How response is defined and counted
Four priorities, five clocks, the rules for pausing them and the arithmetic behind every KPI in your monthly pack.
- Priorities
- Four, P1 to P4
- Clocks
- Response · attendance · make-safe · resolution · closure
- Windows
- Written per site in the contract schedule
- Achievement figure
- Not published. Yours is in every monthly pack.
- First KPI page
- Day 30 of operation
An SLA sets how fast each kind of fault is answered and fixed; KPIs measure whether the contract is working. ITF publishes the structure: four priorities, five clocks, the rules for pausing a clock and how each KPI is counted. The windows are written per site, because staffing and the drive decide them.
- 01 Enquiry
- 02 Site survey
- 03 Scope and quote
- 04 Mobilisation
- 05 Planned schedule
- 06 Execution
- 07 Monthly report
- 08 Management review
- 09 Next cycle
How response is defined and counted sits at step 6 of the fm operating cycle. You receive closed work orders with photo evidence.
Two hours from what
Most FM contracts promise a response time and never say what it is measured from, what counts as a response, or what stops the clock.
So the number means whatever the provider needs it to mean in the month it is checked. A call logged when the technician arrives rather than when you rang. A P2 quietly downgraded to a P3 to protect a statistic, which is the oldest trick in FM reporting. The ticket history is what stops it: a priority can be changed only with a reason written on the ticket, visible to you.
Tight windows are a staffing model, not a sentence in a contract. A P1 attendance window during staffed hours on a site with our own team is one thing. The same window at 02:00 at a site reached by a mobile technician is another, and the drive decides it. So each site's windows are set against four facts: whether our team is on site, the hours it is there, the drive from the nearest mobile technician, and how critical the assets are.
Buy the tight window where downtime is expensive. Paying for it everywhere is how FM budgets bloat.
The KPI set, and how each one is counted
Every KPI in the monthly pack is counted to a written definition, and the way each one is usually gamed is listed beside it so you know what to look for, in our pack or anyone else's.
| KPI | How we count it | How it usually gets gamed |
|---|---|---|
| PPM completion | Tasks completed inside their window, divided by tasks scheduled for the period. A rescheduled task stays in the denominator of its original period. | Moving late tasks into next month so this month looks clean |
| Response compliance | Tickets meeting the response window, per priority. | Logging the ticket when someone arrives rather than when you called |
| Attendance compliance | Tickets meeting the attendance window, per priority. | Counting a phone call as attendance |
| Resolution compliance | Tickets resolved inside the window, per priority. | Splitting one fault into several tickets |
| First-time fix | Faults resolved on the first attendance, divided by all faults attended. | Closing and reopening a new ticket |
| Repeat failure | The same fault on the same asset inside the period set in your contract. | Recording a repeat as a new fault |
| Open work order age | Open tickets by age band, so a backlog building quietly shows up. | Closing tickets without evidence |
| Evidence completeness | Closures carrying the required photographs and readings. | Not measuring it at all |
| Compliance due dates | Statutory and scheme dates met without lapse. | Tracking only the ones that were met |
What goes into your contract
Annex D, the priority and response schedule
P1 to P4 defined, with windows for your site written separately for staffed hours and out-of-hours cover.
The clock and pause rules
When each of the five clocks starts and stops, and the short list of reasons that may pause one.
The KPI definitions sheet
Every KPI with its numerator, denominator and period, attached to the contract as a schedule.
The escalation matrix
Technician, site supervisor, account manager, contract lead: who fills each role on your site and when each step triggers.
The monthly KPI page
Each KPI counted as defined, with every miss listed by ticket and the cause written against it.
The quarterly trend review
Repeat failures and backlog over three months, the input to resetting windows and intervals at the annual review.
Five clocks, and when each stops
A single response time means nothing until it is defined, so every reactive ticket runs five. A ticket logged at 21:40 on the evening before a working day is logged at 21:40, not first thing tomorrow.
- 01
Response
Starts when the ticket is logged. Stops when it is acknowledged and assigned to a named person.
- 02
Attendance
Starts when the ticket is logged. Stops when that person is on site at the fault, scanned in.
- 03
Make-safe
Starts at attendance. Stops when the hazard is contained and the area is safe, photographed.
- 04
Resolution
Starts when the ticket is logged. Stops when the fault is fixed, or a temporary fix is in place with a dated plan for the permanent one.
- 05
Closure
Starts at resolution. Stops when the evidence is attached and the supervisor's sign-off is recorded.
A typical SLA rewrite
A landlord's previous FM contract for a Khobar office building said "response within 2 hours" and nothing else. It never said to what, from when, or what counted as a response, so the service charge dispute with tenants ran on anecdote.
We replace it with Annex D: P1 to P4 defined, five clocks, the pause rules, and windows set separately for staffed hours and for out-of-hours mobile cover. The KPI definitions sheet goes into the contract as a schedule.
From the first monthly pack, the landlord can show each tenant the attendance record for their floor by priority, including any ticket that missed and why.
The landlord holds the priority schedule, the clock and pause rules, the KPI definitions sheet, the escalation matrix and a monthly KPI page it can forward without editing.
Four priorities
Whoever logs a fault assigns its priority against these definitions. A change of priority carries a written reason on the ticket, and you see it.
Risk to life, safety or security, or total loss of a critical service
The window is the tightest in Annex D and is set against whether our team is on site when it happens.
- A fire alarm fault on a live panel
- Flooding
- Total power loss
- Cooling lost to a server room or a cold store
Partial loss of a critical service, or a fault likely to become P1
Treated as the P1 it could become if nobody attends.
- One chiller down with the standby carrying the load
- A leaking valve above a switchroom
A defect affecting comfort or function, with no operational stop
The bulk of reactive work in most buildings, and where downgrading usually happens.
- A room not holding temperature
- A failed light fitting
- A blocked washbasin
Minor work that can wait for the next planned visit
Logged, dated and picked up on the planned round rather than with a separate trip.
- Cosmetic damage
- A sticking cupboard door
- Signage
Questions from tenders and lawyers
Why don't you publish response times?
A single number for every site would be untrue for most of them. We publish the structure and write the windows per site, against the staffing and the drive we have actually agreed. A contractor who advertises one response time for every location has either never been to Ras Al Khair or does not intend to meet it there.
What is your SLA achievement rate?
We do not publish one. A percentage you cannot check is marketing. Your site's figures, including the misses, are in every monthly pack from day 30.
What pauses a clock?
Access refused or unavailable on your side, a part on order after you approved its quote, a permit awaited from your authorised person, or production refusing to release an asset. Each pause is recorded with a reason, a time and a name. Our own staff shortage, our travel time, a stock-out of a critical spare we agreed to hold and waiting for our own supervisor are never a pause.
What happens when you miss?
The clock runs whether we like the result or not, so the miss is in the record and in the pack with a cause and an action, and it goes on the review agenda. Service credits or other remedies are agreed in your contract, since they differ by site and by what the window is worth to you. Government buyers should note that the new Government Tenders and Procurement Law takes effect on 2 January 2027, and its implementing regulations will frame remedies on public contracts.
Can we see the clocks live?
Yes. Your view of the record shows every open ticket with its clocks running, not a monthly snapshot of a system you cannot see into.
Can you work to the KPI set in our own tender?
Yes. If your tender defines the KPIs differently, we count to your definitions and say where they leave a gap that our set would close.
Where the numbers come from
Annual maintenance contract
Annex D sits inside the AMC with the register, the calendar and the exclusions.
Read moreFMWhat you receive
The monthly pack the KPI page sits in, and every other document.
Read moreFMEvidence OS
The system the clocks run in, and your live view of it.
Read moreDigitalDashboards and reporting
The KPI page fed into your own dashboards.
Read morePart of one system
Every division keeps records to the same standard, so the second one you need works the way the first one did.
Request an SLA schedule
Tell us the site, its hours and the assets that stop it. You get a draft Annex D with windows written against the staffing and the drive, and the KPI definitions sheet to put in your tender.