITSM modules (e.g. Incident, Problem, Service Request, Change, Release), there are various meta-data elements you’ll be asked to configure. When a user creates a new Incident record, for example, they will usually be prompted to select the Service being impacted, and possibly the Configuration Items (CIs). In some deployments, the user may also be prompted to select the Category, Sub-Category, and Item values to be appended to the service record. In some ITSM deployments, these meta-data can be uniform across all record types. In others, separate and distinct meta-data may be able to be applied to the different ITSM records.
In any case, you’ll want to avoid giving anyone in the organization the ability to select “Other” as a valid value. If you allow users and technicians to append “Other” to a service record, invariably this will eventually become abused will result in reporting discrepancies in your tool.
The advent of “Other” in the reporting tool arises from a short-sighted or incomplete strategy, so instead of allowing the user to select “Other”, do your homework and determine what the user is really attempting to accomplish. If you allow a user to select the Service being impacted by the Incident, there should be no “Other” service listed in your Service Catalog. Likewise, if you allow the user to select the Configuration Item (CI) being impacted by the Incident, there should be no “Other” CI listed in your CMDB.
Removing “Other” from Category, Sub-Category, and Item drop-down lists can be a little more daunting, but not insurmountable. Most configurations of Category, Sub-Category, and Item that I have witnessed in production have been ad-hoc, short-sighted, and not valuable. Further, I have seen reporting built on top of this flawed implementation approach. My professional opinion is that the combination of Category, Sub-Category, and Item should distinctly identify the configuration item, and the aspect of that CI that is being impacted.
We can all agree that the only service items that exist in an IT environment are hardware and software. Let “Hardware” and “Software” be the only choices for “Category”. Sub-categories under the “Hardware” category would be entries such as, “Data Network”, “Desktop PC”, “Laptop PC”, “Messaging”, “Printers”, “Scanners / Imaging Devices”, “Server – Intel”, “Server – Unix”, “Telephone Network”, and “UPS”. Sub-categories under “Software” would be entries such as, “Antivirus”, “Business Applications”, “Data Files”, “Data Network”, “Database”, “Desktop PC”, “Desktop Publishing”, “Development Tools”, “Infrastructure Tools”, “Laptop PC”, “Messaging”, “Middleware”, “Operating System”, “Printers”, “Reports”, “Scanners / Imaging Devices”, “Security Applications”, “Server – Intel”, “Server – Unix”, “Telephone Network”, “UPS”, “User ID Administration”, and “Web Browsing”.
Now you may have noticed that some of the sub-categories that I listed for “Hardware” are duplicated in “Software”. Take “Printers” for example. There is both hardware and software that goes into a printer, and each could potentially fail and / or require service. You want to be able to report on each of these items separately. Lastly, the “Item” choices would most closely mirror the actual CI setup in your CMDB. Now you may be asking yourself, “Why do I need to have an “Item” entry if I already have a CI entry?” The answer is simple; it’s so that you can more granularly report on exactly what is affected by the ITSM record. For example, if I just choose “Dell Color Laser 5110cn” as the CI affected by the Incident record, I don’t know if its hardware or software that’s the problem. By applying the appropriate Category, Sub-Category, Item, CI, and Service I can accurately report the true nature of all ITSM records across the environment.
Chad Greenslade studied Information Systems at the University of Texas at Arlington and graduated Magna Cum Laude.