SolarSoftwares
Book a demo

What Is Solar Operations and Maintenance (O&M)?

How we keep reviews fair

Quick answer

Solar O&M (operations and maintenance) is the monitoring, inspection and repair work that keeps a PV plant producing. Owners usually buy it under a contract with service levels, such as a yearly availability guarantee. Industry guides often quote that guarantee at 97 to 99%.

Ready to see how it works?

Tell the SolarSoftwares team which tool you are considering. We reply by email to book your demo.

Book a demo
Jump to a section

Key takeaways

  • O&M has three jobs. They are monitoring, planned (preventive) work and repair (corrective) work.
  • Guides quote 97 to 99%, with 98% the most common. About 95% appears for standalone stations or portfolios of small ones.
  • On our invented 2 MWdc plant, one year reads 97.67%, 99.04% or 100%. It depends on how you define it. The guarantee is 98%.
  • Response time ends when a technician arrives, and repair time starts then. A contract should set a repair-time or spares term as well.
  • Of six tools checked, only two document an availability sum with exclusions. The rest handle work orders.

To compare tools that log O&M work orders, see the best solar CMMS software. For availability data, see the best solar monitoring software.

Solar O&M at a glance

O&M is the service after handover. You buy it as a fee plus promises on uptime, speed and reports.

ItemDetail
Core jobsOperations (monitoring, administration), preventive maintenance, corrective maintenance.
Core measuresAvailability (%), response time (hours), sometimes performance ratio, plus a monthly report.
Where it mattersOwner returns, lender and PPA terms, warranty claims and installer service income.
Standard or ruleNo single standard. SolarPower Europe and NREL publish best-practice guides. Each contract sets its own definitions.
In softwareMonitoring platforms, ticketing and CMMS (computerised maintenance management) tools.

Why do O&M terms matter for installers?

The contract says who pays when a plant loses energy. Owner returns, lender and PPA terms, warranty claims and installer service income all depend on how its terms are defined. In the worked example below, the contractor carries only some of the lost energy.

This page covers the contract and how its numbers are measured. For alert rules, see alerts and alarms. For the day-one tests, see the commissioning report page. For the wider process, see commissioning.

What does solar O&M cover?

NREL splits the work into operations and maintenance. Maintenance is preventive or corrective. SolarPower Europe (SPE) adds predictive and extraordinary work.

  • Operations: NREL lists five areas. They are admin, running the plant, work directions, monitoring, and staff knowledge and records.
  • Preventive: planned work. NREL says the schedule depends on the gear, the site and the warranty.
  • Corrective: NREL says it is "required to repair damage or replace failed components".
  • Extraordinary: SPE’s version 3.0 Executive Summary says this is "usually not covered by the O&M fixed fee".

What is in an O&M contract, and how are scope and price measured?

A contract turns scope into numbers. These are a fee, an availability level, response times, remedies and reports. Each needs a clear meaning. If not, nobody can test it.

TermHow it is measured
FeeFixed or variable, with a per-MW sanity check, yearly escalation and a payment schedule
Availability guaranteeAnnual percentage with listed exclusions
Response timeHours, set per fault severity
RemediesLiquidated damages (LDs, agreed payments for a miss) and sometimes bonuses
Liability capShare of contract value or of the annual fee
SparesParts held and who pays
ReportsMonthly. Planned and unplanned work, availability, output against expected
Outside the feeMajor component replacement. Extraordinary work

The terms above draw on the SgurrEnergy contract checklist, the Soltell availability article and the Welsh Government contracts note (the Welsh guide). Where they give figures:

TermFigure or noteSource
Availability guarantee98%, a best-practice minimum over a yearSolarPower Europe v3.0
Availability guarantee98 to 99%Sgurr
Availability guarantee97 to 99%Soltell
Availability guaranteeAbout 95% for standalone stations or portfolios of small onesWelsh guide
Response timeBy severity, no hours givenWelsh guide and Sgurr
Liability capAbout 5 to 20% of contract value, or about 100% of the annual feeSgurr
SparesAbout 0.2% of modules, 12-month warranty on sparesSgurr
  • Remedies: SolarPower Europe v3.0 links availability to both. Sgurr says the market is moving from PR guarantees to availability and response time for independent operators.
  • Spares: the Welsh guide says spares cut avoidable downtime.
  • Outside the fee: LegalDocuments.co.uk guide (components). SolarPower Europe v3.0 (extraordinary work).

The SolarPower Europe figure comes from version 3.0 (printed December 2018), which calls 98% over a year a best-practice minimum. Newer editions exist (version 5.0 in 2021, a sixth in 2025). We did not open them, so check the current edition.

The sources differ. They cover plants of other sizes and years. Sgurr says its benchmarks draw on SPE and market practice. So those two are not independent.

How is an availability guarantee calculated?

Availability is the share of expected output the plant could deliver. The contract figure leaves out downtime. That is time the contractor does not control.

Company pages show two methods. GreenPowerMonitor counts minutes, with manual exceptions. Avathon counts energy. It says contracts typically use the production method: 1 − loss ÷ (actual output + loss). This page uses the reference method: 1 − loss ÷ reference energy. The reference method gives the contractual figure in the worked example. The production method gives about 99.03% on the same events, slightly lower.

  • Exceptions: GreenPowerMonitor lets users enter exceptions that remove periods later.
  • Exclusions: Soltell lists grid blackouts and force majeure. The Welsh guide adds faults in a high-voltage link that is bought on its own.
  • Spares: Sgurr says spare-part delays should not be excluded.

SPE v3.0 says uptime counts all outages. Availability applies exclusions. Its Section 11 formulas are not among the sources read. For a plain meaning and formulas, see availability.

How are response times measured?

SPE v3.0 adds the time to acknowledge an alarm to the time to reach the plant. That sum is response time. Resolution time then runs to the fix. The Welsh guide puts response time on the owner in the availability sum and says to confirm this.

The clocks start at different points. The Welsh guide counts from fault notice to arrival. SPE counts from the alarm. Confirm the start point and who owns each delay.

Step (event A, assumed hours)HoursClock
Alarm to acknowledgement2Acknowledgement time
Acknowledgement to technician on site22Intervention time
Response time (the two above)24Response time
Arrival to fix: 72 waiting for a spare, 48 repairing120Resolution time
Alarm to fix144Total downtime (6 days)

A 24-hour response target is met here. Yet the plant was down for six days. Only a repair-time or spares term covers the other 120 hours. The clocks follow SPE v3.0. The hours are assumed.

Two charts of an invented 2 MWdc plant year: lost energy by four events, and technical, contractual and any-inverter availability against a 98% guarantee.

Figure 1. One year passes or fails a 98% guarantee by definition: 97.67% with every loss, 99.04% after exclusions, 100% if only whole-plant outages count.

Worked example: one year on a 2 MWdc plant

The contractor stocks common spares. The part for event A was not one of them. The owner carries the grid and transformer faults. So the contractor loses 28.77 MWh. That is 47.9% of its 60 MWh allowance.

The assumptions are ours. Reference energy is 3,000 MWh a year (1,500 kWh per kWdc). We spread it evenly at 8.219 MWh a day. Real days vary by season. Winter outages would cost less. Each loss is the share down × days × 8.219 MWh. We then use the energy-based formula above.

EventShare downDaysLost (MWh)Counts against contractor?
A Inverter fault, spare not held25%612.33Yes
B Grid outage100%216.44No (grid)
C Owner-procured HV transformer fault100%324.66No (owner)
D Combiner fuse failures10%2016.44Yes
Contractor total28.77Contractual: 1 − 28.77 ÷ 3,000 = 99.04%
All events69.86Technical: 1 − 69.86 ÷ 3,000 = 97.67%

A 98% guarantee allows 60 MWh of contractor loss. That is about 29 days of a 25% inverter outage. At an assumed 60 USD per MWh, all four events cost about 4,192 USD. The "any inverter" reading gives 100%. The contractor owns no whole-plant outage. Agree which reading the contract uses before you sign.

Check an availability guarantee

This calculator applies the energy-based formula to four events. The example button loads the table above.

Open the calculator on its own page

How solar software handles O&M

Of the six tools we checked in October 2026, only GreenPowerMonitor and Avathon document an availability sum with exclusions. The other four cover work orders, at least in part.

SoftwareAvailability with exclusions?Work orders or tickets?
GreenPowerMonitor (DNV)Yes (time-based)Not stated
Avathon knowledge base (no profile page)Yes (energy-based)Not stated
meteocontrol VCOM CloudNot statedYes
Power Factors UnityNot statedYes
PowerTrackNot statedYes
SolyticNot statedPartly

Not stated: the company pages we read do not say. Partly: the page mentions the feature with few details. What each page says:

  • GreenPowerMonitor (DNV): formula above. You can enter exceptions by hand. That page shows no technical versus contractual split.
  • Avathon: reference-energy and production methods, set at onboarding from contract terms.
  • meteocontrol VCOM Cloud: licence-based CMMS module with work orders, assignees, status and time tracking. Alarms or automations can open tickets. The VCOM datasheet (copyright 2023) lists "recurring events such as annual inspections and vegetation maintenance". The help page does not describe preventive plans.
  • Power Factors Unity: lists "higher availability" as a benefit. Preventive, corrective and emergency work orders, plus dashboards for backlog, response times and SLAs.
  • PowerTrack: work orders created from events with timeline, duration and severity.
  • Solytic: a February 2021 product post says tickets are logged on their own and reports use templates.

For tools that handle work orders and service tickets, see the best solar CMMS software.

Common mistakes with O&M contracts

FAQs

What availability guarantee is normal in solar O&M?

Guides quote 97 to 99%. Of these, 98% is the most common (Soltell). The Welsh guide gives about 95% for standalone stations or portfolios of small ones. Check the exclusions before you compare numbers.

What is the difference between preventive and corrective maintenance?

Preventive work is planned. It keeps gear in good shape. Corrective work fixes damage or swaps failed parts. Confirm that the fixed fee covers both.

What is the difference between response time and resolution time?

Response time runs until a technician reaches the plant. Resolution time runs from arrival to the fix. Contracts can set a target for each, and SPE counts the clock from the alarm.

Does the fixed O&M fee cover everything?

No. LegalDocuments.co.uk puts major component replacement outside the standard fee. SPE says the same of extraordinary work. Read the exclusions as closely as the scope.

Is monitoring the same as O&M?

No. The Welsh guide says a monitoring-only contract gives alarm portals. The owner usually handles the response. Full O&M adds the repair work.

Next step: match your contract’s availability and ticket terms to a tool. Start with the best solar CMMS software for work orders, or the best solar monitoring software for availability data.

Sources

Last updated 30 October 2026. Next scheduled review: 30 January 2027.

See the software on your shortlist.

Tell the SolarSoftwares team what you want in your solar software.

Your request goes by email to the SolarSoftwares team. A demo time is set only after we reply.