Commercial building energy audit and management field guide

Energy audit, and why it starts with the data

An energy audit is a systematic study of where a commercial building wastes energy and money, ending in a ranked list of fixes and what each one saves. The point is not the report. It is a prioritized set of actions, sorted by payback, that an owner can fund in order.

What separates an audit from a guess is that it starts with measurement. You cannot manage what you do not measure, and the first measurements are already sitting in the utility bills and in one number computed from them, the energy use intensity. Those tell you whether the building has a problem and roughly where the problem lives, before anyone walks a mechanical room.

From the data the audit moves to the building, finds the energy conservation measures, ranks them by payback, and proves the savings afterward with measurement and verification. ASHRAE Standard 211 frames the levels of that work, and this guide covers the program around it. For the metering that feeds it, see the submetering and energy monitoring guide. For the analytics that find faults in the trend data, see the building automation fault detection and diagnostics guide.

Where does an energy audit start?

An energy audit starts with the data, not the walk-through. That one rule separates a useful audit from a tour of the building. Before anyone looks at a light fixture, you pull at least twelve months of utility bills, compute the energy use intensity, and benchmark it against similar buildings. The data tells you whether there is a problem worth chasing and where the big users are.

The reason is money and time. A walk-through finds what you can see. The bills find what is actually expensive, which is often invisible on a tour: a base load that never drops at night, a demand spike nobody scheduled, gas use in a month the heat should be off. Walk first and you spend the day on burnt-out lamps while the chiller runs all weekend on an empty building.

Find the big users first. The bills and the benchmark point you at the systems that move the number, so the survey time lands where the savings are. Walking before the data is the most common way an audit wastes the owner's money on small stuff and misses the large stuff.

What is energy use intensity (EUI)?

Energy use intensity, EUI, is a building's total annual energy divided by its floor area, expressed in kBtu per square foot per year. It rolls electricity, gas, steam, and oil into one figure, so a 40,000 sq ft office and a 200,000 sq ft office can be compared on the same footing. A low EUI means the building uses less energy per square foot. A high one means there is room.

The number alone means little until you benchmark it. ENERGY STAR Portfolio Manager compares your building against a national sample of similar buildings and returns a 1 to 100 score, where 75 or above means the building beats 75 percent of its peers. The national median source EUI for a U.S. office runs around 85 kBtu per square foot per year, a hospital far higher, a warehouse far lower. The peer comparison, not the raw EUI, tells you whether the building is a hog or already efficient.

You need twelve or more months of bills to compute it honestly, so a full heating and cooling cycle and the seasonal swings are captured. EUI is the starting number. It will not tell you which valve leaks, but it tells you whether the building deserves a Level 2 audit or just a tune.

Reading the utility bills

The utility bills are the cheapest insight in the whole audit, and most owners have never read them past the amount due. Pull twelve to twenty-four months of every meter, electric and gas, and separate the two things the electric bill charges for: energy, the kilowatt-hours used, and demand, the highest kilowatt draw in the billing period. They are priced differently, and a building can carry a fine energy number and a brutal demand charge from one short spike.

Plot the monthly use against weather and against occupancy. A few patterns jump out. A base load that stays high overnight and on weekends means equipment runs when the building is empty. A summer spike that dwarfs winter points at cooling and the demand charge. A gas bill that never falls near zero in July means heat is leaking on somewhere, often a reheat valve or a simultaneous heat-and-cool fight.

Read the rate schedule too. The building may be on the wrong tariff, paying for demand it could shift, or missing a time-of-use opportunity. None of this needs a site visit. It is an hour with the bills, and it shapes what you go looking for when you do walk the building.

What are the ASHRAE audit levels?

The ASHRAE audit levels, defined in ANSI/ASHRAE/ACCA Standard 211, sort an energy audit into three depths so the effort matches the decision. Level 1 is a walk-through plus the bill analysis, good for quick wins and a ballpark. Level 2 is a detailed survey that puts a savings number and a cost on each measure, enough to decide what to fund. Level 3 is investment-grade, the detailed engineering and modeling behind a large capital project that has to pencil out.

The levels build on each other. A Level 2 includes the Level 1 work, and a Level 3 zeroes in on a few high-cost measures and engineers them to the accuracy a lender or a performance contract needs. Rough cost scales with depth, commonly cited around a few cents per square foot for a Level 1 and several times that for a Level 2, with Level 3 priced per the measures studied.

Match the level to the decision, and confirm the scope against Standard 211 and whatever the incentive program or the lender requires. Standard 211 sets the minimum content of each level; the adopted program rules and the contract control what you actually deliver. Do not order a Level 3 to decide whether to fix a schedule, and do not try to finance a chiller replacement off a Level 1 walk-through.

What each level actually buys

Each level buys a different grade of certainty, and paying for more than the decision needs is its own kind of waste. A Level 1 buys a cheap, rough read: the low and no-cost opportunities, a rough savings range, and a sense of whether a deeper audit is worth it. The numbers are directional, not bankable.

A Level 2 buys decision-grade numbers. Each measure gets a savings estimate, an installed cost, and a payback, built from a real survey of the equipment, the schedules, and the controls, often with some short-term metering or trend data behind it. This is the level most owners actually need, because it is enough to choose which measures to fund and in what order.

A Level 3 buys bankable numbers. The few measures that carry a large capital cost get detailed engineering, calibrated energy modeling, and a savings estimate solid enough to support a loan or a measurement-and-verification clause in a performance contract. It costs the most, and it is overkill for a setpoint change. Under-audit and you fund a project on numbers that do not hold. Over-audit and you spend engineering money to confirm a fix you should have just made.

What is an energy conservation measure (ECM)?

An energy conservation measure, ECM, is a specific change to a building that cuts energy use, from a free schedule fix to a new chiller. The audit's job is to find them, estimate what each saves and costs, and rank them by payback. ECMs sort into a ladder, and you climb from the bottom, because the bottom is where the cheap savings are.

The first rung is no and low-cost: correcting setpoints, fixing schedules, tuning controls, and retro-commissioning equipment back to how it was meant to run. These cross-link straight into the fault detection work, because a control fix is only as durable as the monitoring that holds it in place. The second rung is retrofit: LED lighting and lighting controls, variable frequency drives on fans and pumps, added sensors. The third rung is capital: a new chiller or boiler, envelope work, a full building automation system.

Rank by payback, not by what is exciting. A schedule fix that costs nothing and saves five percent beats a glamorous capital project with a fifteen-year payback every time. The ladder is also the order of operations: capture the free savings first, because they shrink the load the expensive equipment has to serve.

The free savings come first

The biggest return in most audits is the no and low-cost work, and it is the work owners skip on the way to a shiny capital project. Schedules, setpoints, and operations are usually free or close to it, and they are usually where the waste hides. A building running its air handlers, lighting, and pumps twenty-four hours a day in a space occupied ten hours a day is paying for fourteen hours of nothing, every day, and no new equipment fixes that. A schedule does.

This is the part the data already flagged. The overnight base load on the bill, the gas use in summer, the demand spike at three in the morning, those are operational, and they cost little to correct. Studies of existing-building commissioning commonly report energy savings in the range of 10 to 20 percent with paybacks often under two years, and very little of that is new hardware.

Fix the operations before you spend the capital. Every kilowatt you stop wasting at no cost is a kilowatt the new chiller never has to be sized for. Owners who jump to capital first buy bigger equipment than they need and leave the free money on the table.

Operations waste more than worn-out equipment

Operations waste more energy than tired equipment does, and people underestimate it because nothing is visibly broken. The chiller works. The boiler works. They just run when they should not, fight each other, or drift out of tune while everyone assumes the controls are handling it.

The usual suspects are a short list. Night setback that never engages, so the building conditions itself empty. An economizer stuck closed, so the system makes mechanical cooling on a 55-degree day instead of opening a damper to free outside air. Simultaneous heating and cooling, where a reheat valve leaks and a zone heats the same air the central plant just cooled. Setpoints and schedules that drift back after every service call until the building runs on overrides nobody remembers setting.

Every one of these leaves a clear fingerprint in the trend data, and finding them at scale is exactly what fault detection and diagnostics does. See the building automation fault detection and diagnostics guide for that layer. The audit catches them as a snapshot. FDD catches them continuously, which matters because operational waste does not stay fixed on its own.

What the walk-through covers

By the time you walk the building, the data has told you where to look, and the walk-through confirms and fills in what the bills cannot see. Work it systematically so nothing gets skipped: lighting, HVAC, the envelope, the controls, the schedules, the plug load, and how the place is actually operated.

Lighting means the lamp types, the wattage, the hours they burn, and whether any controls exist. HVAC means the equipment, its age and condition, the controls, whether the economizer works, and what the schedules say versus what the building does. The envelope means insulation, glazing, and the air leakage you can feel at a door on a windy day. Plug load means the equipment tenants plug in, which has grown into a real slice of the number and is the hardest to control.

The point of the walk is to connect the expensive patterns in the data to physical causes. A high overnight base load on the bill becomes the air handler you find running in an empty wing. The summer demand spike becomes the chiller staging you watch happen. Walk with the bill analysis in hand, not as a blank tour.

Lighting: the easy retrofit

Lighting is usually the easiest retrofit to justify, because the savings are simple arithmetic: hours burned times watts saved. Swap a 32-watt fluorescent troffer for an LED at half the wattage in a space lit twelve hours a day, and the saving per fixture is predictable and the payback is often short. Add occupancy sensors and daylight controls and you also cut the hours, which the lamp swap alone does not.

The LED conversion is the most common first retrofit for a reason. The technology is mature, utility rebates for it are widely available, and the payback commonly lands in the one to three year range depending on the rate and the hours. Spaces that run long hours, a warehouse, a parking garage, a 24-hour facility, pay back fastest because the hours do the work.

Do not stop at the lamp. Controls are where the second tranche of savings hides, and they cross-link to the same scheduling and occupancy logic that runs the HVAC. A fixture that is efficient but on all night in an empty room is still waste.

HVAC: the largest target

HVAC is usually the largest energy user in a commercial building, so it is where the largest savings live and where the audit spends most of its survey time. The target splits into equipment efficiency, controls, the economizer, and the schedules, and the controls and schedules are almost always the cheaper win.

Equipment efficiency is the capital side: an aging chiller or boiler running well below the efficiency of a current unit. That is a real measure, but it carries a long payback and it belongs after the free fixes, because a right-sized replacement depends on first cutting the load the old unit was carrying. The controls and the economizer are the fast side. An economizer that actually opens on a mild day can cut a large slice of cooling energy at no equipment cost. A schedule that matches occupancy stops the plant from serving an empty building.

Check what the system does, not what the sequence says it does. The drawings describe intent. The trend data describes reality, and the gap between them is the savings. This is the same expected-versus-actual comparison the fault detection guide runs continuously.

The envelope: slow and capital

The envelope is the slow, expensive end of the audit: insulation, air leakage, and glazing. These measures are real and they cut the heating and cooling load at its source, but they trend toward capital cost and long paybacks, so they rank below the operational and lighting work in most buildings.

Air leakage is the one worth chasing early, because it is sometimes cheap to fix and large in effect. A loading dock door that does not seal, a missing weatherstrip, a stairwell that stacks air out the top of the building, these leak conditioned air all year and can sometimes be fixed for the cost of a gasket. Insulation and window replacement are different. They carry serious cost and the payback often runs well past a decade on energy alone, which is why they usually ride along with a re-roof or a facade project rather than standing on their own.

Treat the envelope as the load you reduce before you size new equipment, not as the first place you spend. The exception is a building with an obviously failed envelope, where the comfort complaints and the energy waste point at the same fix.

How do you rank ECMs by payback?

You rank ECMs by simple payback: the installed cost divided by the annual dollar savings, which gives the number of years to recover the cost. A measure that costs $5,000 and saves $5,000 a year pays back in one year. The ranked list, free first, then the fast paybacks, then the capital, is the real deliverable of the audit. The report is just the wrapper around it.

Simple payback is the common screening tool because owners understand it and it sorts a list fast. It ignores the time value of money and the equipment life, so a thorough Level 2 or 3 also runs return on investment or life-cycle cost on the larger measures, where a ten-year payback on a twenty-five-year chiller can still be a good investment. Use simple payback to triage and the fuller analysis to defend the capital.

The prioritized list is what an owner funds against. Put the no-cost operational fixes at the top, because they pay back almost immediately, the lighting and VFD retrofits next, and the capital projects last and only where the numbers and the equipment life support them. An audit that hands over savings with no payback ranking has not finished the job.

Measure it: submetering and data logging

You cannot improve what you cannot see, so the measure step puts instruments on the building to find out which system uses what. The utility bill gives you the whole building. It does not tell you that the kitchen is a third of the load or that one rooftop unit runs around the clock. Submetering and short-term data logging break the total apart.

Permanent submeters on major loads, the chiller plant, the lighting panels, the kitchen, the tenants, turn a single monthly number into a continuous picture of where the energy goes. For an audit you often do not need permanent meters everywhere. A few weeks of portable data loggers on the questionable circuits will localize the waste well enough to size a measure. Either way the principle holds: the meter turns a suspicion into a number you can put a payback on.

The metering side has its own guide. See the submetering and energy monitoring guide for the current transformers, the accuracy classes, and the commissioning that decides whether the reading is true. The audit consumes that data. It does not have to reinvent the metering.

What is measurement and verification (M&V)?

Measurement and verification, M&V, is the practice of proving that an energy measure actually saved what it was supposed to, by comparing measured energy before and after the change. It exists to settle one argument: the owner paid for a fix, the bill did not obviously drop, and now someone has to show whether it worked. M&V is how you avoid the it-did-not-save dispute.

The recognized framework is the International Performance Measurement and Verification Protocol, IPMVP, which defines four options. Options A and B isolate the affected equipment and measure it, A by metering the key parameter and estimating the rest, B by metering all of it. Option C uses the whole-building utility bills, which suits a package of measures large enough to move the meter. Option D uses a calibrated computer simulation when there is no clean baseline to measure against.

Pick the option to fit the measure and whatever the incentive program or the performance contract requires. IPMVP and the program rules control the method, not preference. M&V is rarely free, so it is reserved for measures where the savings are large, contested, or tied to a payment. A $200 schedule fix does not need Option C. A million-dollar performance contract does.

The M&V baseline

The baseline is the before picture, and getting it right is most of M&V. It is the building's energy use before the measure, but a raw before-number is not enough, because the year after a retrofit is never the same as the year before it. A hotter summer, a new tenant, longer hours, any of those move the bill independent of the fix.

So the baseline gets normalized. You adjust for weather, usually by regressing energy against heating and cooling degree days, and for occupancy or production, so the comparison is apples to apples. What you report is avoided energy, what the building would have used without the measure minus what it actually used, not the naked difference in the bills. The bill can rise while the measure still saved, if the weather or the load grew, and the normalized baseline is what proves it.

Establish the baseline before the work, not after, because once the measure is in you cannot go back and meter the old condition. This is the step that turns a savings claim into a number a skeptical CFO or a utility program will accept.

Behavior, operators, and training

The cheapest savings are behavioral, and they are the first to evaporate. People leave lights and equipment on, prop doors, and override thermostats, and an operator who does not understand the control sequence will fight it until it runs on manual. None of that shows up as broken equipment. It shows up on the bill.

The fixes here cost little and stick poorly without attention. Operator training is the one that pays back, because the building engineer who understands why the night setback exists is the person who keeps it from being overridden away. Simple feedback, a dashboard, a monthly number, occupancy reminders, moves behavior more than a memo does. An audit that recommends only hardware misses the slice of savings that comes from running the building the way it was designed to run.

Treat the human side as a measure with its own persistence problem. The behavior reverts, the operators turn over, and the savings fade unless someone keeps measuring and reminding. That is why behavior belongs in the ongoing program, not in a one-time report.

Rebates, incentives, and the ESCO

The measures only happen if someone pays for them, so the audit should hand the owner the funding paths alongside the payback list. Utility rebates and incentives are the first stop, because they cut the installed cost directly and shorten the payback. Most utilities offer prescriptive rebates for common retrofits like LED lighting and VFDs, and custom incentives for engineered savings, and the rules, the rates, and the application paperwork are program-specific, so confirm what the local program actually offers before you bank on it.

For larger work, an energy service company, an ESCO, can finance the project through a performance contract, where the contractor guarantees the savings and gets paid out of them. That guarantee is exactly why M&V matters: the payment depends on a verified savings number, so the contract specifies the IPMVP option up front. Performance contracting suits a big capital package an owner cannot fund out of pocket, and it shifts the savings risk to the contractor, at a cost.

Match the financing to the measure. Free operational fixes need no financing. A lighting retrofit often pencils on rebate plus payback alone. A deep capital project is where the ESCO and the performance contract earn their place.

Why the savings fade, and how to keep them

Savings fade, and the fade is the failure nobody plans for. The setpoints drift back after a comfort complaint. The schedules get overridden during a one-off event and never reset. The economizer linkage seizes a year after it was fixed. Within a couple of years a building tuned to save can be back where it started, and the only sign is a bill creeping up that no one is watching.

This is why an audit is a snapshot and a program is the answer. The savings from operational measures, the cheap ones with the best payback, are exactly the ones that revert, because they live in settings anyone can change. Hardware savings persist better. A high-efficiency motor stays efficient. Control and behavior savings have to be held in place.

Holding them in place is monitoring. Re-audit on a cadence, and put continuous fault detection on the trend data so a reverted setpoint raises a flag in days instead of surfacing in next year's bill. See the building automation fault detection and diagnostics guide for that monitoring layer. Savings you do not monitor are savings you are slowly giving back.

From one audit to an energy program

A single audit is a snapshot. An energy management program is the loop that keeps the building improving. The cycle is steady: benchmark the data, audit to find the measures, implement them in payback order, verify the savings with M&V, monitor so they persist, and re-audit when the building or its use changes. Energy management is continuous, not a one-time report that goes in a drawer.

The loop is what separates a building that saves once from a portfolio that keeps saving. Each pass updates the benchmark, catches the measures that drifted, and surfaces the next tier of opportunity that was not worth funding last cycle. The data improves every loop, because you are metering more of the building and learning where it actually wastes.

A program runs on records and follow-through, which is where a field tool earns its place. FieldOS gives the team a way to log the audit findings, assign the measures as work, capture the before-and-after photos and readings, and keep the M&V and rebate paperwork attached to the building instead of scattered across inboxes. The program is only as durable as the records behind it.

What to record

An energy program lives or dies on its records, because the savings you cannot prove are the savings you will be arguing about a year from now. Keep the bills, the EUI history, the audit report, the ranked ECM list, the M&V baseline and results, and the rebate applications together and attached to the building.

The reason is continuity. Buildings change hands, engineers turn over, and the memory of why the schedule is set the way it is walks out the door with the person who knew. A field tool like FieldOS keeps the record with the asset: log the audit, assign each measure as a tracked job, attach the before-and-after readings and photos, and store the verified savings and the incentive paperwork where the next person can find them.

Common mistakes

• Walking the building before analyzing the bills and benchmarking the EUI.
• Over-auditing or under-auditing for the decision the owner actually faces.
• Jumping to a capital project before capturing the free operational fixes.
• Treating ASHRAE and ENERGY STAR figures as mandates rather than guidance to confirm against the program.
• Skipping measurement and verification, so the savings cannot be proven.
• Letting savings fade because nobody monitors the setpoints and schedules.
• Handing over savings numbers with no payback ranking to prioritize them.

Field checklist

Standards and references

The audit framework comes from ANSI/ASHRAE/ACCA Standard 211, which defines the content of a Level 1, 2, and 3 commercial building energy audit and replaced the older audit guidance with language a jurisdiction can adopt. It is the reference for what each level must contain. Confirm the edition in force and any program-specific scope layered on top of it.

Benchmarking runs on ENERGY STAR Portfolio Manager, the EPA tool behind the 1 to 100 score and the national EUI medians. Measurement and verification follows the International Performance Measurement and Verification Protocol, IPMVP, maintained by the Efficiency Valuation Organization, which defines Options A through D. The federal M&V guidelines build on the same protocol for government projects.

Hedge the figures to their source. The audit levels are ASHRAE's. The EUI medians and scores are ENERGY STAR's. The M&V methods are IPMVP's. The rebate rules, the rates, and the application requirements belong to the local utility or program and change often, so confirm them rather than quoting a number. A certified energy auditor or energy engineer signs off where a project, a lender, or an incentive program requires credentialed work. Start with the data and benchmark the EUI before the walk, capture the free operational savings before the capital, and prove and keep the savings with M&V and monitoring.

Units, terms, and conversions

Energy auditing carries a stack of terms and units that shift between a utility bill, an ASHRAE report, and an incentive application, so the same idea reads differently across the paperwork.

Energy shows up as kilowatt-hours and therms on the bills and as kBtu once it is rolled together for EUI. Demand is measured in kilowatts, separate from energy and priced separately. EUI is kBtu per square foot per year. Savings get expressed as energy, as dollars, and as a payback in years, and the three do not always agree on which measure wins.

Energy audit

A systematic study of where a building wastes energy and money, ending in energy conservation measures ranked by payback.

EUI / benchmark

Energy use intensity, a building's annual energy per square foot in kBtu, compared against similar buildings to judge performance.

ASHRAE audit levels

Standard 211's three depths: Level 1 walk-through, Level 2 detailed survey, Level 3 investment-grade engineering.

Energy conservation measure (ECM)

A specific change that cuts energy use, from a free schedule fix to a capital equipment replacement.

Simple payback

Installed cost divided by annual dollar savings, the years to recover the cost; the common ranking screen.

Measurement and verification (M&V)

Proving actual savings by comparing normalized energy before and after a measure, per the IPMVP options A to D.

Retro-commissioning

Tuning existing equipment and controls back to proper operation; an audit finds the waste, retro-commissioning corrects the operational part.

Persistence

Whether a savings holds over time; operational and control savings drift back without monitoring.

Related tools

Calculators and readiness checks for this work

Compare your options

FAQ

People also ask

• What is electrical submetering?
• What is fault detection and diagnostics?
• What is roof asset management?
• What is anti-icing vs de-icing?
• What is an access control system?