Do IT Helpdesks Evolve? How?

egrizzly

One of my newer experiences in the past few years was to work in an IT helpdesk as it morphed from one stage to another.

I got hired into a company where the entire IT team was a bunch of different functional groups mixed up together. It included Infrastructure Support (e.g routers, servers, switches, e.t.c), Identity Management (e.g. creating and managing accounts in Active Directory), Help Desk, and Deskside/Onsite support. Members of that IT team participated in all of the above functions and no role was separate.

6 months after I joined the company a veteran IT manager was hired. First thing he did was to separate all those functions and tap a first-line manager to manage each functional group. There were now four groups that reported to him, the helpdesk being just one of them. My role at the helpdesk now merely involved taking inbound calls for general issues covering all the listed functional roles. Whatever we didn't resolve at the helpdesk we then passed over to either of any of those groups. Still, within the helpdesk, there was no formal organization. Everybody resolved issues from their own opinions and past experience.

A year later a small group was spun-off again from the helpdesk and was called the QA Team. Their role was simply to listen to the past helpdesk calls and grade that call using a documented process judging if the callers where greeted, assisted, or supported properly.

Next most noticeable change is that now, the helpdesk is having to use documented knowledge articles to resolve issues.

...ok, now, with all that said it certainly looks like our helpdesk changed from a particular configuration, to another, then another.
For the managers out there, is this a known fact in IT best-practice, that helpdesks do in fact evolve. If that is the case, for a company of 8000 employees what is the most ideal state of evolution a helpdesk should be in to be the most effective? Is there any documented model in IT best-practice that a helpdesk can follow to achieve this state?

I wanted to do some objective leg work and actually get an unbiased sense of whether this change in IT helpdesks is known to be repeated across companies all over the world and how to achieve it.

EANx

Organizations will have different approaches to IT services depending on where they are as an organization, both size and maturity as well as the mission of the organization. It's far easier to use an ad hoc approach to IT when the organization is small and immature; get the job done is the call and whoever happens to be nearby does the job. These are organizations where the CFO will answer the main number when the receptionist had a problem with her car or a someone from sales will step into the role of customer support when that group is down a person or two. These organization can't grow past a certain size without discipline.

Discipline in IT takes several forms, structured teams that specialize, a formal ticketing system, knowledge base articles and service-level agreements are all common ways of creating structure and discipline from those ad hoc teams.

The most common set of standards around this is called ITIL (there are others as well). None of the standards tell you how you should do something, only what you should be doing. So you should have a ticketing process, including defined ways of communicating with the customer and how often but ITIL won't tell you which tool to use or if emaill is better than a phone call, those are all highly organization dependent.

All of this is designed to ensure the services IT provides have a "quality" component. It's hard to measure quality, even harder when you're not thinking about adding it in from the start.

DatabaseHead

I've seen it happen sort of like that. Usually happens every time after leadership changes.

We had a guy come in, he split the Help Desk level 1 and 2 to 1 and 2, but one was enterprise application support and infrastructure. Guys like you would of been on the infrastructure side, more "techs" the other side was application support, how to use a system why it's behaving etc.

Then another manager came in and knocked that divide out and went back to 1 and 2.

QA was done by the managers, usually 3 calls a month to keep in your file. QA teams seems a little over the top but if the desk is that large maybe they need something like that....

Welly_59

Tbh the new structure that you have mentioned is the structure that I am used to seeing in IT

TheFORCE

Welly_59 wrote: »

Tbh the new structure that you have mentioned is the structure that I am used to seeing in IT

In mature organizations you would find a mature Helpdesk. He happened to join an organization where there was no structure and he thinks that the Helpdesk evolved.

OP you need to open an ITIL book and read and try to implement some of the ITIL best practices.

egrizzly

I believe it will change again. As it stands now the solution to only half the total incident types we get are documented in knowledge base articles while solutions to just a quarter of the high-level incidents are documented. So, as it stands now if a high-level incident is called in, only 25% of the specialists there can resolve it off the top of their heads, the other 75% will escalate it.

However, these knowledge base solutions are increasing and I think when they get to 100% it will change the helpdesk into another state. It was a hunch however I figured to post it here to see if these changes belong to an already documented metamorphosis.

UncleB

It sounds like the different team approach is the most effective for your organisation - 8k users is a lot to deal with and the chances of any one of you knowing how to do everything as vanishingly small so you need to have an approach that lets you become masters of your own arts then link with the others who are also good at what they do and develop the most effective way to route incoming calls accurately to the people who can resolve them.

This is the best approach to large-ish companies in my experience as it only takes a situation when you lose a few IT staff with wide reaching skills and you suddenly have a near-impossible skill set to recruit for since these guys spanned multiple disciplines and had a bucketload of indocumented knowledge that kept them secure in their jobs.

I find the most effective approach for a manager to take in your sort of environment is to get a consultant in to review older tickets, look for trends (ie problems in the ITIL definition or areas you are weak in), look for areas of poor documentation and then create a requirement for the staff with the skills or knowledge to address these. This takes a big chunk out of your daily overhead of calls with time.

It also works very effectively to get a consultant to look at the documentation matrix that covers all aspects of all the systems you cover and generate a skills match for this matrix for the staff. Get the matches of staff with absent documentation and make it a formal part of their objectives to produce the documentation then have it all peer group reviewed so everyone knows about it and where to find it.

Keep this cloud based (passwords kept separately of course) and you have well on your way to providing a big part of the Business Continuity component for support too.

It is always disruptive for staff to have to surrender their knowledge this way and to become a small cog in a bigger machine, but from a management perspective it is essential to remove reliance on individuals and make IT part of a service. I don't like to diminish individuals but cannot allow them to become a risk of the success of the whole business function of IT.

That is probably the key difference on how a team leader or staff member would see the situation to how a manager would - the manager has to look after how IT performs for the business while a team leader has to look after their team and do what the manager tells them without worrying about the bigger picture.

Welly_59

egrizzly wrote: »

I believe it will change again. As it stands now the solution to only half the total incident types we get are documented in knowledge base articles while solutions to just a quarter of the high-level incidents are documented. So, as it stands now if a high-level incident is called in, only 25% of the specialists there can resolve it off the top of their heads, the other 75% will escalate it.

However, these knowledge base solutions are increasing and I think when they get to 100% it will change the helpdesk into another state. It was a hunch however I figured to post it here to see if these changes belong to an already documented metamorphosis.

High priority incidents will nearly always be escalated. For it to be a high priority it needs to have a high impact and severity. Usually this is a breakdown of an application, hosted service, network infrastructure etc. None of which a 1st line team will usually have the access required to fix

volfkhat

Wait a minute!!
I thought you got a new job offer?

Why aren't you outta there?
:]

EANx

Welly_59 wrote: »

High priority incidents will nearly always be escalated. For it to be a high priority it needs to have a high impact and severity. Usually this is a breakdown of an application, hosted service, network infrastructure etc. None of which a 1st line team will usually have the access required to fix

High impact/severity can also mean the CFO is having problems with his laptop 30 minutes before a presentation to a bank or hedge fund (or that the VP of Marketing can't find his crayons). What is "high" for impact or severity depends on the organization and industry.

Welly_59

To a degree. If your following ITIL best practice then the person affected by the incident is irelevant. It's purely on the effect and impact on the business/client.

I know in real life it can be different. I deal with one client, the service desk will always email direct if it's effecting a "vip" . Doesn't make a blind bit of difference to how I work on the incident though, as I'm often already juggling issues affecting multiple sites, multiple systems etc to a degree that one person with a title isn't going to be getting any extra attention because of it

TechGromit

Different managers have different idea, Help desk could be running at 99% efficiently, and some manager will come in and think they can make things better by changing everything. I remember training on a program called "McAfee Help-desk" to do tickets, it was suppose to allow us to build a knowledge base to help us resolve help desk tickets that someone encountered before. Unfortunately once it was deployed, the project lead got pull onto other projects and no one was assigned to build that knowledge database. Two year later we were training on a help-desk program called "Heat" again with big promises. Far as i was concerned, there was nothing wrong with the last help desk program, but some manager was talked into buying it by a quick talking salesperson. I recall a time, there were so many changes in upper management IT, that we couldn't even finish changes the last management wanted before we were redoing our process for the next manager of the day.

EANx wrote: »

High impact/severity can also mean the CFO is having problems with his laptop 30 minutes before a presentation to a bank or hedge fund ...

Reminds me, everyone has to put in help desk tickets for IT service but Director level and above management is always except, drop everything and help them ticket or not. I recall seeing a job posting 6 months ago in my company, Senior Level IT position solely responsible for troubleshooting VP computer issues. Thanks but no thanks.

egrizzly

I am...for the most part. I gave 2 weeks notice haha.

volfkhat wrote: »

Wait a minute!!
I thought you got a new job offer?

Why aren't you outta there?
:]

egrizzly

I would like to put together a 6 month plan for our VP to help transition the department from its current stage to another state where the helpdesk is 90% efficient on all inbound calls. That's 90% overall FCR as they say.

UncleB

egrizzly wrote: »

I would like to put together a 6 month plan for our VP to help transition the department from its current stage to another state where the helpdesk is 90% efficient on all inbound calls. That's 90% overall FCR as they say.

...I gave 2 weeks notice

Do you think they will take any notice of someone about to leave the company? Save your breath and energy for your next employer.

Welly_59

biggest win for helpdesk is improving first time fix rate. Saves incidents being passed to 2nd/3rd line teams, and also ensures a quicker fix for the user as it doesnt have to get bounced around multiple teams to get resolved.

But then you have the issue's of SLA's. If the SLA's are all focused on initial contact time, time to answer etc then you find that first time fix goes out the window as everyone worries about their call stats too much