SLA attainment
+0.5 pp96.7%
% of tickets within target · target ≥ 95.0%
Above target for six straight months, tracking the fall in MTTR.
Example dashboard — demo data.Synthetic figures built to illustrate the instrumentation. No employer data.
These four boards are worked examples by Alvaro Garcia, built to demonstrate business acumen: what to measure, who to measure it for, and what to conclude from the reading. The numbers are invented; the judgment is the point.
ITIL v4
IT Service Operations
- Audience
- COO, CIO, service delivery
- The question it answers
- Are we meeting our service commitments, and where is the operation about to break?
- Measurement period
- Rolling 12 months, Sep 2025 – Aug 2026 · headline values as of 31 Aug 2026 · incident queue as of 09:00
Service health
Are we meeting what we committed to?
Every metric carries its ITIL target. A number without a target is trivia.
MTTR
-0.3h3.9h
hours to restore · target ≤ 4.0h
Down 43% across the year. Runbook automation did most of this, not headcount.
MTTA
-0.1h0.8h
hours to acknowledge · target ≤ 1.0h
Acknowledgement is now under an hour — the on-call rotation change landed.
Change success rate
+0.4 pp98.1%
% of changes without rollback · target ≥ 95.0%
December's dip was the freeze-period backlog released in one window.
First-contact resolution
+1.1 pp71.4%
% resolved on first touch · target ≥ 75.0%
Improving steadily but still short of target — the knowledge base lags the product.
Critical service availability
99.93%
% uptime, 5 services · target ≥ 99.90%
99.93% is roughly 29 minutes of downtime in a month across all five services.
Ticket backlog
-11171
open tickets · target ≤ 200
Down 40% from the December peak, with only 5 tickets older than 30 days.
Cost per ticket
-$0.80$24.30
USD, fully loaded · target ≤ $26.00
Falling because first-contact resolution rose, not because service was cut.
SLA attainment
Unit: % of tickets resolved within target
Above the 95% target for six consecutive months. Read this against the response times beside it — when attainment rises while MTTR stays flat, someone is re-classifying tickets rather than resolving them faster.
Show data table
| Month | SLA attainment | MTTR | MTTA |
|---|---|---|---|
| Sep | 91.2% | 6.8h | 1.9h |
| Oct | 91.8% | 6.5h | 1.8h |
| Nov | 92.4% | 6.2h | 1.7h |
| Dec | 92.1% | 6.4h | 1.8h |
| Jan | 93.3% | 5.8h | 1.5h |
| Feb | 93.9% | 5.5h | 1.4h |
| Mar | 94.4% | 5.2h | 1.3h |
| Apr | 94.9% | 4.9h | 1.2h |
| May | 95.2% | 4.7h | 1.1h |
| Jun | 95.8% | 4.4h | 1.0h |
| Jul | 96.2% | 4.2h | 0.9h |
| Aug | 96.7% | 3.9h | 0.8h |
Time to acknowledge and time to restore
Unit: hours
MTTA fell 58% while MTTR fell 43%. Acknowledgement improving faster is the right order — it means triage got better, not that people worked harder.
Show data table
| Month | MTTA | MTTR |
|---|---|---|
| Sep | 1.9h | 6.8h |
| Oct | 1.8h | 6.5h |
| Nov | 1.7h | 6.2h |
| Dec | 1.8h | 6.4h |
| Jan | 1.5h | 5.8h |
| Feb | 1.4h | 5.5h |
| Mar | 1.3h | 5.2h |
| Apr | 1.2h | 4.9h |
| May | 1.1h | 4.7h |
| Jun | 1.0h | 4.4h |
| Jul | 0.9h | 4.2h |
| Aug | 0.8h | 3.9h |
- MTTR
- MTTA
Incidents by severity
Unit: incident count per month
The December spike is a release-freeze artefact, left in place deliberately. A dashboard where the holiday freeze is invisible is a dashboard that is smoothing its data.
Show data table
| Month | P1 | P2 | P3 | P4 | Total |
|---|---|---|---|---|---|
| Sep | 4 | 14 | 58 | 96 | 172 |
| Oct | 3 | 13 | 55 | 94 | 165 |
| Nov | 3 | 12 | 54 | 91 | 160 |
| Dec | 5 | 16 | 61 | 103 | 185 |
| Jan | 2 | 11 | 52 | 89 | 154 |
| Feb | 2 | 11 | 50 | 87 | 150 |
| Mar | 2 | 10 | 49 | 85 | 146 |
| Apr | 1 | 9 | 47 | 84 | 141 |
| May | 2 | 9 | 46 | 82 | 139 |
| Jun | 1 | 8 | 44 | 80 | 133 |
| Jul | 1 | 8 | 43 | 79 | 131 |
| Aug | 1 | 7 | 41 | 77 | 126 |
- P1
- P2
- P3
- P4
Open backlog by age
Unit: ticket count
129 of 171 open tickets are under a week old. A backlog that is large but young is a throughput problem; one that is small but old is an ownership problem.
Show data table
| Age band | Tickets | Share of backlog |
|---|---|---|
| 0–2 days | 78 | 45.6% |
| 3–7 days | 51 | 29.8% |
| 8–14 days | 26 | 15.2% |
| 15–30 days | 11 | 6.4% |
| 30+ days | 5 | 2.9% |
Rates against target
Three rates, each with its threshold marked
The tick on each arc is the target. Amber means the gap is real, not that the number fell.
Change success rate
First-contact resolution
Critical service availability
Active queue
Open incidents right now
Severity, owner and time remaining against the SLA target. One ticket is already past its clock — see the reading below.
| Incident | Severity | Service | Owner | Open | SLA |
|---|---|---|---|---|---|
INC-4471 Card authorization latency above 900ms in EU region | P1 | Payment gateway | M. Torres | 1.4h | 2.6h lefttarget 4h |
INC-4468 Batch settlement reconciliation lagging by two cycles | P2 | Settlement engine | K. Raghavan | 5.2h | 2.8h lefttarget 8h |
INC-4462 SSO token refresh failing for a single tenant | P2 | Identity platform | D. Okafor | 7.8h | 0.2h lefttarget 8h |
INC-4455 Document ingestion queue depth above threshold | P3 | Document intelligence | S. Bianchi | 19.5h | 4.5h lefttarget 24h |
INC-4449 Reporting exports timing out over 50k rows | P3 | Analytics | J. Whitfield | 26.0h | Breached by 2.0htarget 24h |
How to read this
What I would say in the room
The numbers above are instrumentation. This is the part that is actually the job — what the pattern means, what it does not mean, and what I would do about it.
- Act now
One incident is already past its SLA clock
INC-4449 has been open 26 hours against a 24-hour P3 target. Reporting exports are not business-critical, but a breached SLA on a low-severity ticket is usually a routing problem rather than a capacity one — it sat unassigned, then got picked up late. Check the queue rules before adding people.
- Watch
First-contact resolution is the constraint on cost per ticket
FCR at 71.4% against a 75% target is the single biggest lever left. Every point of FCR removes roughly 1.6 tickets per hundred from second-line, and second-line costs about four times first-line. Closing that 3.6-point gap is worth more than any further tooling spend.
- Working
The December spike was a change-window artefact, not a regression
P1s tripled and change success fell to 93.8% in December, then both recovered immediately. That is the signature of a release freeze followed by a single large window — the fix is to stagger the post-freeze batch, not to tighten change control, which is already at 98.1%.
- Working
MTTA fell faster than MTTR, which is the right order
Acknowledgement dropped 58% while restoration dropped 43%. Getting the right person looking sooner is what makes every downstream number move; a team that improves MTTR without improving MTTA is usually just working harder on the same broken triage.
- Context
Aging is healthy — 75% of the backlog is under a week old
129 of 171 open tickets are less than seven days old and only 5 are past thirty days. A backlog that is large but young is a throughput question; a backlog that is small but old is a competence or ownership question. This one is the former, and it is shrinking.