A production line starts, chillers recover after a shutdown, and air compressors ramp up at the same time. The event may last only 15 minutes, yet it can influence a large share of the month’s electricity bill. That is why are demand charges high is a question that deserves more than a tariff-level answer. For medium-voltage industrial and commercial users, demand charges reflect the cost of being ready to supply power at the site’s highest moment of need.
Unlike energy charges, which apply to every kilowatt-hour consumed, demand charges are generally based on the highest measured demand in kilowatts during a billing period. A site can therefore reduce total energy consumption and still face high electricity costs if its maximum demand remains unmanaged.
Why are demand charges high?
Demand charges are high because electricity networks and generation assets must be sized for peaks, not averages. When a facility draws a sudden, substantial amount of power, the utility must have sufficient capacity across cables, substations and generation resources to serve it reliably. The charge helps recover the cost of maintaining that capacity.
For a factory, warehouse, hospital, data-intensive operation or large commercial estate, the financial effect can be material. A single peak can set the billing demand for the month, depending on the tariff structure. Some tariffs also apply ratchet clauses, where a portion of a previous high demand influences future bills. In practice, an avoidable operating event can become a recurring cost exposure.
The rate itself varies by tariff, voltage level, location and contract arrangement. A high demand charge is not always a billing error, nor is it always a sign of waste. A site operating genuinely energy-intensive equipment to meet production commitments may need a high capacity level. The opportunity is to separate necessary demand from unplanned or poorly timed demand.
Demand is measured in time intervals, not monthly averages
The most common misunderstanding is treating demand as a monthly consumption figure. It is a rate of use. If a site consumes 20,000 kWh over a month, that figure says little about whether its highest 15-minute or 30-minute interval was 300 kW or 1,200 kW.
A brief overlap between loads can create the highest interval. Starting multiple motors, reheating electric furnaces, charging a battery fleet, or recovering HVAC after a scheduled setback can all raise demand sharply. The site may consume no more energy overall, but the timing of consumption makes the bill more expensive.
Operational coincidence drives expensive peaks
High demand charges often originate in normal operating decisions made across different teams. Production prioritises output. Facilities teams prioritise comfort and uptime. Engineering responds to process requirements. Each decision can be sensible in isolation, while their combined load creates an unnecessary peak.
Typical contributors include simultaneous equipment start-ups, compressor cycling, chilled-water plant demand, poor power-factor management where applicable, electric vehicle charging, and equipment operating outside planned schedules. Solar variability can also matter. If cloud cover reduces on-site generation while large loads rise, grid import can climb quickly.
This is why monthly bills alone are insufficient for diagnosis. Decision-makers need interval data, asset-level visibility and operational context. Without them, a maximum-demand charge can look like a fixed cost rather than a controllable outcome.
The tariff can amplify the cost of one peak
Tariff design determines how strongly peak demand affects the final bill. Some arrangements apply a straightforward charge per kilowatt of measured maximum demand. Others have time-of-use components, seasonal pricing, contracted-capacity thresholds or penalties for exceeding an agreed level.
For high maximum-demand users, this creates a trade-off. Reducing the declared or contracted capacity too aggressively can expose the business to excess-demand costs or operational risk. Keeping too much capacity may mean paying for headroom that the site rarely needs. The right position is based on measured load behaviour, production requirements and resilience planning, not a generic target.
Finance teams should also consider that demand reduction has a different value from energy reduction. Cutting one kilowatt-hour saves the unit energy rate. Preventing a sustained peak can reduce a capacity-related charge across an entire billing period. The strongest business case often combines both.
How to identify avoidable demand charges
Start by reviewing at least 12 months of billing data alongside interval-meter data. Look for the highest demand interval each month, when it occurred, and whether the event aligns with production, weather, shift changes or equipment start-up routines. A recurring pattern points towards a controllable operating condition. An isolated spike may require event investigation.
Then compare demand against on-site solar output, battery state of charge and key asset performance. If a site has solar but its peak occurs after generation declines, solar alone will not resolve the demand issue. If a battery is available but is discharged too early, held in reserve without a defined strategy, or unable to respond at the right interval, its financial value will be constrained.
The most useful analysis identifies a demand limit that protects operations. Setting the limit too low can force unnecessary load curtailment or battery cycling. Setting it too high delivers little bill reduction. It should be reviewed as production volumes, operating hours and equipment change.
Smarter control turns visibility into savings
Monitoring shows what happened. Demand control determines what happens next. For industrial and commercial energy users, that means using real-time data and forecasting to anticipate a developing peak before it is recorded by the meter.
An enterprise-grade energy platform can coordinate solar prioritisation, battery dispatch and flexible-load control around a defined demand threshold. When demand approaches the limit, the system can discharge a BESS, defer non-critical loads, adjust HVAC setpoints within approved comfort parameters, or sequence equipment start-ups more intelligently. Once the risk has passed, loads can return to normal operation under controlled rules.
This approach must be engineered around site realities. Critical production assets, safety systems and quality-sensitive processes should not be treated as generic flexible loads. Battery control also requires careful planning around state of charge, degradation, reserve requirements and expected tariff windows. Peak shaving that compromises uptime is not a commercial success.
Amsolar AI combines real-time analytics, forecasting and automated control with practical solar and BESS expertise. Its Management Reporting System brings energy, environmental and operational signals into one reporting environment, helping teams move from reactive bill reviews to measurable action. The objective is real savings, not theoretical reductions: lower avoidable demand, stronger asset visibility and dependable operational performance.
Demand management supports more than lower bills
Reducing avoidable peaks can improve planning across operations, finance and sustainability. It gives energy managers a clearer view of how production behaviour affects grid reliance. It helps finance teams forecast utility exposure with greater confidence. It also improves the quality of Scope 2 reporting by connecting electricity consumption with the conditions driving it.
Demand management is not a one-off optimisation exercise. New machinery, altered shifts, seasonal cooling loads and expanded solar capacity can all change the site’s load profile. Regular reporting, transparent alarms and periodic control reviews keep the strategy aligned with the business.
The next high-demand interval will not wait for the next monthly bill. Treat it as an operational event that can be predicted, measured and managed, while keeping the capacity your business genuinely needs.
