Do you produce something like this each month?
You probably do. Most corporate IT support teams have response and resolution targets for each incident priority.
They’re expressed like this, “We aim to resolve 90% of P2s within 4 hours”. Data from ITSM software is used to create a report showing the percentage of tickets that meet those targets. The report is regularly shared with customers (to show them what a great job IT is doing) and once a year a Service Delivery Manager gets together with the business to review the targets. Although neither party would admit it, both are clueless as to what those targets should be.
Can we all agree to stop this madness?
“Your support SLAs are awesome!” (Said no customer ever).
Time-based Service Level Agreements (SLAs) are a poor way to warrant and measure IT support performance.
Forrester Research found that approximately 60% of IT support teams think they provide high quality support and only 35% of businesses feel they get it. I think time-based SLAs are to blame. They’re bad practice widely accepted as best practice.
Green SLA traffic lights mean nothing. They certainly don’t tell you that your customers should be happy.
Your traffic lights could be green while your customers feel they’re not getting good IT support because:
- Your response and resolution targets are too slack.
- Tickets are being assigned lower priorities than customers need.
- The customer experience is poor (e.g. unfriendly, ticket passed from team to team, poor communication).
- Resolution (or next step) timeframe expectations are not set or well managed.
- Tickets that miss their targets do so by a long way (the incentive for speed is gone once a target is missed).
- SLA breaches are avoided by overuse of the ‘stop the clock’ function and/or by closing tickets prematurely.
When your traffic lights are all green you don’t realise you need to improve service quality because it looks like you’re already doing a good job.
The opposite is true if your traffic lights are red:
- Your response and resolution targets are tougher than they need to be.
- Tickets are being assigned higher priorities than customers need.
- The customer experience is great (e.g. friendly, helpful, great communication).
- Resolution (or next step) timeframe expectations are set and well managed.
- Tickets are only just missing their targets.
- Tickets are allowed to breach SLAs without cheating (stopping the clock or prematurely closing the tickets).
When they’re red, it looks like the only way you can make customers happy is by adding more resources or increasing productivity.
It’s not that we shouldn’t measure timeliness for our own internal use. Rather, the problem comes from using timeliness as THE sole indicator of service levels. It’s a poor proxy for service quality and it drives the wrong behaviour .
Here’s some real examples, overheard in IT departments:
Service Delivery Manager: “We’ve missed our target for the month for Sev 2s. I want everyone to focus on Sev 3s now!”
Service Desk Manager: “I don’t want to automate password resets. They’re really quick to resolve. Without them it’d be harder to meet our SLAs”.
Application Support Manager: “Why would we want to create Knowledge Articles so the Service Desk can do more? If we gave the easy stuff to the Service Desk we’d breach our own SLA targets”.
Crazy but true.
Time-based support SLAs take the focus away from what really counts – delivering a level of IT support that customers are happy with.
What do I recommend? Spend every minute you used to spend on setting, reporting and discussing SLAs on collecting, analysing and acting on customer feedback instead. You’ll get a proper gauge on how happy or unhappy your customers are, and the verbatim feedback will tell you all you need to know about where to improve.