Every pay-for-performance programme in Ontario pays on savings that were verified, not savings that were projected. That makes M&V the step between doing the work and getting paid for it — and the step that has to be designed before the work starts, because the baseline disappears the moment anything changes.
A projected saving is an argument. A verified saving is a fact, and increasingly it is the thing being bought. Programmes moved to pay-for-performance because projections were systematically optimistic, and the only defence against that is to measure.
| Programme | What verification controls |
|---|---|
| IESO XLerate | Payment follows approval of the M&V report and requires achieving at least 50% of anticipated savings. Baseline measurement required before any system changes. |
| IESO Energy Performance Program | Three years of annual incentives calculated from measured performance against an auto-generated baseline model, built from twelve months of hourly interval data. |
| IESO EBCx | Implementation pays on confirmed savings; persistence pays on savings still present after twelve months. |
| Enbridge Commercial Custom Retrofit | Paid per cubic metre saved, within eight weeks of commissioning. |
IPMVP — the International Performance Measurement and Verification Protocol — defines four approaches. Choosing the right one is most of the skill, because each trades cost against certainty.
| Option | Method | Use it when |
|---|---|---|
| A — Retrofit isolation, key parameter | Measure the parameter that matters; estimate the rest from stipulated values | The saving depends on one variable you can measure cheaply, such as operating hours on a known load |
| B — Retrofit isolation, all parameters | Measure everything affecting the isolated system | A single system's saving must be defended precisely and other building changes would contaminate a whole-building method |
| C — Whole facility | Compare whole-building metered consumption before and after, weather-normalized | The saving is large relative to total consumption, and the utility meter can actually see it |
| D — Calibrated simulation | A building model tuned to reproduce metered consumption | No usable baseline exists, several measures interact, or the building changed at the same time as the retrofit |
Choosing the option after the work is done. Each option requires different data collected at different times, and most of it has to be collected before anything changes. An owner who installs first and asks about verification afterwards is usually left with Option D and a modelling exercise that could have been a meter.
Everything rests on it. A defensible baseline needs the consumption data, the independent variables that drive it — weather, occupancy, production volume, operating hours — and a stated method for normalizing between the before and after periods. It also needs the non-routine adjustments written down in advance: what happens to the calculation if a tenant leaves, a shift is added, or half the building is renovated mid-term.
Those adjustment rules are where M&V disputes actually happen, and they are cheap to agree before the work and expensive to argue afterwards.
We prepare M&V plans as part of audit reporting — what will be measured, how, against what baseline, over what period, with what adjustment rules — and we prepare the measurement and reporting that programmes require against that plan.
The Certified Measurement and Verification Professional (CMVP) designation, administered by the Association of Energy Engineers, is the recognised credential for IPMVP work, and some owners and programmes require it.
In our October 2026 review of thirteen Canadian firms, two were selling M&V services without naming a CMVP anywhere on their site. So the question to put in writing to any bidder is: "Who on this project holds the CMVP, and will they sign the M&V report?"
Put it to us and we will answer it in writing before you engage us. Where a deliverable requires a CMVP signature, it is delivered by or under a CMVP — we will tell you which, by name, in the proposal. We would rather lose the work than imply a designation.
Partly, and at a cost in both money and certainty. If utility data for the pre-retrofit period exists and the building was otherwise stable, a whole-facility method may still work. If several things changed at once, you are into calibrated simulation, and the uncertainty is wider than anyone wanted.
Programme-dependent. XLerate pays after M&V report approval; the Energy Performance Program runs annual reporting for three years; EBCx persistence monitors for twelve months. Design the plan around the longest obligation, not the shortest.
You do. We will also tell you which of it is worth keeping permanently — the baseline data and the adjustment rules in particular, because they are what the next project will be measured against.
Related
Request a proposal
A compliance deadline, an incentive application, a financing requirement, or a capital plan. Tell us which, and we will tell you what you need and what it will cost — as a fixed fee, before any work starts.
If you do not need an audit, we will say so.