The follow-up report for our recent solar case estimated 8,898 kWh of annual PV generation. That is a useful number for planning. To find out what the system actually produced, and what it saved on the bill, the next step is to follow the meter records.
In the paired EnerGuide case, the net rating improved by 30 GJ/year while rated consumption was almost unchanged. The homeowner sheet showed a renewable contribution of 32 GJ/year, rounded, alongside the 8,898 kWh estimate. Those figures describe the model. No inverter production records, bills or matched billing periods were included in that review.
Start with the question you want answered
“Did the solar work?” can mean three different things: did it produce the expected electricity, did the household buy less from the grid, or did the owner spend less? Each question needs its own comparison. A successful production year can coincide with higher household use, a different electricity rate or a new electric vehicle.
A utility meter does not see every kilowatt-hour
Some solar electricity is used in the house before it reaches the grid meter. That direct use reduces imports, but it is not usually shown as a separate production reading on the electricity bill. The inverter or a dedicated production meter is the starting point for generation.
For a simplified system with no storage or other generation, using consistent AC measurements over the same period: household electricity use = grid imports + PV generation − grid exports. If only a reliable net-import energy quantity is available, adding same-period PV generation can reconstruct use under these assumptions. Separate import and export records are still needed to analyse self-consumption and price the flows where rates differ.
A battery adds charging, discharging, storage changes and losses to the accounting. Multiple meters or generators also change the boundary. Draw what each meter measures before adding the readings together.
The record set worth asking for
| Record | What it establishes | Check before comparing |
|---|---|---|
| PV production export | Recorded generation over the selected period | AC versus DC basis, missing intervals, resets and system downtime |
| Utility import and export data | Electricity crossing the grid connection | Actual versus estimated reads; matching start and end dates |
| Commissioning and change dates | When the comparison can begin | Operational start date, equipment changes and outages |
| Bills and tariff history | Prices and charges that applied | Import rates, export credits, fixed charges and taxes |
| Household and weather context | Other reasons consumption might change | Occupancy, heating, new loads and comparable seasons |
For this home, those are the next records to review. Their absence from the present analysis is a limit on the conclusion, not a finding that the system underperformed.
Compare like periods before calculating savings
Use a complete operating year where possible. A partial first year should be labelled as such rather than quietly scaled into an annual result. Match the production dates to the import and export dates, and retain gaps rather than filling them with an unlabelled average.
A before-and-after import difference can still be useful, but it combines the effect of solar with changes in household demand and weather. If the aim is to estimate savings caused by solar, state the reference: what would grid purchases have been over the same period without it? Use a documented method, and explain the effect of any uncertain inputs.
NRCan explains why household utility use need not match an EnerGuide rating: standard operating assumptions differ from actual occupancy, weather and some loads. A mismatch is a prompt to inspect those differences, not an automatic verdict against either record.
Then work out the dollars
Generation used directly can avoid purchases; exported electricity can earn a credit. The applicable rates may differ, and fixed charges may remain. Apply the actual tariff to the relevant quantities instead of multiplying the rating improvement by a single electricity price.
That calculation establishes an operating-bill effect. A return-on-investment calculation goes further: it also needs the installed cost, financing, incentives, maintenance assumptions and the period being assessed.
For this case, the next useful result would be a measured production total and a reconciled import/export balance for the same period. Those would let the conversation move from what the model expects to what the operating records show.
Source and scope
The 8,898 kWh/year figure and rounded 32 GJ/year contribution come from the follow-up Homeowner Information Sheet for the anonymized 2026 case. This article lays out a proposed check of its operating performance; it does not report measured production, a bill discrepancy or achieved savings. The diagram is an energy-accounting guide, not a representation of verified site wiring. Private property identifiers and source files are omitted.
