Commercial solar performance is easiest to manage when generation is recorded as an operating measure rather than checked only after a problem appears. This guide explains what businesses should monitor, how to compare output with a reasonable baseline, how to investigate underperformance, and when to create evidence for an installer or warranty claim.
Overview
A solar PV system can produce a different amount of electricity every day without anything being wrong. Weather, daylight hours, roof orientation, shading, soiling, temperature, grid restrictions and planned shutdowns all affect output. The useful question is not whether this month matches last month, but whether the system is behaving as expected for its size, site conditions and season.
For a business, this matters for more than energy savings. Lower generation can increase electricity purchased from the grid, reduce the value of on-site consumption and affect export income. A reliable monitoring routine also gives an operations team evidence when speaking to an installer, maintenance contractor, inverter manufacturer or warranty provider.
Start by recording the system’s design information: installed capacity in kilowatt-peak (kWp), panel layout, commissioning date, inverter model, battery capacity if fitted, known shading, and the meter or monitoring platform used. Keep the original design estimate and handover documents with these records. If you are assessing a new system, the guide to comparing commercial solar installers and quotes explains which performance information should be requested before work begins.
What to track
1. Generation and system availability
Record total solar generation in kilowatt-hours (kWh) for each month. Where the monitoring platform allows it, also record daily generation, peak output and any periods when the inverter was offline. A monthly total alone may hide a short outage, so an occasional review of the daily graph is worthwhile.
For larger sites, note the output of each inverter or monitoring zone. Comparing zones can help identify a fault affecting one roof area, string or inverter rather than the whole installation. Keep generation separate from consumption: a business may use less solar on site because of a quiet trading period, but that does not mean the panels generated less.
2. Normalised output
Divide generation by installed capacity to calculate specific yield:
Specific yield = solar generation in kWh ÷ installed capacity in kWp
This makes comparison more useful across sites or between different stages of an expansion. For example, a 100 kWp system producing 8,000 kWh has a specific yield of 80 kWh/kWp for that period. Use the same capacity basis every time, and note any panels or inverters that were added, removed or unavailable.
3. Site and operational conditions
Add brief notes beside the monthly figure. Useful entries include prolonged cloud, snow, exceptional heat, heavy dust, nearby construction, new shading, roof work, access restrictions, planned shutdowns and grid outages. For commercial premises, also record changes to operating hours, weekend use and machinery demand. These factors may alter self-consumption without changing PV generation.
4. Energy flows and financial measures
If the system includes a battery, record solar generation, electricity imported from the grid, electricity exported and battery charge or discharge where available. A battery can change when solar electricity is used, but it does not increase the energy produced by the panels. Check that the monitoring platform’s definitions are clear: “solar production”, “load”, “self-consumption” and “export” are not interchangeable.
Businesses should also track the value of avoided purchases and export payments separately. Export arrangements can change, so review the relevant tariff terms rather than assuming that a previous payment rate still applies. See the guide to comparing SEG tariffs and solar export payments for a separate review of that issue.
Simple monthly performance log
| Month | Generation (kWh) | Specific yield (kWh/kWp) | Inverter availability | Weather or site notes | Action |
|---|---|---|---|---|---|
| January | Record reading | Calculate | Check alerts | Record relevant conditions | None or investigate |
| February | Record reading | Calculate | Check alerts | Record relevant conditions | None or investigate |
| March | Record reading | Calculate | Check alerts | Record relevant conditions | None or investigate |
Extend this table through the year and save the underlying monitoring export where possible. For a multi-site business, use the same fields at every location.
Cadence and checkpoints
Daily or weekly: look for interruptions
Assign a named person to check whether the monitoring portal is receiving data and whether the inverter shows an active fault. This does not require a detailed production analysis every day. A missing data connection is not automatically a generation failure, but it should be distinguished from an inverter shutdown promptly.
Monthly: record and normalise
At the end of each month, save the generation figure, calculate specific yield and compare it with the same month in previous years if those records exist. Compare like with like: a month with a planned shutdown, incomplete data or a changed system should be marked rather than treated as a clean benchmark.
Quarterly: inspect patterns
Review the previous three months together. Look for one inverter consistently trailing other inverters, repeated alarms, a sudden change in the shape of the daily output curve, or a growing gap between generation and the original design expectation. A quarterly review is also a sensible time to check whether roof work, new plant, vegetation or site changes have introduced shading.
Annually: reconcile performance and documents
Compare annual generation with the original forecast, allowing for the assumptions used in that forecast. A forecast is not a guarantee: roof orientation, shading, local weather and equipment availability all affect results. Confirm that the system capacity, monitoring records and export statements agree, and review whether planned maintenance has been completed. The commercial solar maintenance and monitoring schedule can help organise this wider review.
Indicative reference bands
The table below is a planning aid, not a performance guarantee. UK solar yield varies considerably by location, orientation, tilt, shading, design quality and availability. Use the original site-specific forecast as the primary reference where it is available.
| Comparison | What it can tell you | What it cannot prove |
|---|---|---|
| Current month versus same month last year | Whether output has changed under broadly similar seasonal conditions | Whether weather or temporary downtime explains the difference |
| One inverter versus another | Whether a fault may be localised to a zone or equipment group | Whether the inverters have identical orientation, shading or capacity |
| Actual specific yield versus design forecast | Whether further technical review may be justified | Whether the forecast assumptions were realistic or still apply |
How to interpret changes
Begin with the simplest explanation and avoid treating a single low-output day as evidence of a fault. Cloud cover and short winter daylight can create large normal variations. Instead, investigate a persistent change, a sudden step down, repeated alarms or a mismatch between inverter readings and meter data.
- Check the data first. Confirm the monitoring period, time zone, meter reading and communications status. Look for missing intervals or a platform reset.
- Check the inverter. Read the displayed status and fault code without opening electrical equipment. Photograph the screen or save the alert, including the date and time.
- Compare zones. If only one inverter or roof section is affected, record the difference and avoid assuming that all panels are underperforming.
- Review site conditions. Consider new shade, dirt, leaves, snow, construction, roof access restrictions or a change in the grid connection.
- Check energy-flow definitions. A battery control change or altered business hours may reduce export while generation remains normal.
- Escalate with evidence. Contact the responsible installer or maintenance provider with dates, readings, alerts, photographs and a description of what changed.
Do not climb onto a roof, remove panel covers or open an inverter enclosure to investigate. Commercial PV systems contain electrical hazards and may remain live in daylight. Use a qualified professional for physical inspection and electrical testing.
A useful evidence pack includes the system design and commissioning documents, serial numbers, monitoring exports, monthly log, inverter alerts, meter readings, weather or site notes, maintenance history and any photographs taken from a safe location. State the practical impact clearly: for example, “inverter two has recorded no generation since the logged date” is more actionable than “the panels seem weak”.
When to revisit
Keep the log active throughout the year, with a short monthly entry and a deeper quarterly review. Revisit the baseline whenever the system changes: adding a battery, replacing an inverter, expanding the array, changing monitoring software, altering the grid connection or carrying out roof work can make earlier comparisons misleading.
Update the review after a prolonged outage, repeated inverter warning, major shading change, unexplained fall in specific yield or a maintenance visit. Check warranty and installer procedures at the same time, because the required evidence, notification route and response process depend on the equipment and contract. Do not wait until the end of a warranty period to report a persistent issue.
For your next review, download the last 12 months of generation data, calculate monthly specific yield, mark every known outage, and compare each inverter or monitoring zone. If the pattern remains unexplained, request a documented technical assessment from a suitably qualified commercial solar professional. For related planning work, see the guides to solar payback and savings, commercial battery storage and comparing solar quotes.