5 Steps to Turn Around A Weak IT Service Delivery Reputation
So, you’ve taken the reigns of a State Government IT shop and are beginning to find that not all is well with your service delivery functions. Maybe your IT managers are repeating the same status report each week on the same system updates or projects. Maybe you’ve noticed your IT technicians appear to be very busy but nothing seems to actually be getting done. Possibly you have just received the that dreaded shot across the IT bow in the weekly executive staff meeting:…”The status of my initiative is the same as last week and the week before that! We’re waiting on IT!”
The root cause here is that either your team doesn’t really understand how to operate efficiently or (more likely) the service expectations of your customers out of touch with what you are capable of delivering; either way here are 5 steps to help you start untangling the situation and establish a method to monitor your progress.
#1 – A task-based timekeeping system for IT personnel – we’re not talking about your average time clock here, you need a system that relates the time each IT staff members expends back to an individual assigned task. If the system “rolls up” tasks into projects that is all the better. There are many good systems on the market and you could take a look at actiTime or MS Project Server 2007 as basic examples.
Why is this important?Time allocated to completion of a task is a critical measure for both process improvement and project management. If your budget is ever under the axe this data will be an outright life-saver. If you are managing projects you simply cannot measure earned value without near real-time task/time data. If you are running a “maintenance only shop” historical task/time data can be a lifesaver. Look at it this way, if you don’t know how many hours it took to complete the last maintenance refresh cycle how could you possibly defend your estimates for the next cycle when you include that cost into your budgets? All of this data rolls into the IT budget – how many times have you been challenged on the “cost of IT” or the cost of maintaining a current systems vs. replacing it? If you are in Florida state government you’ve been challenged every session and lately a few times in between.
The usual complaint when implementing a time system in Florida government is ”We already enter our time into the state time system. I don’t want to waste staff time by keeping two time-sheets. Wewill just keep on estimating how long each task took.” Well everyone – including your executive management – knows that most of your IT work statistics are “estimated” and because of this most will assume your numbers are either padded or just an outright guess. While it may take a while to gather substantial work statistics, in the interim you will be able to measure the earned value of tasks, how assignments are made across all resources (who is getting work piled on them and who is not), and how many times a task is being re-assigned to a different person.
Problem Resolution: Look into time keeping systems. Check your current Help Desk ticketing system first.
#2 – (Florida government) Define your Strategic Services for Schedule IV-C – if you don’t yet know about the TRW Schedule IV-C, here is an excerpt straight from the TRW documentation: “The Schedule IV-C provides a consistent approach and data collection tool to help an agency better align its investments in IT resources (staff, hardware, software, etc.) with the business needs and policies of the agency”.
So is this just another legislative budget tool? Well yes it is a budget tool, but inside it actually contains so much more. Within the Schedule IV-C documentation you will find a “turn key” IT service catalog which covers the common, utility services provided by every IT shop in every state agency. These utility services are referred to as “non-Strategic” services. In addition, you will find instructions on defining the unique, specialized IT services provided by your Agency. These are referred to as “Strategic Services”.
Why is this important? Most IT leaders are keen on defining the IT service catalog as soon as they take ownership of the team because this data provides they lay of the land. In addition, this data is a key input into processes such as IT budget control, IT governance, and service level management. It is nice to know if you are with a Florida State Agency or Department that you do not need to re-invent the wheel! More than half of your IT service catalog is already documented in the IV-C Non-Strategic section. Once you define your Strategic Services you can immediately begin matching the hourly data from item #1 to show how IT dollars impact your agency’s customers (the people) and once you have step 3 in place you can begin organizing an IT Governance structure (if that was your plan) based on the IT services that have the greatest visibility to executive leaders.
Problem Resolution: Go to http://trw.state.fl.us/iv-c_downloads.cfm, download the templates and instructions, start filling them out. Simple.
#3 – Define a Service Request policy and process- probably the most important policy document a government IT unit can have is an Agency-wide policy describing how to request services from IT. Most agencies have policies and procedures on Internet usage, desktop software, or information security but in many organizations there is no policy direction to internal customers of IT describing how to obtain services from IT. This lack of direction leaves many business customers to find their own path to IT and that path usually involves emailing or calling their favorite IT worker, who incidentally is never the IT manager you left in charge.
Why is this important? First and foremost, if you don’t know how requests are coming into IT and from where you can’t even begin to manage them. Without a control point in place to record incoming requests you have no hope of task prioritization or IT governance. If your managers have no idea what direction or at what frequency new requests are likely to come in they cannot forecast workloads or prioritize team work effort. Lack of request organization leads to a reactionary resource management style where work assignments are regularly reallocated (switched) to address previously unknown “agreements” and customer expectations.
At this point most IT managers will point to their Help Desk ticket system or something similar and say that all customer requests are recorded here. The Help Desk ticket system is probably capable of tracking the request but ask yourself “Did I ever tell my customers they are required to ask for service through my help desk?” If not, then the “path of least resistance” your customers have probably already discovered is to go around that help desk unless they just need a simple task like a printer installed or their password reset.
Problem Resolution: Write a procedure at the agency level describing how to request services from IT. Make sure you have the infrastructure (Help Desk) to support the request process first or it will back-fire!
#4 – Develop a RACI matrix for the IT Organization - if you are not already familiar with the RACI matrix it is a table-based diagram which describes the roles and responsibilities of team members in delivery of a process, project, or service. To get the most from a RACI matrix in state IT, remember the following suggestions:
- (R)esponsible means someone who is able to respond. If the task is to restore service to a server then a manager who cannot trouble-shoot a server is NOT able to respond; the server technician is.
- (A)ccountable should really be a single person and this should be the person who is held accountable if the server is not restored. This could also be the technician but it should be the manager; otherwise why do you need the manager.
- (C)onsulted usually represents someone who has answers… if the server goes down I could ask my dog for advice but I might be better off calling Microsoft. On another note, if I am making a change to the server I should “consult” the Change Advisory Board (CAB) on the potential risks.
- (I)nformed someone who needs to know that an incident occurred or the status. In the server outage scenario this could be the manager, the CIO, the customer, or any number of stakeholders.
Why is this important? So you can see who you are relying on to deliver service. You can read the matrix any number of ways but here are my observations on common issues exposed by a RACI matrix.
- Your line staff (technicians) are mostly listed as Responsible and Accountable and your management staff are (I)nformed. This indicates that your line staff are running the show and the results are passed up to you through management – who may be filtering the truth.
- Your Consulted list has more than a few entries per item. You have a bureaucratic system in place; the one Responsible person must cut through a phalanx of consultants before he or she can respond.
- Your Accountable list has more than one entry per item. No one can be held accountable and if challenged finger-pointing will ensue among the “Accountables”. The finger-pointing will likely end with all Accountable parties pointing to the one Responsible party.
- Your Informed list has more than 3 persons or groups per item. Your Responsible person (the one who should be doing the work) probably spends more time generating status reports than doing work.
So what would I consider a good matrix balance?
The line staff (technicians) are mostly listed as Responsible and Accountable, the management staff are mixed across Accountable and Consulted, and the CIO and/or assigned Project Manager are Informed.
This indicates that the line staff do the work and are mostly accountable for their results. The managers are accountable for some results but primarily act as mentors (Consulted) making them generally accountable for the quality of the results provided by the line staff. Status reporting is limited to no more than two sources.
Problem Resolution:Use the service catalog from step #2 and complete a RACI matrix for each system in each service. Use the matrix to sort out your “heroes” from the people who don’t seem to add much value. Remember “heroes” are a higher risk than someone who does almost nothing; sometimes your “slackers” are the way they are because the heroes don’t share information or collaborate. The goal is to have a level matrix.
#5 – Re-think any Open Door Policy that may be in place
This last one is almost a repeat of #3 but honestly if I had not already seen too many “Open Door” policies enacted in state government (with disastrous consequences) I would not even bring up the topic. This may seem counterintuitive but I will make my case and you can decide. Basically an “open door” policy allows any business customer to walk into IT and discuss a concept or solution or maybe just ask for help directly at the source. While this may initially seem to be a good idea to improve customer service, if you have read or taken action on Step #3 then you see the root problem here. This policy actually opens the door for (and condones) customer behaviors which undermine the request process. The point that many IT managers may miss is that any interaction with the business customers they serve results in an agreement to provide service at some level. No one cares that you don’t have the authority to take action or that you weren’t the right person to talk to – they told “IT” and now “IT” owes them a solution to the problem.
The argument for an “open door” policy will usually be heavy on customer service ideals, so to illustrate my point I will use a highly customer service oriented industry; the car dealership repair shop. When you pull into a dealership repair shop you are greeted up front by a repair shop manager and your needs are assessed. You are given an estimated time of completion based on the number of customers in front of you. There is a sign on the door to the shop floor that says “No Customers Beyond This Point”. This “closed door” policy is in place to protect both the customer and the mechanics. Protection against customer injury is obvious but if a mechanic is distracted he or she may make a mistake on the repair job or may damage something. The customer could also inadvertently damage another customer vehicle.
The reason the customer’s don’t seem to mind the “closed door” is that they receive adequate support for their needs from the service manager.
If you have an open door policy it was likely established to address a lack of connection between IT and business by “opening up the doors” and making the IT section more inviting. It will be much easier to maintain a positive IT presentation by closing the doors to the back office, straightening up the showroom (Service Desk), and holding the managers accountable for customer satisfaction.
