How To: A Deeper Dive Into Service Level Management: How to Configure SLAs

By Chris Mahanes

There is no question that with the rise of digital technologies, there’s been a corresponding rise in demand for better customer service, which is not only impacting how companies must respond to their customers, but how IT must now respond to their colleagues, or in the case of Managed Service Providers, their clients.

We know how important Service Level Management is to you and your organization. Provance ITSM is PinkVerified for the ITIL® Service Level Management process (along with 10 other processes). And in this first post of our Deeper Dive Into SLA Management blog series, I’ll discuss how to configure an SLA for Provance ITSM.

But first, it’s time for some definitions:

Service Level Agreement (SLA): Response and resolution targets that has been agreed to by two parties, either with an organization and their customer or between internal users.

Operations Level Agreement (OLA): Interdepartmental response and resolution targets agreed upon to meet the KPIs of the SLA.

Underpinning Contracts (UC): Response and resolution targets between an organization and their vendors.

This post will solely discuss SLAs and will not go into the functionality concerning OLAs and UCs, but we will discuss these later in the series.

SLA Types

To understand how to configure, first I’d like to list for you the various SLA Types:

Configuration Item
Asset Specific
Contact Specific
User Specific
Service Specific
Customer Based
Timezone
Country Specific

The SLA Type determines the Provance ITSM SLA engine process. Note that an SLA can also be defined as a combination of the types outlined about, such as a specific SLA applied to a specific customer for a specific service. In other words, customers can have different SLAs for the same service.

How to Configure an SLA

I’m now going to show you how to configure an SLA based on the Time Zone SLA. First, let’s start by clicking on the Service Management section in the top-level navigation. Click on SLAs. Then, click New. Now, let’s say we have to configure our support hours to be in the Pacific time zone. So, as you can see in the screenshot below, I’ve named our new SLA “PST Time Zone” and I’m choosing from the drop down menu the SLA Type “Time Zone SLA”.

Continuing through the form, you can see in the screenshot below, I need to define which process this SLA will relate to, and in this instance, I have decided it will be Incidents, however, I can choose from Service Request, Problems, Change Request or Incident. I’ve added Chris Mahanes as the Service Level Manager and I’ve left the level to be blank, but you can specify Gold, Silver or Bronze. I’ve chosen the Time Zone to be Pacific. I then chose the start and end times for the week as well as the start and end times for the weekend.

Now that I’ve filled in the business hours, I can choose the Service Level Targets (see the screenshot below). Here, I chose the response time based on Priority (I can specify the time either as hours or minutes). If you only have one response time I would simply add the same values I have for P1 for P2, P3 and P4. Under Resolution Times, I chose 2 days (16 hours) for Priority 1. For Priority 2, I chose 24 hours, for Priority 3, I chose 36 hours and for Priority 4, I chose 40 hours. The next section I have filled out is Threshold for Risk and Threshold for Breach. Threshold for Risk is based on a percentage of the total resolution time. So, here I’ve added 90%. When we begin to approach the risk of breach, I’ve chosen 75%. This is important as our SLA engine will calculate when we are approaching risk and when the record is about to breach. This metric is captured in the SLA Reporting dashboard to ensure the organization is meeting their SLAs.

Next, I want to set up break times. You can apply breaks to certain priority levels. In the screenshot below, I’ve set up a lunch time break as an example. Here, I’ve decided for Priority 1 the lunch break won’t apply as we have an extremely limited time to resolve the issue.

Next, I’ll set up the hierarchical escalation (see in the screenshot below). For Stage 1, Escalate to First Level (%), you can add in the percentage that will escalate it and then choose the level (0 through 4) you’d like to escalate it to. Then you can choose to assign it automatically, although I’ve chosen Do Not Assign Automatically. There’s a drop down menu of choices including Assign to Ticket Owner’s Manager and Assign to Service Owner. And you can make the same choices for Stage 2. Then you can set up your notifications. The first field is to do with whether you want to allow your customers to see that it’s been escalated, while the other fields let you automatically notify the owner and manager, the team or an additional team.

Now, how does it calculate you ask? Based on the priority that is assigned our SLA engine will calculate at run time what the Response and Resolution Date & time should be. Let’s say we want to review KPI based on a certain account. Navigate to an account and change your form to Provance Analytics. To provide Modern Service Management we need to be able to aggregate how we are meeting our customer’s need from an Organization, Service and agreement perspective and react to those changes efficiently. In the screenshot provided below, we provide this information with a simple form change. We can see from incidents being logged against a certain customer and then review how we are performing per service and by each agreement. As an account manager, if our support fell under a certain threshold we can proactively investigate why we are not meeting the needs of this particular customer.