Utility Rate Optimization: Are You on the Wrong Tariff?

Two identical buildings with identical consumption can pay very different electricity bills, and the reason is often the rate they are on. Rate class and tariff choice are set once, usually when an account opens, and then rarely revisited even as the building's load changes. A facility that grew, shrank, or shifted its operating hours can spend years on a rate that no longer fits.

Rate optimization is the discipline of checking that fit. It asks a few concrete questions: Is this account in the correct rate class for its demand? Is it on the best available price plan for its usage pattern? And would switching, where the utility allows it, actually lower the bill? None of these can be answered from a single invoice. They need a load profile and a comparison.

This guide walks through rate classes, the difference between time-of-use, tiered, and demand rates, the signs a facility is on the wrong one, and the data you need to model a switch with confidence. The examples lead with Ontario and Canadian utilities, where the rules are well documented.

Rate classes: how utilities sort customers by size

Utilities assign each account to a rate class based on demand and sometimes annual energy, and the class determines whether you are billed mostly on energy or on demand. In Ontario the pivotal line is 50 kW. Hydro One assigns non-residential accounts to energy billing when average monthly maximum demand is less than 50 kW, and to demand billing when it is equal to or greater than 50 kW, at which point the variable delivery charge is based on peak demand rather than total kWh.

Above that line, classes ladder up by size. Hydro One's structure runs from general service under 50 kW, to 50 to 4,999 kW, to Large Use at 5,000 kW and above. Toronto Hydro uses four business classes: General Service under 50 kW, 50 to 999 kW, 1,000 to 4,999 kW, and Large Use over 5,000 kW, and notes your rate class may change depending on your average monthly demand. BC Hydro classifies on both demand and annual energy, with Medium General Service covering 35 to 150 kW and under 550,000 kWh per year and Large General Service above those levels.

The single most important threshold across Ontario distributors is that 50 kW switch from energy billing to demand billing, because it changes which lever controls your bill entirely.

Time-of-use, tiered, and demand rates

For smaller commercial accounts on Ontario's Regulated Price Plan, the choice is between three price structures. The Ontario Energy Board (OEB) sets these prices; the figures below are effective November 1, 2025 and run to October 31, 2026.

• Time-of-use (TOU) prices vary by hour. Per the OEB, off-peak is 9.8 cents, mid-peak 15.7 cents, and on-peak 20.3 cents per kWh, with on-peak and mid-peak windows swapping between the winter (November 1 to April 30) and summer (May 1 to October 31) seasons.
• Tiered prices vary by volume. Non-residential customers pay 12.0 cents per kWh for the first 750 kWh each month and 14.2 cents above that.
• Ultra-Low Overnight (ULO) rewards overnight load. ULO is 3.9 cents from 11 p.m. to 7 a.m. daily, but on-peak (weekdays 4 to 9 p.m.) is 39.1 cents, nearly double TOU on-peak.

The OEB publishes all three on its electricity rates page, and confirms that eligible customers may select from TOU, Tiered, or ULO prices. The right choice depends entirely on shape: a facility that runs steadily through the day may do better on tiered, one that can defer load to nights may win with ULO, and one with a pronounced off-peak profile benefits from TOU. Larger accounts move into demand-based rates, where the peak kW discussed above dominates.

Signs a facility is on the wrong rate

A few patterns reliably signal a mismatch worth investigating:

• Demand consistently below the class threshold. Toronto Hydro lets a customer in the 50 to 999 kW class request a review if monthly average peak demand falls below 50 kW for five consecutive months.
• Demand consistently above a threshold you are not billed on, suggesting a class or rate that would price your peak differently.
• A flat, steady load on a TOU rate, which captures little off-peak benefit and may fare better on tiered.
• A load that could shift to nights but sits on TOU rather than ULO.
• A low load factor, meaning short spikes against a low average, which drives demand charges and ratchets up quickly.

Toronto Hydro publishes the reassignment mechanism directly: a customer can request a review if monthly average peak demand falls below 50 kW for five consecutive months. Utilities do not usually move you to a cheaper class on their own, so catching the pattern is on you.

Load factor: the number that points to the right rate

Load factor is the clearest single indicator of whether a demand rate or an energy rate suits a facility. It is average demand divided by peak demand, calculated as monthly kWh divided by (peak kW times days in the period times 24 hours), and it is always below 1. A high load factor means steady, predictable use; a low one means spiky use that idles capacity.

The link to rates is direct. As EnergyCAP puts it, meters with consistently high load factors often qualify for preferential low-cost rate schedules, so always check whether yours does. The economic principle, per a standard reference, is that electrical rates are designed so that customers with high load factor are charged less overall per kWh) because their load imposes lower system cost. A high-load-factor site often benefits from a demand rate; a spiky, low-load-factor site is punished by one.

The data you need to model a rate change

A credible rate comparison is a modelling exercise, and it needs the right inputs:

1. Twelve or more months of bills, to capture seasonal swings in both energy and demand.
2. Interval (smart meter) data, so consumption can be re-priced hour by hour under each candidate rate.
3. A demand profile, to test class thresholds and demand-charge exposure.
4. A TOU consumption breakdown by on-peak, mid-peak, and off-peak, for both seasons.

In Ontario this data is available by right. The OEB explains that Green Button is a standardized data format giving residential and business customers an option to easily access and securely transfer their energy usage data, and that rate-regulated utilities were required to provide access by November 1, 2023. The standardized file can include up to a maximum of two years of interval readings, which is exactly the input a rate model needs. We cover the format in detail in our Green Button guide.

How MartinAI helps

Rate optimization stalls on data assembly. The information you need is spread across a year of bills in different layouts, plus interval files in a separate format, and pulling it together by hand for a portfolio is where most reviews die. MartinAI reads utility bills across every commodity and layout, capturing rate class, demand, consumption, and the charges tied to each into a clean whole-building record, and validating them so the comparison starts from correct numbers.

With that record, you can flag accounts sitting on the wrong side of a class threshold, calculate load factor by meter, and compare each site's usage shape against the available rate structures. MartinAI produces the clean, structured cost history that rate modelling, benchmarking, and reporting depend on, so a switch is proposed on evidence and defended with a year of real data.

Conclusion

The right rate is not the one you were assigned years ago, it is the one that matches how the building uses power today. Check the class threshold, know your load factor, and compare your real usage shape against the TOU, tiered, and demand options open to you. The utility will not do it for you, but with a clean load profile in hand, the answer is usually clear, and sometimes it is worth a lot.