Field service KPIs and how to calculate them

A customer complains that this is the third time someone has come to look at the same machine. The service manager checks and finds that the first visit was a diagnosis, the second was missing a spare part and the third was a different technician who didn’t know the model. Each visit was “closed”, so the monthly report looks fine. The problem isn’t missing data: it’s that nobody is measuring it with a formula.

In field service, what you sell is hours and visits, and they slip away easily. These six indicators tell you where. For all of them we’ll use the same hypothetical month: 8 technicians, 210 closed orders, 22 working days of 8 hours.

1. First-time fix rate (FTF)

FTF = orders resolved on the first visit / eligible orders × 100

“Eligible” means you explicitly exclude the ones that didn’t depend on you: customer not present, out-of-scope work, cancellations. Write the rule down and don’t change it every month.

Suppose that out of 210 orders you exclude 10 because the customer wasn’t there. Of the remaining 200, 164 were resolved on the first visit: FTF = 164 / 200 = 82%.

How to read it: every point of FTF is one less visit’s worth of travel, hours and customer patience. When it drops, look at the reason for the second visit: if it’s “missing part”, the issue is van stock; if it’s “couldn’t fix it”, the issue is assignment or training.

2. MTTR (mean time to repair)

MTTR = total repair hours / number of repaired orders

Define when the clock starts and stops. A practical definition: from when the technician starts work until the equipment is operational, adding up every visit on that order.

In the month, the 164 orders fixed the first time took 2.5 h on average (410 h), and the 36 that needed a return took 6.5 h (234 h). MTTR = 644 h / 200 = 3.2 h.

How to read it: the average hides what matters. Orders with a second visit take more than twice as long, so improving FTF lowers MTTR almost automatically. Also look at MTTR by equipment type and by technician.

3. SLA compliance

Compliance = orders handled within the deadline / orders under SLA × 100

Measure the response deadline (arriving or making contact) and the resolution deadline separately, because they fail for different reasons.

Suppose 180 of the month’s orders were under an SLA contract. 171 got a response on time and 162 were resolved on time:

Deadline	Within	Total	Compliance
Response	171	180	95%
Resolution	162	180	90%

How to read it: if response is fine and resolution isn’t, the bottleneck is diagnosis or spare parts, not dispatch. Also check how many orders are due soon today: the monthly percentage arrives too late to save them.

4. Cycle time

Cycle time = close date and time − creation date and time

This is what the customer experiences. Breaking it into segments shows where the time goes. Suppose this typical order:

Segment	Hours
Creation → assignment	4
Assignment → arrival	18
On-site work	3
Close-out (report, signature, sign-off)	24
Total	49

How to read it: in the example, only 3 of 49 hours are work; the rest is waiting. The 24-hour administrative close-out also delays invoicing. Use the median as well as the average: a few never-ending orders distort the average.

5. Technician utilization

Utilization = productive hours / available hours × 100

Available hours: 8 technicians × 22 days × 8 h = 1,408 h. Decide whether travel counts as productive and show it both ways:

• On-site work only: 644 / 1,408 = 45.7%
• On-site + travel (300 h): 944 / 1,408 = 67.0%

How to read it: a big gap between the two figures points to poorly planned routes or areas that are too wide per technician. Utilization very close to 100% isn’t good either: it leaves no room for emergencies and ends up breaking the SLA.

6. Contract profitability

Contract margin = contract revenue − (hours × loaded hourly cost + parts + transport)

The loaded hourly cost includes the technician’s salary plus social and labor charges, divided by their available hours; the components and percentages should be set by your accountant. Suppose it comes to Bs 80 per hour.

Item	Contract A	Contract B
Monthly revenue	Bs 12,000	Bs 8,000
Hours (× Bs 80)	40 h = Bs 3,200	70 h = Bs 5,600
Parts	Bs 2,500	Bs 1,800
Transport	Bs 800	Bs 1,400
Margin	Bs 5,500 (45.8%)	−Bs 800 (−10%)

How to read it: contract B looks like “a good customer” because it generates a lot of work, but it loses money. Before renewing it, check whether the problem is repeat visits (low FTF), old equipment that eats spare parts or a badly calculated rate.

What to record on every visit

No indicator comes out right if the source data doesn’t exist. On every order and visit you need:

• Creation date and time, priority and the SLA that applies.
• Contract, customer and equipment serviced.
• Assigned technician and assignment time.
• Departure, arrival, work start and work end times.
• Outcome: resolved or not; if not, the reason (part, skill, customer absent, diagnosis pending).
• Parts and materials used.
• Customer signature or sign-off and close time.

Common mistakes

1. Counting visits instead of orders. FTF is calculated on orders; if you count visits, repeats look like productivity.
2. Changing exclusions every month so the number looks better.
3. Logging hours from memory at month end. Real hours are recorded during the visit, not afterwards.
4. Measuring the SLA only at month end, when you can no longer save any order.
5. Calculating profitability without transport or parts, which on contracts with old equipment are most of the cost.
6. Looking at indicators in isolation. A low MTTR with a low FTF can mean quick visits that fix nothing.

If you want to automate it

With discipline, all of this can live in a shared spreadsheet. As the team grows, an ERP like MKA can record times, signature and parts for every visit from the technician’s phone and calculate first-time fix, MTTR, SLA compliance and utilization on its own (see field service).