OLA between Service Desk and Service Level Manager

General discussion on all aspects of the IT Infrastructure Library (ITIL)
Post Reply
User avatar
ITIL Expert
ITIL Expert
Posts: 403
Joined: Sun Mar 09, 2008 8:00 pm
Location: Sunderland

Wed Nov 17, 2010 2:00 pm

Bear with me on this one.....

In my organisation we have a group of people, lets call them Service Level Managers (SLMs).......these guys are responsible for driving service improvement and availability in particular. They are also responsible for producing reports that demonstrate that they have a handle on this and are doing their jobs for their allocated business customer.

They are accountable for this but they are not responsible - the lines of support/delivery are responsible for doing the best job they can to underpin the Service Level Managers who front up to the customers.

Service Levels across the board are not good and the Service Level Managers are struggling to qualify the reasons why. They get their information from tickets logged and since we offshored our service desk to India the quality there has gone down the toilet - we have tickets incorrectly prioritised, sent to the wrong technical groups, left open (accruing downtime) for no particular reason, with no details on them......the list of transgressions goes on.

Our Problem Management team are neither proactive or reactive. They have completed the odd e-learning course on ITIL and proclaim it as their god, but the truth is that they don't raise any Problem tickets unless someone screams from the high heavens and by that time it's too late.

Our technical groups are self-serving, they work in silos and the customer is something that other folks worry about.

Our priority/severity levels are like nothing you would believe - we are a BPO organisation so our internal users are actually being paid for by our clients so our Severity/priority levels are high. 1 production user who is unable/partially unable to work is a Sev2, 5 or more is a Sev1 (our highest Severity). Severity 1 incidents are escalated to part of our Service Desk that I'll call Situation Management for them to manage as an urgent priority. We have dozens of Severity 1 incidents every day so these guys are very busy and again the quality therein is lacking.

That's a flavour of whats going on there......there's a heck of a lot thats wrong and because we're a silo'd operation I haven't got the leverage to drive these groups to rebuild things from the bottom up.

I, Personally am responsible for fronting up to our CEO, CIO, General Managers and VPs with a report demonstrating what our SLMs are doing to drive Service Improvement,

An ITIL Manual would extoll the virtues of integrated processes that work in harmony with metrics that demonstrate that all your processes are working as they should.

An ISO20000 Auditor would come in and ask the first question "Do you have management commitment", any answer other than "no" would be a lie - our executives are not going to have a root and branch review of what they do, partly because it would be expensive to do that and partly because it would probably reveal that in exercising their desire to save a few dollars/rupees they have sacrificed quality/service performance.

So its broken and as things stand I will be rocking up each month to meet with our senior execs to report something wholly inadequate.

So in the interests of parochiality I am looking to be more personally effective and take a top down approach that helps me to deliver.

I'm looking to have OLAs between myself and the SLMs, to make sure they give me their reports in a timely fashion and they're fit for purpose (e.g. no program where Service Availability is below target without and accompanying 'get well' plan). However, I'm very much in the same boat as the SLMs and I recognise that the biggest issues lie with the lack of quality/performance in the incident management process and the lack of follow up from Problem Management.

I want to put SLAs in place between Service Desk/Sit.Management and SLMs and also Problem Management and SLMs, with associated metrics and reporting. Does anyone have any experience in doing this and if so what kinds of things did you measure? I'm thinking of correct prioritisation of incidents, incidents referred to correct group, incidents resolved with SLA timescales, problem ticket coverage (not sure on this one - maybe an automatic problem record for every Sev1 with an impact greater than 'X', timely/accurate production of post incident reports) there are no doubt more.

If I can get these OLAs in place my hope is that the lack of performance in certain groups will be highlighted and addressed, driving more quality, performance and robust processes - thats not mine to drive I can only influence.

As I said earlier, I know the true answer is management commitment, robust processes, etc, etc but thats not something I can control.....I can only look to be personally effective and drive out the things I need in a more subtle way.

I have some thoughts on how I might do this (as you can see) but I would really value the thoughts and ideas of others especially if you have experience of doing this.

I'm looking forward to a lively debate :lol:

Post Reply