This article originally appeared in the January/February 2015 issue of SupportWorld magazine.
Looking for the perfect dashboard, the one that will tell the story of your support organization is not easy. The road is paved with good intention, bad execution and ugly results. But the need to tell your story and finally get to your destination is real. The goal is to show you how you can build a simple, yet telling, dashboard and how you can use those metrics to tell your story at every level of your organization and to give you the means to understand what is going on in your support organization.
Like most support organizations, chances are that you’re pretty good at collecting information with an ITSM tool and produce some sort of report. Typically, the format is not so great, so you end up putting this information in a spreadsheet. You massage the data until you have something to show at your next meeting. In the end, it gets the job done, which is collecting the information and getting it out there.
There is often confusion between the information we collect and the information needed to make a decision. We all want to show what we have done, but this is not necessarily the right information to make good decisions. The number of tickets opened or closed doesn’t mean much to most people. This is why we often don’t get any feedback on those numbers.
Most of the data we collect and show are a series of numbers having no context or relation between them. Most of which shows some trends to see our numbers evolution, but that’s about as good as it gets. The numbers are not linked to anything, and the information we produce is just that, a number. Who cares how many tickets you closed or how many calls you’ve got? 911 centers get millions of calls a year – they don’t really care about the number of calls because they are staffed to handle it. Instead, they are more concerned about the one they missed because life depends on it.
Have you ever been in a meeting where someone said, "In January, we received 2,950 incidents, and we resolved 2,971. We currently have 406 opened incidents. We have an average resolution time of 6.9 hours. 51.3% are resolved at level 1 and 88.7% within SLA." Outside of your perspective, is there any meaning to these numbers? Are they good or bad? Comparing these numbers to the ones in your head is a start, because alone they mean very little.
And so begins our quest for the perfect dashboard that will answer fundamental questions for your organization (full IT or just the service desk, the principle is the same). These fundamental questions are the ones you get asked every week, like:
- "Are your people really overloaded?"
- "How fast can you deliver this service?"
- "Can we handle this new workload?"
- "Are we efficient?"
- "Are our customers satisfied with our service?"
- "What does IT do?"
Put in a certain context, validated, and balanced against other data in your system, these questions and their metrics will provide you with the information needed, but for that, you’ll need a balanced scorecard.
By showing many aspects of one metric, a balanced scorecard can provide the vision you need and will bring context and relation between the different data.
By showing you many aspects of one metric, a balanced scorecard can provide the vision you need and will bring context and relation between the different data. It will allow you to understand one metric using another. It will also answer your questions while raising many new ones. In the end, you will have the tools you need to make the right decision. As new questions are raised, you will add more metrics and perhaps remove some. You will also need to produce a different scorecard for each level of your organization because upper management does not need the same information as the service desk.
The idea of the balanced scorecard started in the finance world. Numbers needed to be explained with other numbers. Sales alone are nothing if not associated with profit, and that is the basis of a simple balanced scorecard. Pushed further, the balanced scorecard’s purpose is to present information from four different, yet linked, perspectives. For example, you want to balance a series of metrics of a certain type, like customer satisfaction, against a series of metrics of another type, like cost and efficiency, and see how they affect each other.
What follows is a model for you to start with. This model has four perspectives or quadrants:
- Customer Satisfaction, to measure how you are perceived by your customers
- Employee Satisfaction, to see how is your main resources are feeling about themselves
- Cost and Productivity, to understand your performance in the organization
- Organization Maturity, to see how you can sustain growth
Each of these quadrants represents a distinct aspect of your support organization, and yet, each of these quadrants is linked to one another. How can you have customer satisfaction without employee satisfaction or productivity? How can your productivity go up if your maturity stays the same? The metrics we will identify for each of these quadrants will help explain metrics from another quadrant.
Obviously, these metrics are just suggestions, but if you are looking for a place to begin building your metrics, this is a good place to start. There is more than one way of doing this and those fundamental questions we talked about at the beginning will help define your own metrics for these quadrants.
You may not know how to measure everything right now, and that’s alright. Most organization can measure Customer Satisfaction and Cost and Productivity pretty easily, but the other quadrants are not so simple and will depend on your maturity level. Look to them as goals for the future, but don’t forget about them. In my organization we leave certain metrics blank as a reminder.
What drives customer satisfaction? For the most part, we are looking at metrics that will answer questions like, "How fast are we delivering?", "Are we missing calls?", "What is our backlog?" etc. Here is a list of metrics I would suggest:
|Customer satisfaction rate
||Quite obvious, but many organizations don’t have this information|
||A bigger abandon rate may indicate less satisfaction|
|Number of tickets closed at first contact
||A higher number drives up Client Satisfaction and drives down support costs |
|Percentage of reopened tickets
||A lower percentage drives up customer satisfaction|
|Average resolution time
||A faster time drives up customer satisfaction|
|Percentage of tickets resolved within SLA
||A higher value shows your commitment to service quality and a higher satisfaction rate|
Cost and Productivity
How well are we doing? This part of the quadrants tries to explain how efficient the organization is, but also, if there is any room to be more efficient.
||A positive ratio indicates you close more tickets than you open. A negative value is the reverse. You want to stay close to zero; otherwise it may indicate a different problem.|
|Cost per tickets
||A lower number indicates productivity not efficiency |
|Number of customer per Full Time Equivalent
||A higher number can indicate a higher efficiency. |
|Support Personnel Utilization Rate
||A higher number indicates a better utilization of resources and a drives down a lower cost per ticket|
How do our people feel about themselves? This part of the quadrants tries to measure the happiness and the commitment of employees.
|Employee satisfaction level
||A higher number will drive down cost per ticket and increase customer and employee satisfaction|
||A higher number may indicate employee dissatisfaction|
|Average number of training per full time equivalent
||A higher number increases efficiency on the long run and customer and employee satisfaction|
|Percentage of employee evaluation performed as scheduled
||A higher number drives satisfaction up because employees like to get feedback. It shows that we care. |
How are we dealing with growth and how well organized are we? This part of the quadrants gives insight to how we can deal with unresolved issues and changes, and how our process evolves over time to support the organization.
|Team skills diversity
||A greater diversity increases efficiency and satisfaction|
|Score on a process maturity model
||A higher score will drive down cost and increase customer satisfaction|
|Number of changes implemented with success
||A higher rate indicates a more mature organization|
|Number of problems/known errors opened
||A higher number indicates a more mature organization that knows how to handle long-term issues|
In the end, your balanced scorecard will help collect all the information needed to explain just about anything. For example, a high resolution time can be explained by low team diversity, a low process maturity, or a raise in problem tickets. The abandon rate can be explained by the number of FTE. At this point the key is the balance between the information.
This scorecard will help you and your manager understand how any decision affects your client, employee, cost, and organization. Doing this simple scorecard can help you become a world class support organization.
Note: You can get a lot of information on how balanced scorecard works and how to calculate the different metrics from the book The Metrics Reference Guide Vol 1-4, available through HDI, which inspired part of the current model.
Different Levels of the Dashboard
Now that you have the main scorecard/dashboard, you’ll use it to create different views for different audiences. You cannot present and use the same information for your level 1 support and your top management. One will be more interested in the daily operation details, while the other will require more high-level information like cost and efficiency. One may need live data, while the other might be happy with yesterday’s news. Certainly, your CEO or VP of IT does not care how many tickets you did this month, they want to know if you have what you need to do it.
You will therefore end up with different levels of the dashboard. A typical setup will have four levels:
- Level 1 – Operational: These are the metrics closest to the ground (Level 1-2 support), as shown in our example.
- Level 2 – Lower Management: These metrics will be used by the managers of your support groups and they are usually a summary of the level 1 metrics.
- Level 3 – Upper Management: These metrics are for your IT leadership, who will be more interested in metrics around efficiency, cost, and resource management. They want to know what you do and how well you do it.
- Level 4 – Executive: If your organization is not an IT organization, then this is the business level or the corporate level. These people want to understand what you do and if you are in good shape.
In effect, you present the same information you collected in your scorecard, but with a different twist at each level.
As illustrated here, certain indicators can be used at every level. The Support Personnel Utilization rate is certainly a good example. Others will be used to form a new metric or to present the same information differently.
In order to do this, you will need to set goals or threshold for your metrics. For customer Satisfaction Rate, you might decide your target is between 80 and 85 percent. When you present the data at level 1, the actual numbers might be good. At level 2 you may decide to show the number and an indicator (Green, Yellow or Red) to show that you are meeting your target or not. As you go higher, you may need to show just the indicator, since the number becomes meaningless to them.
Building an efficient dashboard is no small task, but the information here can be used as a starting point. As such, don’t forget to start with what you can measure and with raw data that you refine as you go up the ladder. One size does not fit all, and the higher the information, the simpler the information needs to be.
An IT veteran, Benoit has over 20 years of professional IT Service Management experience ranging from process implementation to technical integrations to leadership roles. With extensive experience, education, and ITIL certifications, he brings a refined focus to the intricacies of Service Management.