The best SLA measure?

by on 06/05/2011

The Problem

Service Level Agreements (SLAs) are notoriously difficult to monitor.

So what if 27% of issues were resolved over SLA in period? Is that a good result or a bad one?

It is bad if the business perception is that calls are not being resolved in a timely fashion. If that 27% is made up of calls that were held up by a third party, by the business itself or a delay is agreed (perhaps because other calls have a higher priority), then it is completely acceptable.

There is little or no correlation between the number of calls over SLA and how good a job the supplier is doing!

SLAs are often seen as a one way thing, showing how well the supplier is doing. In fact, they represent a triangle of responsibility, with third parties and the customer themselves as well as the supplier playing their parts in the resolution of issues.

So what can you use instead?

This chart provides the single most useful way of analysing the service a supplier is providing.

pie chart showing causes of calls over SLA

To produce this chart, the supplier should:

  • Track the time from when a call was logged to when it was closed.
  • I.e. do not exclude the time the call was with a third party or back with the customer.
  • As a call approaches its target SLA resolution time, discuss with the customer whether this is acceptable or not. If it is, then set the cause to “agreed”.
  • When a call is closed that is over target SLA, assign a cause if it was not agreed as acceptable with the customer.
  • Produce the chart shown.

Then, when you sit down with your supplier to discuss the support service they have been providing (you do do that don’t you?), you have information available to inform that conversation.

If the cause is “agreed”, then that’s a good thing.  If the cause is “customer delay”, then the customer needs to take action. And if it is “lack of resource” or “third party”, then the supplier is responsible and either needs to recruit / assign more resource or manage their own suppliers harder.


Of course, there is more that can be done.

It is important to look at trends over time, not just a snapshot position. So, is the trend getting better, staying about the same, or getting worse. How many calls are being raised, and how many by the same group of users. Etc. etc.

But if you want one chart to start the discussion, then this is a great place to start.

Leave a Comment

Previous post:

Next post: