Field service KPIs are the metrics a service organization uses to measure how efficiently it resolves customer issues and how much value the service business generates. The core operational set is first time fix rate, mean time to repair, resolution time, technician utilization and avoidable dispatch rate; the commercial set covers service contract attach rate, renewal rate and the share of profit coming from aftermarket services.

Most published KPI guides are written for residential trades — HVAC, plumbing, electrical — where the job is short, the asset is simple and the technician is dispatched from a van. This page is written for the other case: industrial equipment manufacturers, machine builders and OEM service organizations, where a single asset can be highly configured, downtime is expensive, and the expert who knows the machine is rarely near it.

Every benchmark below is attributed to a named source with a year. Where sources disagree, both are shown rather than one being picked.

Field service KPI benchmark table

KPIFormulaWhat good looks likeSource (year)
First Time Fix Rate(Jobs resolved on first visit ÷ Total jobs) × 100Median 75%, top 20% 86%, bottom 20% 53% (30-day window)Aquant, 2025 Field Service Benchmark Report
First Visit Repair RateOn-site incidents resolved in one visit ÷ on-site incidentsIndustry median 87%TSIA (2021)
Resolution TimeCase closure date − case creation dateMedian 5 days, top 20% 3 days, bottom 20% 11 daysAquant (2025)
Avoidable Dispatch Rate(Dispatches that did not need to happen ÷ Total dispatches) × 100Median 14%, top 20% 3%, bottom 20% 24%Aquant (2025)
Recovery time after unplanned downtimeTime from stoppage to production restarting81 min average, up from 49 min five years earlierSiemens / Senseye (2024)
Technician / billable utilization(Productive or billable hours ÷ Total available hours) × 10075–85% typical; Pacesetters ~90.2%TSIA (2025)
Engineer utilizationAs above, engineering populationIndustry average 83%; Pacesetters 90.6%TSIA (2025)
Time Between Service VisitsMean days between visits to the same assetMedian 86 days, top 20% 131 days, bottom 20% 50 daysAquant (2025)
Service contract attach rate(Units sold with a service contract ÷ Units sold) × 100Overall average 48.9%; industrial equipment only 32.7%TSIA (2021)
Aftermarket profit shareService profit ÷ Total profit40–50% of overall profits at many manufacturersDeloitte (2020)
Installed base under contract(Units under contract ÷ Installed base) × 10080%+ at service-focused manufacturersDeloitte (2020)

Last reviewed July 2026. Aquant’s benchmark reports are vendor-published — Aquant sells field-service AI — but the methodology is disclosed: 157 service organizations, more than 21 million service events, 6.5 million assets, over 602,000 technicians and an average of three years of data per company. Siemens’ figures come from 181 completed online interviews with maintenance, engineering and IT professionals at large industrial organizations across automotive, FMCG, heavy industry and oil & gas, covering April 2019 to March 2023.

The operational KPIs, and what each actually tells you

First time fix rate

First time fix rate is the percentage of jobs resolved on the first visit. It is the most widely tracked field service KPI because a failed first visit consumes cost, capacity and customer goodwill simultaneously. Aquant’s data quantifies that: when the first visit fails, the job takes an average of 2.7 total visits and adds roughly 13 days to resolution time.

The measurement window is where most comparisons fall apart. Aquant recommends 30 days and warns that 7- or 14-day windows split related repeat visits into separate tickets, making performance “look better than it is.” If your first time fix rate is above 90%, check the window before celebrating.

Industrial machinery sits at 71% in Aquant’s 2025 industry breakdown — below the 75% all-industry median, alongside medical devices at 69%. Complex, highly configured capital equipment is simply harder to diagnose correctly from a phone description.

Mean time to repair

MTTR measures how long restoring a failed asset takes. Its weakness as a benchmark is that the R is ambiguous — repair, recovery, respond or resolve — and each starts a different clock. Mean time to recovery includes detection, waiting and travel; mean time to repair often counts only hands-on work. The customer experiences the former.

The most striking industrial data point on this metric is that it is moving in the wrong direction. Siemens found that recovery after unplanned downtime now takes an average of 81 minutes, up from 49 minutes five years earlier — even though the number of incidents fell over the same period, from 42 a month per facility to 25. Siemens attributes the slower recovery partly to skilled maintenance labour lost during the post-COVID “great resignation,” and partly to supply-chain difficulty sourcing emergency replacements.

That is worth reading carefully, because it separates two different problems. Fewer failures is a reliability achievement. Slower recovery is an access-to-expertise problem.

Resolution time

Resolution time is the elapsed time between case creation and closure — the metric that most closely matches what the customer perceives. Aquant’s 2025 median is 5 days, with the top 20% at 3 days and the bottom 20% at 11 days: bottom performers take almost four times longer than top performers.

Resolution time and first time fix rate should always be read together. A team can post a strong first time fix rate while customers wait a fortnight for the visit.

Technician utilization

Utilization is productive or billable hours divided by total available hours. TSIA reports that most field service organizations run 75–85% billable utilization, with Pacesetter organizations at approximately 90.2%, and an engineer utilization industry average of 83% against 90.6% for Pacesetters.

Utilization is the KPI most often pushed past its useful range. Capacity that is 95% committed cannot absorb an emergency, so response times lengthen and overtime costs rise. It is a metric to keep in a band, not to maximise.

Avoidable dispatch rate

Avoidable dispatch rate is the share of on-site visits that did not need to happen. Aquant puts the median at 14% — roughly one in seven truck rolls — with the top 20% of organizations at 3% and the bottom 20% at 24%. Top performers are therefore about eight times more efficient on this measure than the bottom quintile, which is a far wider spread than most operational KPIs show.

This is the metric that most directly captures the shift the rest of this page describes, and it has a companion figure in the same dataset: Aquant found 33% of service queries are solvable without a service professional’s help at all.

The KPIs service organizations are adding

Industry conference decks increasingly contrast “traditional” service metrics with a new set built around remote resolution, prediction and outcomes. The framing is genuinely useful. The practical caveat is that these metrics are at very different stages of maturity, and it is worth being honest about which ones you can actually benchmark.

Measurable today, with published benchmarks:

  • Avoidable dispatch rate — median 14% (Aquant, 2025).
  • Cost of downtime in currency, not hours — see below.

Measurable internally, but with no published industry benchmark:

  • Remote resolution rate — the share of cases closed without any site visit. There is no credible published benchmark for this figure; the closest proxy is avoidable dispatch rate. Track it against your own baseline and define whether “remote” includes phone-only resolution.
  • Percentage of cases diagnosed remotely before dispatch — a leading indicator for first time fix rate rather than an outcome in its own right.

Not yet benchmarkable, and worth treating with caution:

  • Percentage of incidents predicted and predictive accuracy — these require condition-monitoring infrastructure, and no industry benchmark for them was found in published sources.
  • Percentage of energy savings achieved — a real objective, but not currently a standardised service KPI.

The honest position is that these last items are strategic intentions rather than comparable measures. Adopting them as KPIs is reasonable; presenting a number for them as an industry benchmark is not, because no such benchmark exists.

Cost of downtime

Expressing downtime in currency rather than hours is the change that most often shifts a service conversation from cost centre to risk management. Siemens’ 2024 figures give the range: a lost hour costs $36,000 in fast moving consumer goods at the low end and $2.3 million in automotive at the high end — more than $600 a second — while SME manufacturers can reach $150,000 an hour. Across the world’s 500 biggest companies, unplanned downtime is estimated at $1.4 trillion a year, equal to 11% of revenues, up from 8% in 2019.

For an OEM selling service contracts, this number is the customer’s number, not yours — which is exactly why it belongs in the KPI set. It sizes what your response time is worth to them.

Commercial KPIs: from warranty to recurring service revenue

The commercial half of the picture is where industrial service organizations have the most headroom, and the benchmark data says so bluntly.

TSIA reports an overall average initial service contract attach rate of 48.9% — but a breakdown that should give machine builders pause: healthcare technology achieves 53.8%, while industrial equipment manages only 32.7%. Roughly two-thirds of industrial equipment leaves the factory without an attached service contract.

The prize for closing that gap is well documented. Deloitte finds that aftermarket operating margins average about 2.5 times the operating margin on new equipment sales, that many manufacturers generate 40–50% of overall profits from services, and that manufacturers who focus on services often have 80%+ of their installed base under service contracts. Deloitte also reports that manufacturing leaders anticipate the business model shifting from roughly 75:25 in favour of products today to 50% products and 50% outcomes.

Academic work points the same way. Research from Aston University’s Advanced Services Group, funded by the Economic and Social Research Council and published in January 2026, found that for every one percentage-point increase in the share of revenue earned from services rather than products, firms experience over 2% total revenue growth and almost 2% profit growth.

The connection back to the operational KPIs is direct: contracts are only profitable if the cost to serve them is low enough. An organization with a 14% avoidable dispatch rate and a 71% first time fix rate has a very different cost-to-serve than one at 3% and 86% — and can therefore offer responsive service tiers that the first organization cannot price.

What these KPIs trade off against each other

This is the part most KPI lists omit, and it is where measurement programmes usually go wrong. Field service metrics are not independent, and improving one in isolation frequently degrades another.

  • Utilization vs. response time. Utilization above the mid-80s removes the slack that same-day response depends on. Pushed far enough, a “productivity” gain shows up as an SLA breach.
  • Measurement window vs. honesty. Shortening the first-time-fix window improves the reported number without improving anything real. Aquant’s own analysis is explicit that sub-30-day windows overestimate first time fix rate.
  • Average handle time vs. dispatch avoidance. Compressing average handle time in the contact centre suppresses exactly the extra questions that would have avoided an on-site visit. A saved four minutes can buy a $500 dispatch.
  • First time fix rate vs. resolution time. A visit deferred until the right part and person are available raises first time fix rate and lengthens the customer’s wait. Neither number alone tells you whether that was the right call.
  • MTBF vs. MTTR. These are different projects with different owners — reliability engineering versus service execution. Reporting them as one “uptime” figure hides which lever is actually stuck.

The practical response is to track metrics in pairs, and to fix the definitions before comparing anything to an industry benchmark.

How to choose which KPIs to track

  1. Start with two operational and one commercial. First time fix rate and resolution time, plus contract attach rate, will surface most problems worth having.
  2. Write down the definition, including the window. Most benchmarking arguments are definitional, not factual.
  3. Add a cost figure the customer recognises. Downtime cost per hour converts service performance into their language.
  4. Only then add leading indicators such as the share of cases diagnosed remotely before dispatch.
  5. Review the benchmark sources annually. These reports are re-issued, and figures move.

Where remote assistance moves these numbers

AR remote assistance affects a specific and limited part of this KPI set, and it is worth being exact rather than expansive.

It moves: avoidable dispatch rate and the share of cases resolved without a visit, because an expert can see the asset before a truck is committed. First time fix rate, because the fault and the required part are identified from an observed diagnosis rather than a verbal description. The diagnosis and expert-travel components of MTTR, because the senior person joins in minutes instead of after a scheduling and travel window — which addresses precisely the expertise gap Siemens identifies behind slower recovery times. The productive share of technician and expert time, because windshield hours fall. And indirectly, contract economics, because a lower cost to serve makes responsive service tiers viable in geographies where dispatch was never economic.

It does not move: parts lead time or inventory fill rate. MTBF or equipment reliability. And it does not predict failures — VSight performs no sensor-based prediction and no IoT condition monitoring, so the “percentage of incidents predicted” family of metrics is outside what it can influence.

VSight Remote puts a remote expert on a technician’s or customer’s live camera view with AR annotation directly on the equipment, on phones, tablets and smart glasses, with customer-facing sessions opening in the mobile browser without an app install. VSight Workflow delivers procedures as standardized digital work instructions. VSight is a connected worker platform, and is GDPR, HIPAA and ISO 27001 certified.

If you want to size the effect on your own numbers rather than take a vendor’s word for it, the field service ROI calculator works from your inputs.

Frequently asked questions

What are field service KPIs? Field service KPIs are the metrics a service organization uses to measure how efficiently it resolves customer issues and how much value the service business generates. The core operational set is first time fix rate, mean time to repair, resolution time, technician utilization and avoidable dispatch rate; the commercial set covers service contract attach rate, renewal rate and the share of profit coming from aftermarket services.

What is the most important field service KPI? There is no single most important one, and organizations that pick one usually distort it. First time fix rate is the most widely used because it sits at the intersection of cost, capacity and customer experience — a failed first visit consumes all three at once. But it is only meaningful alongside a measure of how long the customer waited, because a high first time fix rate achieved by slow scheduling is not good service.

What is a good first time fix rate? Aquant’s 2025 benchmark data across 157 service organizations puts the median at 75%, the top 20% at 86% and the bottom 20% at 53%, measured over a 30-day window. TSIA reports a higher median of 87% for on-site incidents resolved in one visit. The two are not directly comparable because they measure different populations and windows, so 75–86% is the realistic working range.

How much does unplanned downtime cost per hour? It varies enormously by sector. Siemens reports the cost of a lost hour at $36,000 in fast moving consumer goods at the low end and $2.3 million in automotive at the high end, with SME manufacturers reaching $150,000 an hour at the top end. Across the world’s 500 biggest companies, unplanned downtime is estimated at $1.4 trillion a year, or 11% of revenues.

Which field service KPIs are service organizations adding? The measurable additions are avoidable dispatch rate — the share of on-site visits that did not need to happen, with an Aquant median of 14% — and the cost of downtime expressed in currency rather than hours. Several metrics discussed at industry conferences, such as percentage of incidents predicted and percentage of energy savings achieved, currently have no published industry benchmark, so they can only be tracked against your own baseline.

Do field service KPIs conflict with each other? Yes, and this is the most common measurement failure. Pushing technician utilization too high removes the slack that same-day response depends on. Shortening the first-time-fix measurement window flatters the number by splitting related repeat visits into separate tickets. Optimising average handle time in the contact centre can suppress the questions that would have avoided a dispatch. Metrics should be read in pairs, not chased individually.

First time fix rate · MTTF and MTTR · MTBF · truck roll · expert utilization rate · service level agreement · first-contact resolution · average handle time · uptime and downtime · OEE · field service management

Sources