Utility budgeting is one of the most consequential forecasting exercises that commercial property teams undertake each year, and it is also one of the most error-prone. Unlike rent revenue, which is contractually fixed and predictable, utility costs are influenced by a complex set of variables that shift from year to year: rate changes imposed by utility commissions, weather patterns that drive heating and cooling demand, occupancy fluctuations that alter consumption profiles, and equipment performance that degrades over time. A forecast that fails to account for any of these factors can miss the mark by 10 to 20 percent, creating budget variances that erode investor confidence and complicate financial reporting.
The challenge is compounded by the fragmented nature of utility data in most commercial real estate organizations. Utility bills arrive in different formats from dozens of providers. Rate schedules are published in dense tariff documents that require specialized knowledge to interpret. Weather data must be sourced from external providers and normalized against historical patterns. And occupancy projections, which are critical for forecasting consumption, live in leasing systems that are rarely connected to utility management tools.
This guide provides a step-by-step framework for building utility budgets that account for the variables that matter most. Whether you are budgeting for a single building or a portfolio of hundreds, the methodology is the same: start with clean historical data, layer in known rate changes, adjust for weather and occupancy, and build in appropriate contingency margins.
Step 1: Assemble and Clean Historical Consumption Data
Every credible utility budget starts with historical consumption data. You need at least 24 months of actual consumption data for each meter in your portfolio, ideally 36 months. This data serves as the foundation for your forecast because it captures the real-world consumption patterns of your buildings, including seasonal variation, base load characteristics, and the impact of any operational changes that occurred during the historical period.
Before using historical data for forecasting, you must clean it. Estimated reads should be identified and, where possible, replaced with interpolated values based on actual reads from adjacent periods. Billing period irregularities, such as short or long billing cycles caused by meter reading schedule changes, should be normalized to standard 30-day periods. Any anomalous consumption events, such as construction-related spikes or extended outages, should be flagged and excluded from the baseline unless you expect similar events to recur in the forecast period.
Common Data Quality Issues
- Estimated reads: Look for bills flagged with estimated meter readings. These are common during periods when utility companies cannot access meters and can significantly distort consumption patterns.
- Billing period misalignment: Different utilities bill on different cycles. Normalizing all consumption to calendar months ensures consistent comparison across commodities and properties.
- Meter changes: When a meter is replaced, the read sequence resets. If the transition is not properly documented, it can appear as a massive negative consumption event followed by a spike.
- Account changes: Properties that have changed utility accounts during the historical period may have gaps in their data. These gaps need to be filled with estimates based on adjacent periods or comparable properties.
Step 2: Identify and Quantify Rate Changes
Utility rates change every year, and failing to account for these changes is one of the most common sources of budget variance. Rate changes are driven by utility commission proceedings, fuel cost adjustments, renewable energy surcharges, capacity market results, and legislative mandates. The magnitude and timing of these changes vary by utility, commodity, and rate class.
For electricity, the most important rate components to track are the generation charge, transmission charge, distribution charge, and any surcharges or riders. In deregulated markets, the generation charge is set by your energy supplier and may be fixed under a supply contract or floating with market prices. The transmission and distribution charges are set by the local utility and approved by the state public utility commission. These charges typically change once or twice per year based on regulatory proceedings.
For natural gas, the commodity cost is driven by wholesale gas market prices, which can fluctuate significantly based on supply, demand, and weather patterns. Distribution charges are regulated and tend to be more stable. Water and sewer rates are set by municipal authorities and typically change annually, with increases in the 3 to 7 percent range being common in most markets.
Where to Find Rate Change Information
State public utility commission websites publish proposed and approved rate changes for regulated utilities. Most commissions maintain docket tracking systems where you can search for pending rate cases by utility name. Industry associations and energy consultants publish annual rate outlook reports that summarize expected changes across major markets. Your utility account representatives can also provide advance notice of pending rate changes, though the timeline and specificity of this information varies by utility.
Step 3: Apply Weather Normalization
Weather is the single largest source of year-over-year variation in utility consumption for most commercial buildings. A mild winter reduces heating demand, while an unusually hot summer increases cooling costs. If your forecast is based on a historically mild year, you will under-budget for a year with more extreme weather. Conversely, basing your forecast on an extreme weather year will result in budget surplus that, while less operationally problematic, reduces credibility with investors and lenders.
Weather normalization adjusts historical consumption to reflect what it would have been under normal weather conditions. The standard approach uses heating degree days (HDD) and cooling degree days (CDD) to quantify the relationship between weather and consumption. A degree day represents one degree of average daily temperature deviation from a base temperature, typically 65 degrees Fahrenheit. Days colder than the base generate heating degree days, while days warmer than the base generate cooling degree days.
To weather-normalize consumption, you establish the statistical relationship between degree days and consumption using regression analysis on your historical data. This regression produces coefficients that represent the building's sensitivity to weather: how many additional kilowatt-hours or therms the building consumes per degree day. You then apply these coefficients to normal weather data (typically a 10 or 20-year average of degree days for the building's location) to produce a weather-normalized consumption estimate.
A building that consumed 1.2 million kWh in a year with 4,200 CDD might be expected to consume 1.35 million kWh in a normal year with 4,800 CDD. Without weather normalization, your budget would be 12 percent too low.
Step 4: Adjust for Occupancy Changes
Occupancy is the second most significant driver of utility consumption variation after weather. A building at 95 percent occupancy consumes meaningfully more energy and water than the same building at 70 percent occupancy. Your utility forecast must reflect the occupancy levels you expect during the budget period, not the occupancy levels that existed during the historical period.
The impact of occupancy on consumption is not linear. Base building systems such as common area lighting, elevator operations, lobby HVAC, and garage ventilation consume energy regardless of tenant occupancy. This base load typically represents 30 to 50 percent of total building consumption. The remaining consumption is occupancy-driven, scaling with the number of occupied spaces, operating hours, and tenant density.
Building an Occupancy-Adjusted Forecast
- Determine the base load: Identify the minimum consumption level your building maintains at zero tenant occupancy. If you have historical data from periods of low occupancy, this data provides the most reliable base load estimate.
- Calculate the per-unit occupancy increment: Determine how much additional consumption each occupied unit or floor adds to the base load. This can be estimated from historical data by comparing periods with different occupancy levels.
- Apply projected occupancy: Work with your leasing team to obtain the expected occupancy trajectory for the budget period. Account for known move-ins, move-outs, and any speculative leasing assumptions in the operating plan.
- Adjust for tenant type: Different tenant types have different consumption intensities. A data center tenant consumes significantly more electricity per square foot than a standard office tenant. If your occupancy projections include changes in tenant mix, adjust the per-unit increment accordingly.
Step 5: Account for Capital Projects and Operational Changes
Any capital projects or operational changes planned for the budget period that will affect utility consumption should be incorporated into the forecast. Common examples include LED lighting retrofits, HVAC equipment replacements, building automation system upgrades, and changes to operating schedules.
For each project, estimate the expected consumption impact based on engineering calculations or vendor-provided savings estimates. Be conservative in your assumptions: actual savings from energy efficiency projects typically fall 10 to 20 percent below engineering estimates due to real-world operating conditions, tenant behavior, and interactive effects between building systems.
Also account for the timing of project completion. A lighting retrofit that begins in April and completes in August will only deliver partial- year savings in the first budget year. Pro-rate the savings based on the expected completion timeline, and budget the full-year savings for the following year.
Step 6: Build in Contingency and Document Assumptions
Even the most rigorous utility budget will have some degree of uncertainty. Weather can deviate significantly from normal patterns. Rate changes can be larger than projected. Occupancy timelines can slip. Equipment can fail in ways that increase consumption. Building a contingency margin of 3 to 5 percent into your utility budget provides a buffer against these uncertainties without creating excessive budget surplus.
Equally important is documenting every assumption that underlies your forecast. When the budget period ends and actual costs are compared to the forecast, you need to be able to explain variances by pointing to specific assumptions that differed from reality. Did weather deviate from normal? Did a rate increase exceed projections? Did a planned capital project get delayed? Clear documentation of assumptions transforms variance analysis from a blame exercise into a learning exercise that improves future forecasts.
Conduit automates much of this forecasting workflow by maintaining clean historical consumption data, tracking rate changes across utilities and markets, integrating weather normalization into consumption analysis, and producing budget templates that document assumptions alongside the numbers. The platform allows property teams to build utility budgets in hours rather than weeks, with confidence that the underlying data and methodology are sound.