A Workflow Distribution SOP for IT MSPs that Works!
“Which of these hundreds of tickets on my Service Manager dashboard merits my attention?”
--John S, 2018
I was speechless! I had no idea how to answer the question, and those words set me on a quest to find the answer….
Today, in general, the answer is “the ones that monitor Service Delivery performance.” Too many times, in the case of John’s question, they’re looking for the tickets needing to be run-to-ground. In my humble opinion, that’s actually the work of the Service Coordinator, and at the time, John did say we had a well-rounded, highly seasoned, highly trained, very experienced team of Service Coordinators. His time would have been better spent monitoring the Intake and Open Ticket performance (aka Service Coordinator responsibilities), the Service Delivery profitability performance (which is what the MSP owner is monitoring already) and the Tech performance.
While dashboards are awesome and provide dynamic, proactive reporting of transactional information in the near-term (day or week), for performance reporting, it takes Autotask Live Reports – but, then again, we’re kinda biased. A pre-requisite for great Autotask live reporting is great, underlying data. Conversely, the old adage “garbage in, garbage out” applies here.
The Road to a Zen Service Delivery Environment
The road to a proactive, data-driven, Zen Service Delivery environment, starts with focusing on dividing the many, 100s, 1000s of client requests into different workflows. And in doing so, the performance of each one can be found, benchmarked against other MSPs, tracked and optimized.
Over the years of working with MSPs from around the world, Advanced Global has identified 11 recommended Workflows that all MSPs should be using. That said, some MSPs may have more than 11, as they use Autotask tickets for Sales, Procurement, vCIO engagements, etc. Even with Autotask continually improving the software to provide two types of Recurring Tickets (Just-in-Time and Scheduled), some MSPs may need two different Workflows. Though each Workflow will need its own SOP, this article/SOP is about how to do the divide-and-conquer SOP.
A simple All Tickets Completed Last Month report will do the trick. The fields we recommend are: Ticket Number, Ticket Title, Priority, Total Hours Worked and Contract. A wealth of information can be gathered from this raw data ticket search if requests have been divided into at least 11 different workflows. Export the list of tickets to Excel (if you don’t see the Export button, contact your Autotask Sys Admin), and in Excel, turn on filters for the 1st row (column headings). Before segmenting client requests into different workflows, filter Total Worked Hours on 0, and copy all rows to another tab include the header tab) – name the tab noise. Go back to the raw data tab and select all Total Hours Worked values except 0.
If you’re already drinking the Advanced Global Kool-Aid, then the next step is easy: Set the Priority filter to the first priority (usually Critical) and copy the list to another tab (again, include the header), rename the tab immediately after the Priority tab, and unselect that priority in the priority filter and select the next one. Continue through the entire list of priorities until all have been copied to their own respective tab.
Segmenting Client Requests
If you haven’t previously segmented client requests, there is a way to get into the ballpark. If, like most MSPs, the priority field is underutilized (which is why we use the priority field to call out what Workflow a client’s request has been Triaged into), the default priority (Medium or Standard [“Low”] will be 80-90% of the tickets, with the rest being mostly Critical. There are two problems with this situation:
1) All tickets are treated with the same SOP
2) The software can’t distinguish the importance of one ticket from another
Without further segmentation, you’ll be dead in the water in your quest to resolving Service Delivery Issues.
Have no fear… Advanced Global is here! Rather than filtering on Priority, filter on Total Worked Hours instead. Please note this process is an approximation and is meant to demonstrate the value of Request Segmentation. You can use the data to make strategic decisions but ignore the actual tickets because some will be miscoded and won’t be in the right Workflow. But for this exercise, that’s OK; the purpose here is to develop a workflow thought process and demonstrate the value of fully triaging client requests. The Workflows we’ll be using are Widget (tickets with less than 4 hours of Total Worked Hours), Calendar (Tickets with 4-16 hours of work), and Projects (Tickets with more than 16 hours of work).
The Thought Behind this Workflow Distribution SOP
Here’s what’s driving our thoughts on this: After working with MSPs from around the world, we see most MSPs dump all tickets into a queue and ask the techs to figure what to work on next. This process is chaotic, inefficient and a barrier to growth. We also see most MSPs wait until the client requests comes in, then try to figure out what to do with it. When it takes too long to figure out, or the next client request takes over their attention, these tickets get dumped into the aforementioned catch-all queue. We’ve come to realize this process will work for tickets needing less than 4 hours to remediate, and it’s the tickets taking over 4 hours causing the chaos. Which means that from an SOP point of view, Incident or Service Request is less important and Total Worked Hours is more important. We’ve also recognized that project work (however it’s defined at your MSP...) needs a completely different process to be finished on-time and on-budget.
Moving From a Reactive to Proactive MSP Operation
The first step in moving from a reactive, break/fix, chaotic Service Delivery operation to a proactive, data-driven, Zen operation is taking the raw list of tickets, sorting the non-zero Total Hours Worked tickets by Total Worked Hours. Copying those greater than 16 hours to a tab renamed Projects, and then 4-16 Total Worked Hours to a tab renamed Calendar and the rest to a tab renamed Widgets.
Now that we have the tickets segmented into different tabs, we can start thinking about them differently. We also know the volume of tickets and how many hours each segment took last month. Dividing both numbers by 12/52 or multiplying by 0.23 will get you to a weekly number, yielding how many hours per week is needed to respond to the Ready to Engage widget on the tech’s combined dashboards. This is an important number to keep from over-scheduling techs. Take care to ensure any work scheduled in the calendar Workflow leaves room for any Widget work. And similarly, of course, all Projects need to be scheduled with the same considerations in-mind.
Speaking of Projects, they’re often unfairly accused of being the culprit of all the chaos. In fact, it’s the excitement of scheduling projects (or even worse, making promises when Projects will start and finish) without considering the number of hours needed for Widget and Calendar remediation.
At this point, we feel we’ve done more damage than good. We’ve opened the Pandora’s box of transitioning from a Reactive, Break/Fix, Chaotic Service Delivery Operation to a Pro-Active, Data-Driven, Zen operation. The truth is, based on those who have gone before you, on your own, the transition will take 7-10 years. Those who have been following Advanced Global for two years or more are in the final stages of our processes. For our clients, we can get them through the process in less than a year. Hang in there! It’s worth the effort and most likely will improve the bottom line by more than $380,000 per year without hiring another tech.
- Steve & Co
