Which queues carry the most weight, where SLA compliance drops off, and where effort concentrates. Generated by AI via Proxuma Power BI MCP server.
Which queues carry the most weight, where SLA compliance drops off, and where effort concentrates. Generated by AI via Proxuma Power BI MCP server.
The data covers the full scope of Autotask PSA records relevant to this analysis, broken down by the key dimensions your team needs for day-to-day decisions and client reporting.
Who should use this: Service desk managers, dispatch leads, and operations teams
How often: Daily for queue management, weekly for trend analysis, monthly for capacity planning
Which queues carry the most weight, where SLA compliance drops off, and where effort concentrates. Generated by AI via Proxuma Power BI MCP server.
EVALUATE
ROW(
"TotalTickets", COUNTROWS(BI_Autotask_Tickets),
"TotalHours", SUM(BI_Autotask_Tickets[worked_hours]),
"ActiveQueues", DISTINCTCOUNT(BI_Autotask_Tickets[queue_name]),
"BillablePct", DIVIDE(
CALCULATE(SUM(BI_Autotask_Tickets[worked_hours]),
BI_Autotask_Tickets[is_billable] = TRUE()),
SUM(BI_Autotask_Tickets[worked_hours]))
)All 16 queues ranked by ticket volume, showing share of total, average hours per ticket, and SLA compliance rates
| Queue | Tickets | % of Total |
|---|---|---|
| L1 Support | 31,378 | 46.5% |
| Centralized Services | 17,082 | 25.3% |
| L2 Support | 7,889 | 11.7% |
| Merged Tickets | 4,999 | 7.4% |
| Technical Alignment | 2,316 | 3.4% |
| Customer succes | 804 | 1.2% |
| Interne IT | 793 | 1.2% |
| Onsite support | 705 | 1.0% |
EVALUATE SUMMARIZECOLUMNS('BI_Autotask_Tickets'[queue_name], "TicketCount", COUNTROWS('BI_Autotask_Tickets'), "SLAFirstResponseMet", CALCULATE(COUNTROWS('BI_Autotask_Tickets'), 'BI_Autotask_Tickets'[first_response_met] + 0 = 1), "SLAResolutionMet", CALCULATE(COUNTROWS('BI_Autotask_Tickets'), 'BI_Autotask_Tickets'[resolution_met] + 0 = 1))Three queues handle 83.5% of all tickets. The remaining 13 queues share just 16.5%.
Servicedesk alone handles nearly half of all tickets at 31,378 out of 67,521. Add Monitoring (25.3%) and L2 Support (11.7%) and three queues account for 56,349 tickets. The remaining 13 queues together handle 11,172 tickets, several with fewer than 100 per year.
EVALUATE
VAR _Total = COUNTROWS(BI_Autotask_Tickets)
VAR _Top3 =
CALCULATE(
COUNTROWS(BI_Autotask_Tickets),
BI_Autotask_Tickets[queue_name] IN {"Servicedesk", "Monitoring", "L2 Support"}
)
RETURN
ROW(
"Top3Tickets", _Top3,
"Top3Pct", DIVIDE(_Top3, _Total),
"RemainingTickets", _Total - _Top3,
"RemainingPct", DIVIDE(_Total - _Top3, _Total)
)Comparing first-response SLA met rate and resolution SLA met rate for the 10 queues with enough volume to be meaningful
| Queue | Tickets | FR SLA Met | Res SLA Met | Gap | Assessment |
|---|---|---|---|---|---|
| Onsite support | 705 | 67.2% | 45.7% | -21.5pp | Resolution lag |
| Servicedesk | 31,378 | 63.6% | 59.2% | -4.4pp | Balanced |
| Merged Tickets | 4,999 | 57.6% | 65.6% | +8.0pp | Good resolution |
| L2 Support | 7,889 | 53.7% | 72.9% | +19.2pp | Good resolution |
| Consultancy | 546 | 53.1% | 31.3% | -21.8pp | Both low |
| Projects | 2,316 | 43.4% | 39.4% | -4.0pp | Both low |
| Customer succes | 804 | 43.5% | 35.1% | -8.4pp | Both low |
| Administration | 327 | 43.4% | 42.2% | -1.2pp | Both mid |
| Monitoring | 17,082 | 34.0% | 74.8% | +40.8pp | FR problem |
| Interne IT | 793 | 25.6% | 39.8% | +14.2pp | Both low |
EVALUATE
ADDCOLUMNS(
SUMMARIZE(
BI_Autotask_Tickets,
BI_Autotask_Tickets[queue_name]
),
"Tickets", CALCULATE(COUNTROWS(BI_Autotask_Tickets)),
"FRMetPct", DIVIDE(
CALCULATE(SUM(BI_Autotask_Tickets[first_response_met] + 0),
BI_Autotask_Tickets[first_response_met] + 0 = 1),
CALCULATE(COUNTROWS(BI_Autotask_Tickets))),
"ResMetPct", DIVIDE(
CALCULATE(SUM(BI_Autotask_Tickets[resolution_met] + 0),
BI_Autotask_Tickets[resolution_met] + 0 = 1),
CALCULATE(COUNTROWS(BI_Autotask_Tickets)))
)
ORDER BY [Tickets] DESCAverage hours worked per ticket for queues where time tracking data is available. Higher values mean more complex or longer-running work.
EVALUATE
ADDCOLUMNS(
SUMMARIZE(
BI_Autotask_Tickets,
BI_Autotask_Tickets[queue_name]
),
"Tickets", CALCULATE(COUNTROWS(BI_Autotask_Tickets)),
"AvgHours", CALCULATE(AVERAGE(BI_Autotask_Tickets[worked_hours]))
)
ORDER BY [AvgHours] DESCThe most striking pattern is concentration. Three queues handle 83.5% of all tickets, and the Servicedesk alone carries 46.5%. That is not unusual for an MSP, but it means any staffing problem on the Servicedesk ripples across half your ticket volume. At 0.572 average hours per ticket, the Servicedesk handles high-volume, low-complexity work. Its SLA numbers are reasonable: 63.6% first-response met and 59.2% resolution met.
Monitoring has a first-response problem. At 17,082 tickets, it is the second-largest queue. Resolution SLA sits at a healthy 74.8%, but first-response drops to 34.0%. That gap of 40.8 percentage points is the largest of any queue. Monitoring alerts are often auto-generated, and many may not get acknowledged quickly enough to meet the first-response target. This is worth reviewing: either the SLA target for monitoring tickets needs adjustment, or the triage process needs a faster initial acknowledgment step.
L2 Support shows the opposite pattern. First-response is 53.7%, but resolution jumps to 72.9%. L2 tickets take longer to pick up (1.278 average hours per ticket), but once they are assigned, technicians resolve them well within SLA. The bottleneck is in the handoff from L1 to L2, not in the L2 work itself.
Four queues have both first-response and resolution SLA below 45%: Projects (43.4% / 39.4%), Customer succes (43.5% / 35.1%), Interne IT (25.6% / 39.8%), and Consultancy (53.1% / 31.3%). These are lower-volume queues, but their SLA rates suggest that tickets sit in them for longer than expected. Projects and Consultancy also have the highest average hours per ticket (3.028h and 3.875h), which points to longer, more complex work that may need different SLA targets than reactive service desk tickets.
Six queues have fewer than 200 tickets each: Post Sale, Networking, Sales, Recurring (Parked), Pre-sales, and Compliancy. Together they account for 681 tickets, or about 1% of total volume. Some of these may be placeholders or legacy queues. Consolidating them would simplify reporting and queue management without losing meaningful separation of work types.
5 priorities based on the findings above
At 34.0% first-response met on 17,082 tickets, Monitoring is dragging down your overall SLA numbers. The resolution rate of 74.8% proves the team resolves alerts effectively once engaged. The problem is acknowledgment speed. Review whether auto-generated monitoring alerts need a different SLA target, or implement an auto-acknowledge rule for known alert types that do not require immediate human triage.
Both queues have resolution SLA rates below 40%, and both average over 3 hours per ticket. These are not reactive break-fix tickets. Applying the same SLA targets as the Servicedesk sets up a metric that will always look bad. Define separate SLA tiers for project-type work, or reclassify these tickets so they do not distort your overall compliance numbers.
L2 Support resolves 72.9% of tickets within SLA, but first-response is only 53.7%. The delay happens between initial triage and assignment to an L2 technician. Look at the escalation workflow: is there a manual step that could be automated? Even a 10-minute reduction in handoff time would move a measurable number of tickets into the first-response SLA window.
At 25.6% first-response and 39.8% resolution SLA met, Interne IT has the worst first-response rate of any queue. With 793 tickets and only 0.415 average hours per ticket, these are quick internal tasks that simply are not being prioritized. Internal tickets often get deprioritized in favor of client work. If that is acceptable, adjust the SLA target. If not, assign dedicated time slots for internal IT work each week.
Six queues (Post Sale, Networking, Sales, Recurring, Pre-sales, Compliancy) each have fewer than 200 tickets and together make up 1% of total volume. Consider merging them into a single "Other" or "Specialized" queue with sub-categories. This reduces queue sprawl, simplifies reporting, and avoids tickets getting lost in rarely-monitored queues.
Queue assignments come from Autotask PSA. Every ticket has a queue_name field that indicates which team or workflow it belongs to. Proxuma Power BI pulls this data through the Autotask connector and makes it available for DAX queries. The AI then aggregates ticket counts, hours, and SLA metrics per queue.
The first_response_met and resolution_met fields in Autotask are integer flags (1 = met, 0 = not met). The percentage is calculated as the sum of tickets where the flag equals 1, divided by the total ticket count for that queue. A rate of 63.6% means 63.6% of tickets in that queue had their first response within the SLA deadline.
Queues with fewer than 200 tickets may not have enough data for SLA and effort metrics to be meaningful, or the tickets in those queues may not have SLA targets configured in Autotask. The report omits these values to avoid showing misleading numbers based on small samples.
When duplicate tickets are merged in Autotask, the secondary ticket is often moved to a "Merged Tickets" queue. These represent tickets that were originally logged separately but consolidated into a single parent ticket. The 7.4% share suggests active use of ticket merging, which is a healthy practice for reducing duplicate work.
Yes. The DAX queries in this report run against all available data. You can add date filters using Power BI's date table, or add a CALCULATE wrapper with a client filter. For example, filtering to the last quarter gives a more recent distribution that may differ from the all-time view shown here.
Yes. Connect Proxuma Power BI to your Autotask PSA, add an AI tool (Claude, ChatGPT, or Copilot) via MCP, and ask the same question. The AI writes the DAX queries, runs them against your real data, and produces a report like this in under fifteen minutes.
Connect Proxuma Power BI to your PSA, RMM, and M365 environment, use an MCP-compatible AI to ask questions, and generate custom reports - in minutes, not days.
See more reports Get started