What gets measured gets improved – HelpDesk
In today’s episode of the IT Provider Network we are going to discuss ways you can improve your service team and specifically your helpdesk by holding your team accountable to some metrics.
- Volume – How many tickets are you generating on a daily, weekly, monthly basis
- Response – How fast are you responding to and solving customer problems
- Noise – What are your ticket types and who are your noisy customers
- Production – Who on your team is getting it done and who is playing on their phone
How many customers, end points, reactive and proactive tickets are you servicing and how is this going to effect your helpdesk.
Number of customers
This is important because smaller customers are much harder to service that fewer larger customers. You will find as you grow your business you might need to weed out some of the smaller older customers that you have because they will abuse you!
In general, we strive for a customer with 20 users or more and we like them to have 30 users or more.
Number of Locations
Same rules apply – more locations more places to roll trucks too, more internet connections to go down and more network devices to support. If you are supporting many multi-location customers make sure you include a built in adder in your pricing so that you can compensate for these extra costs.
Number of End Points
This might be your #1 determination in your monthly billing. Many MSP’s bill their clients by having their RMM tools report to the PSA the number of agents you have installed in the field and then this drives quantity counts to your agreements which in turn you use to bill your clients. We use ConnectWise Automate (formally Labtech) and with Automate we have the ability to count and report through the Plugin, server, workstation, printer, router, other device counts to Connectwise Manage for billing. We have a mix of clients. Older Ones that we bill by endpoint but our newer clients are all user based. This alleviates any fear a client has over extra laptops, tablets or mobile phones.
We’ll take about pricing in an upcoming episode of the IT provider network but we have moved to user based pricing because it seems to align better with what our clients want to measure…headcount.
Reactive / Proactive Ticket Counts
These counts are very important to build and scale your service delivery team and running an effective helpdesk. Knowing and monitoring your ticket counts is key to correct staffing.
At our company, we track everything in real-time with dashboards and our dispatch portal but we report via our daily huddles on yesterday’s metrics. We use a combination of Connectwise’s Report Writer along with a great third party tool called Brightgauge.
Brightguage integrates with many tools including Connectwise, Quickbooks, IT Glue, BMS by Kayesa, Datto, Continuum, Autotask and many others. Brightgauge allows you to securely import your data, organize and analyze it then create outstanding dashboards scorecards and reports.
Pricing starts at $229 a month and goes up to over a thousand a month, so it’s not inexpensive but it is well worth the price. You can find more information as well as the Brightgauge Podcast at www.brightgauge.com
Getting back to helpdesk ticket counts. We report # of tickets opened, # of tickets closed, #total open tickets for ticket volume. This gives us two data points to make sure we are keeping up with the volume. Are we closing more that we opened and is the total # of tickets moving down?
How fast are you able to respond and resolve your customer’s issues.
At the end of the day, this is one of only two things your clients really care about. How fast can you solve my problem and how much does your offering cost? If you do a good job at the first one they will care much less about the second one. But know this, if you don’t service them well then they will definitely bitch about your bill.
If you want to see or listen to Episode 7 of the IT Provider Network we talked about 5 Things to Keep your Customers Happy and Reduce Churn
Average Time to Respond
Average Time to Respond to service issues is the first response time metric that we track. Your PSA will often give you this metric IF you have your system set up correctly.
This is the time that a support person actually started working on the ticket, not the time when your system sends back an automated response that the ticket was received.
You need to balance the response time with the cost of labor but generally, you want this to be a low number, less than 10 minutes is acceptable for most clients.
Average Time to Resolve
This one is a little harder to manage and greatly depends on how you setup your helpdesk and service boards. If you have all your proactive and reactive tickets living on one service board then it is really hard to get any kind of meaningful number out of this metric. Take for example if you have an outlook problem on the same board as a server project. One should be resolved in 20 minutes and the other may take 20 days.
When we first implemented our Connectwise PSA, we had 3 service boards and we worked this way for along time until we started to try to measure and improve our service delivery.
- Our Helpdesk Board – this is where any customer generated service issue is logged. It doesn’t matter whether the ticket is originated from an email, phone call or our taskbar icon – if the customer originated it, it starts here.
- NOC board – This is where all of the “software generated” tickets land. This includes all the RMM tickets as well as any other 3rd party integration that is not related to data backups. We get a lot of noise on this board and if I am being honest sometimes I just go in there and wipe out the noise. We should do a better job of fixing the problem at the source by tidying up our monitors and alerts. Sometimes it’s hard to fix things like a disc full alert, you may have an older customer that has a system volume that is 85% full – that’s never going to change until they upgrade or replace the hardware so you either have to have a bunch of one off monitors or you just end up ignoring the alerts. I don’t recommend you do the later so… do as I say not as I do.
- Project Board – Our project board was the third original board. This is where we have any time and material projects that have been quoted outside of our managed service agreements. We determine a project to be any job that is over 8 hours and/or involves more than one person.
We have since added three additional boards based on experiences we have had over the years.
- Backup issue board – we wanted to break these from our NOC issues so we could use our overnight team could work the issues BUT we also wanted full visibility into backup problems. We have been in business for over 21 years and we had two situations where we ended up in real trouble with clients – both were 100% our fault and both were caused by us not fixing a backup problem in a timely fashion. If you ignore a backup problem in our company now we call this a CLM – Career Limiting Move.
- VCIO board is the second new board we added – VCIO issues sometimes don’t have a quick resolution or they can stay open for a long time. For example, the customer might want you to look into something based on a quarterly business report but it never ends up moving forward, or you may show the customer that they need a new server but that spawns a long conversation about virtualization. By adding a VCIO board we separate these types of consultative issues from true task based service tickets.
- Sales Board is the last new board we added – we have a couple special statuses to move a service ticket back and forth from service to sales and possibly back again. This often happens when a tech uncovers a need for a hardware upgrade. He assigns it to sales, sales works with the client to discover the true need, and this is often different than the tech uncovered. The tech is thinking tactical and hopefully, your VCIO or salesperson is being more strategic. Sales might need to send it back to service for additional analysis. By sending the ticket between the boards you can do a better job of managing the clients SLA and your metrics won’t be skewed.
Noise is a number of tickets from a client, user, device or application that is screwing with your profit! You need to be on the lookout for trends that are above the norm and try to silence them.
We track the top 5 noisiest clients for the week. We do this by tracking total tickets per customer. Sometimes a customer will get noisy for a specific reason. Some that come to mind are:
- LOB upgrades – while this shouldn’t create noise if and often do if you usually it happens because you aren’t doing your job. If you screwed the pooch on the upgrade, or if the client wasn’t trained on the new version and you allowed that to be OK, or if permissions we such that a user did the upgrade themselves SHRIEK
- Acts of God – A good storm, lightning, flood or other act of God can really cause a user to become noisy. Not much you can do about that, other than attack the problem quickly and get them whole again as soon as possible.
- New Management -sometimes a new manager comes on board and usually, you see what I call pigeon management. They swoop in, shit all over everything, and fly away. Sometimes this includes you too.
Many other reasons too but the point is you want to make sure you are monitoring all sources of noise and controlling the ones you can. If after doing that, you find that a customer is still too noisy then perhaps you should suggest they find another MSP where they can be more successful!
Production is the final group of metrics that we monitor in the service desk. This one, like all of them, is a little tricky. Each tech is not going to produce the same amount of work but there are reasonable expectations that you should expect from them. For our team members, it is 7-10 tickets closed per day per man plus the following:
- Working Hours – On-time and ready for work
- Submitted Time Sheets A completed time sheet for the day before by 9:10 am the next morning – this is the time of our daily huddle. This includes truing up the time sheet to account for 8 hours.
- Fully documented tickets Complete with problem and resolution documentation and communications with the customer.
- Contributions to the Knowledgebase – We measure the updates in IT Glue as well as the new documents created.
- Work completed on your Employee Strategic Plan – this usually includes one internal lunch and learn and one process improvement suggestion per quarter. We talked about ESP’s in Episode 9 of the IT Provider Network on Hiring – Process and Tools to Find Great Talent
- Individual Customer Satisfaction Scores – we don’t require a certain number of ticket surveys but the ones you have better be over 95% satisfied.
Ask your questions or share your feedback
•Comment on the shownotes below
•Email firstname.lastname@example.org (audio files welcome)
Low Volume by it is THE PLACE TO BE for all of my insider information. If you're on the list, you'll be the first to know. Signup Here
Tell 'Em I Sent You
If you investigate or purchase any product or service that I have talked about, please remember to mention that I sent you. I may or may not get a small commission but more importantly I hope to spread the word about the IT Provider Network.Please Subscribe
Please consider Subscribing to the Podcast it is the only measure of listenership I have.
Give me a Review
Your written iTunes reviews encourage me and they help other people find the podcast. If you appreciate the information you are learning from the IT Provider Network then please do me a HUGE FAVOR and write a review on iTunes or on Stitcher!
Connect with Me
On the socials as @TerryRossi or @ITProviderNet
Help a Brother Out