Our operating numbers, methodology included.
Updated monthly. Every figure states how it was calculated, because a number without a method is a marketing claim rather than a metric. Where we missed, the miss is on the page.
11minutes
Target for P1 is 15 minutes
Mean elapsed time from ticket creation to first human response, across every ticket of every priority. Automated acknowledgements are excluded — a robot replying is not a response.
99.2%
2 misses in 249 P1 tickets
P1 tickets answered within 15 minutes, divided by all P1 tickets. Two misses in the window. Both were credited automatically.
74%
Category benchmark is roughly 60%
Tickets closed by the first engineer who touched them, without escalation or a second session. Reopened tickets count against the original.
97%
31% survey response rate
Every closed ticket triggers a two-question survey. Response rate over the window was 31%. All responses count, including the ones we would rather not publish.
0.81per month
Maturity threshold is 1.0
Total tickets divided by managed endpoints. Lower is better: it means the environment is stable, not that the help desk is hard to reach. Above 1.0 is generally read as an unstable estate.
$1,840last quarter
Against roughly $687,000 invoiced in the quarter
Dollar value of service credits applied to client invoices for missed P1 and P2 first-response targets. We publish this because a remedy nobody ever collects is not a remedy.
The numbers that take longer to move.
Operating metrics tell you what this month felt like. These tell you what five years produced.
How these are produced, and what we do not do.
Every figure on this page comes from HaloPSA, our ticketing system, over a rolling 90-day window. It is pulled on the first business day of each month by our operations manager and reviewed by the CTO before it goes up. The window moves; the numbers are not cherry-picked from a good quarter.
What is excluded, and why
- Automated acknowledgements do not count as a first response. A robot replying to your ticket is not a response, and counting it would roughly halve our average.
- Internal and project tickets are excluded from ticket counts and CSAT. Including scheduled project work would inflate volume and flatter first-contact resolution.
- Reopened tickets count against the original ticket for first-contact resolution. Closing something optimistically and reopening it the next day is not a resolution.
- All CSAT responses are counted, including the bad ones. We publish the response rate alongside the score so you can judge how much the score is worth. A 97% CSAT on a 4% response rate would not be worth much.
What we would publish if it went wrong
This one is worth saying in advance, because it is easy to promise transparency in a quarter that went well. If P1 attainment drops below 95%, if CSAT drops below 90%, or if we have a security incident affecting a client environment, it goes on this page in the month it happens, with what we did about it.
The two P1 misses in the current window were a 19-minute response during a regional Comcast outage that also took down our own primary line, and a 22-minute response caused by an on-call handoff error on a Sunday. Both were credited automatically. The handoff process changed the following week.
Why the SLA credit figure is on here
Because a remedy nobody ever collects is not a remedy. A provider who publishes a service credit clause but has never issued one is telling you either that they are perfect or that the clause is decorative, and only one of those is likely. Last quarter we issued $1,840.
The commitments these are measured against