What Utility Data a Commercial Solar Assessment Needs

A commercial solar utility bill analysis needs more than a total annual electric spend. It needs a traceable record of which utility account serves which load, what each meter measured, which rate and riders produced each bill, how electricity use changed over time, and where the source data are incomplete or unrepresentative.

The practical starting package is:

• original electric bills for every relevant account and meter across at least one complete operating and seasonal cycle, when available;
• the current utility name, rate schedule, riders, and effective source for each account;
• a map connecting accounts, physical meters, buildings, tenants, and loads;
• the original interval export for each meter, if the utility or facility can provide it;
• an operating chronology covering shutdowns, vacancies, new shifts, equipment changes, construction, and expected future loads;
• separate production, import, export, and storage records when on-site energy equipment already exists; and
• a manifest, exception log, and assumption register that follow the data into the model.

One complete cycle is a starting point, not proof that the data are sufficient. A prior tenant, an estimated bill, a meter exchange, a new production line, or an existing solar array can make a full-looking dataset misleading. The right question is not “Do we have 12 bills?” It is “Can a reviewer reproduce what each record means and see every material limitation?”

This guide owns that input-readiness decision. It does not calculate system size, predict savings, assign a tariff, or produce a return on investment. Those decisions require a defined technical and financial scope after the records are validated.

Build the utility-data package before asking for a result

An owner should be able to answer four questions before a solar model begins:

1. Coverage: Do the records represent every account, meter, and load in the proposed project boundary?
2. Identity: Can every value be traced to a named utility, account, meter, time period, channel, and unit?
3. Representativeness: Does the observed period resemble the facility the owner expects to operate, or are there known breaks in continuity?
4. Limitations: Are gaps, transformations, and assumptions documented rather than silently repaired?

Use a status table to make the decision explicit:

Status	Meaning	Next action
Usable as provided	Source, identity, time basis, units, and coverage are documented for the intended question	Preserve the original and include it in the handoff
Usable with stated limitation	The record supports some analysis, but a known issue limits interpretation	State the affected dates, meters, and model decisions
Investigation required	A mismatch, missing field, or unexplained discontinuity could change the result	Ask the utility, facility team, meter vendor, or analyst to resolve it
Not representative	The record is real but reflects an operating condition that should not become the only base case	Keep it, explain it, and define a separate approved case
Missing	The expected record is not available	Document the request, possible substitute, and consequence

Do not let “received” become a synonym for “validated.” A folder can contain a year of PDFs and a large CSV while still omitting a second account, confusing kilowatts with kilowatt-hours, or mixing a pre-expansion facility with the proposed operating plan.

Start with an account-and-meter map

Commercial properties often have more than one electrical boundary. A campus may have separate services for production, offices, exterior lighting, a warehouse, a pump, or a tenant. A consolidated statement may collect several accounts on one invoice. Submeters may divide internal use without corresponding to a utility billing meter. The property owner may control some accounts while tenants control others.

Create one row for every utility account, physical billing meter, owner submeter, generation meter, and storage data stream that could enter the assessment.

Manifest field	What to record
Utility	Full legal provider name shown on the current bill
Account	Owner-controlled label; keep the exact account identifier in the controlled source record
Service location	Address, building, suite, parcel, or equipment served
Meter	Meter identifier and whether it is a billing meter, submeter, generation meter, or virtual allocation
Account holder	Entity named on the account and the party authorized to request records
Rate	Schedule code, rider names, voltage or service class if shown, source URL/document, and checked date
Data channel	Imported energy, exported energy, net energy, demand, generation, reactive power, storage, or another named value
Coverage	First and last bill/interval dates and any known gaps
Change history	Meter replacement, tenant change, account transfer, consolidation, service modification, or equipment change

Walk the map with someone who knows the facility. Match bill identifiers to meter labels and physical loads when access and responsibilities allow. Do not assume that similar service addresses refer to the same meter or that the account called “main” supplies the whole property.

This step also prevents double counting. If a common-area submeter is already included behind a utility master meter, adding both series creates a load that never existed. If one invoice is merely a payment summary for five underlying accounts, the invoice total should not be added again after the five accounts are modeled.

The serving utility cannot be assigned from a South Carolina city name alone. Provider territories, cooperative service, municipal service, and individual account arrangements can vary. Verify the legal utility on the current bill and use the Sunburst utility hub only as a starting point for provider-specific information.

Preserve the original bills, not just a summary

A worksheet with monthly kWh and dollars is convenient, but it removes evidence the analyst may need. Retain the original statement for every billing period and make the worksheet a traceable index rather than the sole source.

Capture at least:

• bill-period start and end dates and the number of billed days;
• current and prior meter reads when shown, including actual or estimated status;
• energy consumption and its unit;
• every billed-demand field and unit;
• utility name, account, service address, and meter identifier;
• rate schedule, riders, voltage/service class, or other rate identifiers shown;
• fixed, minimum, demand, energy, time-of-use, power-factor, fuel, tax, adjustment, and credit lines that may be relevant to the later model;
• meter multipliers when shown;
• unusual corrections, rebills, late charges, deposits, or non-energy items; and
• a file name that identifies the account and bill period without overwriting earlier statements.

The U.S. Department of Energy explains that utility costs depend on the site’s load magnitude and shape and its bill structure. DOE also notes that bills commonly identify the rate, monthly kWh, peak demand, and historical use. DOE’s utility-rate evaluation guidance supports collecting both distribution and supply bills when those charges arrive separately.

Total dollars alone cannot preserve that structure. Two months with similar consumption can have different costs because the number of days, time pattern, billing demand, fixed components, adjustments, or effective tariff differed. Conversely, a bill comparison can be distorted by a deposit, prior balance, late fee, or credit unrelated to facility load.

If a bill shows only a short historical bar chart, treat it as a visual cross-check, not a replacement for the statements behind those months. A bar chart may omit exact dates, demand, rate changes, estimated reads, and adjustment lines.

Verify the rate schedule without assigning one

The current bill should identify the rate schedule or give the account information needed to ask the utility. Then check the current official tariff source and record the date of that check. Keep the source document with the handoff rather than copying only a price into a spreadsheet.

That distinction matters in South Carolina. Dominion Energy South Carolina’s current rates and tariffs hub lists multiple nonresidential schedules and riders. Santee Cooper likewise maintains a current rates hub for different retail service classes. Duke Energy Carolinas and Duke Energy Progress are separate legal utilities even though both use Duke branding. Cooperatives and municipal providers can use their own schedules.

Do not select a rate because it looks close to a proposal description or because a rate database associates it with the property’s ZIP code. The National Laboratory of the Rockies’ REopt guidance tells users to compare a database rate with the facility’s bill to confirm that the entry matches. REopt’s rate-input tutorial is a useful validation method, not authority to change or assign the customer’s tariff.

Record:

• exact schedule and rider names;
• utility and jurisdiction;
• effective date or revision date;
• source URL or tariff file;
• voltage, service class, contract demand, minimum demand, or other qualifiers when relevant;
• any known pending change that requires a new case; and
• who confirmed the account assignment.

The assessment can model a documented current rate or an explicitly labeled scenario. It should not tell the owner that another tariff applies without current utility confirmation. Rate switching, solar program eligibility, export treatment, and grandfathering are separate account-specific decisions.

Obtain interval data when available, then document what it means

Monthly bills describe billing periods. Interval data describe how a meter’s recorded value changes within a day. Depending on the account and meter, a provider or facility system may offer 15-minute, 30-minute, hourly, daily, or another resolution.

The National Laboratory of the Rockies’ REopt load-profile guidance distinguishes actual 15-minute, 30-minute, and hourly load data from a simulated building profile. It also explains why the calendar year matters: the correct day of the week affects how a time-varying rate is applied. That is a modeling lesson, not a universal requirement to reshape every raw utility file into REopt’s example format.

Preserve the original export. Create a separate working copy only after the following fields are understood:

Field	Question to answer
Meter and channel	Which physical meter and which quantity does this column represent?
Timestamp	Does it mark the start or end of the interval?
Time zone	Is it local standard time, local clock time, or UTC? How is daylight saving handled?
Duration	Does each value cover 15 minutes, 30 minutes, one hour, or another period?
Unit	Is the value kWh for the interval, average kW, instantaneous kW, kVA, kvar, or something else?
Direction	Is it import, export, net flow, generation, or a signed combination?
Multiplier	Has the meter multiplier already been applied?
Quality	Does the source identify actual, estimated, substituted, or missing values?
Calendar	What year is represented, and does it include leap day?

Do not convert interval kWh to kW, or combine channels, without documenting the formula and interval duration. Do not treat negative values as errors until the sign convention is known. They may indicate exports, a data convention, a correction, or a problem requiring investigation.

Green Button is one possible data format, not a universal South Carolina availability claim. DOE’s Green Button overview explains that customer data can be provided in a standardized machine-readable format and that the available time resolution depends on the utility’s data. Ask the named provider what the specific account can obtain, in what format, for what dates, and for which meters.

For one current provider-specific example, Dominion Energy South Carolina describes an interval-data service for qualifying interval-metered large-business customers, including CSV export and meter-profile aggregation. Dominion’s large-business page does not establish that every Dominion business account—or any other South Carolina utility account—has the same access.

Reconcile intervals to matching bill periods

Before using a time series, aggregate it to each bill’s actual start and end dates and compare the result with the billed energy for the same meter and channel. DOE’s energy-management information-system guidance identifies comparison of totalized interval data with utility-bill totals as a bill-validation capability.

This is not an instruction to force a match. It is a test designed to reveal what has not yet been explained.

Investigate differences such as:

• calendar-month totals compared with billing periods that begin midmonth;
• missing or duplicated intervals;
• an export file and bill using different time zones or daylight-saving conventions;
• kW treated as interval kWh, or the reverse;
• a meter multiplier applied twice or not applied;
• a replaced meter or account transfer inside the period;
• interval data for one meter compared with a consolidated account total;
• utility adjustments, estimated reads, rebills, or rounding;
• import-only, export-only, or net channels compared with a different bill quantity;
• generation or storage behind the utility meter; and
• a demand channel confused with an energy channel.

Keep a reconciliation table with the account, meter, bill dates, billed value, interval aggregate, difference, status, likely cause, evidence, and final treatment. If the reason remains unknown, preserve that fact as a model limitation. A clean spreadsheet created by deleting inconvenient rows is not stronger evidence than a messy file with a transparent exception log.

Keep demand data ready without promising a demand outcome

Commercial bills may show one or several demand quantities. The interval export may include average kW, kWh that can be converted for each interval, kVA, or no demand channel at all. A visible maximum in the file may not equal billed demand because the tariff may define a particular measurement interval, time window, ratchet, minimum, contract-demand rule, power-factor treatment, or billing adjustment.

For input readiness, retain:

• every demand field from each original bill;
• its label and unit;
• the corresponding account, meter, and bill period;
• the interval resolution and channel definition;
• the current tariff language governing demand; and
• any known meter or rate change.

Then hand those records to the demand analysis. Do not relabel the largest spreadsheet value as “the demand charge” or assume solar will reduce the billed quantity. The commercial demand-charge decision guide explains the interval and tariff validation needed for that separate conclusion.

Add a dated operating chronology

Electric records describe what the meter observed; they do not explain why it happened or whether it will continue. Build a month-by-month chronology with facilities, operations, finance, and property-management staff.

Include:

• normal operating days, hours, shifts, and seasonal schedules;
• planned and unplanned closures;
• vacancies, tenant starts or departures, and changes in leased area;
• storms, outages, equipment failures, or temporary generators;
• renovations, construction loads, and commissioning periods;
• production volume, refrigeration, HVAC, pumping, or process changes that materially affect electricity use;
• energy-efficiency work and control changes;
• service, meter, or tariff changes;
• new EV charging, electrification, storage, or major equipment;
• approved expansion or contraction plans; and
• the internal owner of every future-load forecast.

DOE’s interval-analytics guidance shows that daily and weekly profiles, heat maps, weather, occupancy, and operating variables can help explain commercial energy patterns. The goal here is not to diagnose every end use. It is to prevent an abnormal period from silently becoming the forecast.

Keep observed history separate from future cases. If a plant has approved a new line, retain the historical meter series as observed, document the line’s forecast and source separately, and label the modeled future case. Do not edit the historical series upward to make it resemble a plan. If a tenant departure is possible but unsigned, treat it as an uncertainty, not an assumed fact.

Existing solar or storage changes what utility data represent

When a property already has on-site generation, the utility record may show grid imports, exports, net grid use, or some combination. It may not show the building’s total electricity requirement. Storage can further alter the time pattern measured at the utility meter.

EPA’s ENERGY STAR research on commercial buildings and onsite renewable energy distinguishes renewable production, onsite renewable use, exports, grid imports, and net grid consumption. It warns that net consumption alone may not reveal how much electricity the building actually uses.

Collect, when available and relevant:

• the existing system’s nameplate and commissioning records;
• inverter or monitoring-platform production exports;
• utility import, export, or net channels and their definitions;
• a one-line or meter diagram showing where production and storage are measured;
• storage charge/discharge records and control changes;
• curtailment, outage, maintenance, or monitoring-gap history;
• meter replacement and configuration dates; and
• matching time zones, intervals, and units for every series intended to be combined.

Do not add monthly solar production to monthly net usage and call the result a complete load profile unless the meter relationships and energy flows support that calculation. Even when the monthly arithmetic is valid for one purpose, it does not create the subhourly facility profile needed for a different purpose. State what was measured, what was reconstructed, how it was reconstructed, and what remains unknown.

An existing system also creates a separate expansion, warranty, monitoring, and interconnection question. This data page should prepare the records; it should not assume the original and new equipment can be combined or that a tariff or authorization will remain unchanged.

Treat missing data as a decision, not a cleanup nuisance

A gap does not automatically stop an assessment, and a substitute does not automatically make the package complete. The consequence depends on the question being asked.

For every missing or suspect period, record:

• exact account, meter, channel, and dates;
• how the issue was found;
• whether the utility or facility was asked for a replacement;
• whether the gap affects a peak, seasonal period, tariff change, shutdown, or other material condition;
• any proposed substitution or synthesis method;
• who approved that method;
• which outputs may rely on it; and
• whether a new scenario or sensitivity is required.

Monthly bills may provide a useful consumption basis when interval data are unavailable, while a modeled reference-building profile may support a clearly labeled preliminary screen. NLR’s REopt materials expressly allow simulated profiles based on building type and annual or monthly consumption when actual interval data are absent. That does not make a generic building shape equivalent to this facility’s observed load.

Avoid these shortcuts:

• copying the prior weekday into a missing holiday without documentation;
• filling a summer gap with a winter average;
• rescaling a generic profile and calling it measured interval data;
• combining tenant and master-meter data without proving the meter hierarchy;
• deleting a meter exchange discontinuity;
• assuming a screenshot contains the same fields as the underlying export; or
• presenting an estimated read as actual because it appears on an official bill.

The analyst can decide whether a limitation is immaterial, requires a sensitivity, or prevents a particular conclusion. The owner should be able to see that decision and its basis.

Create a reproducible model handoff

The final input package should let another qualified reviewer reproduce what entered the model without reconstructing the history from email threads.

Use a controlled folder or document system with:

1. Read-me: site, decision being evaluated, current-through date, owner contact, and package scope.
2. Account-and-meter manifest: every account, meter, channel, rate, holder, and coverage period.
3. Original records: unaltered bills, utility exports, meter files, tariff documents, and relevant monitoring exports.
4. Working data: clearly named transformed files, never substituted for the raw sources.
5. Transformation log: formulas, unit conversions, time-zone treatments, aggregation, joins, and excluded records.
6. Reconciliation log: bill-to-interval comparisons and resolution status.
7. Operations chronology: observed disruptions, facility changes, and approved future cases.
8. Exception register: missing, estimated, inconsistent, or nonrepresentative data and their consequences.
9. Assumption register: every analyst assumption, source, owner, date, uncertainty, and output affected.
10. Model handoff note: model/version, intended question, input mapping, scenario names, and unresolved limitations.

Apply a current-through date to the package. A new tariff, meter, tenant, shift, service configuration, solar system, or major load can make an earlier handoff obsolete even when its files remain accurate records of the past.

The package should also say what it cannot support. A monthly bill set may be adequate for one preliminary energy screen but not for a conclusion that depends on subhourly timing. An interval file may show time shape but omit the correct billing account or current tariff. More rows do not compensate for missing identity.

Use a proceed-or-pause gate before modeling

Proceed to the defined assessment stage only when the data owner and analyst agree that:

• the project boundary and decision question are written;
• every relevant account and meter is either mapped or explicitly excluded;
• original bills and rate identifiers are retained;
• interval data have documented time, unit, meter, and channel definitions where available;
• interval and bill totals have been compared for matching periods;
• operating changes and future cases are dated and sourced;
• existing generation and storage are separated from grid records appropriately;
• gaps, estimates, and transformations are visible;
• the current tariff source has been checked without assuming eligibility or assignment; and
• the handoff states which conclusions the available inputs cannot support.

Pause when a missing account, unexplained meter hierarchy, unknown data unit, large unreconciled difference, existing-generation boundary, or nonrepresentative operating period could materially change the question being modeled. The next action may be a utility request, facilities interview, meter investigation, new data collection period, or a limited scenario—not a guessed answer.

Sunburst publicly describes commercial work that considers load profile, demand charges, tariff, and feasibility. If your team has assembled the manifest and source records, request a commercial solar assessment and ask Sunburst to confirm which records it accepts, the transfer method, the assessment scope, and how data limitations will be documented before you send files. That conversation defines the engagement; this page does not establish an upload workflow or promise a particular deliverable.

The model that follows remains a separate decision. For full cash-flow inputs and audit questions, use the commercial solar ROI guide. For the broader property, electrical, structural, interconnection, and approval evidence packet, use the commercial solar feasibility guide. Installed-cost scope belongs in the commercial solar cost guide.

How Sunburst handles the data stage

We treat the data package as a deliverable in its own right. A Sunburst commercial assessment returns your account-and-meter map, the rate schedule we verified (rather than assumed), the interval files with their measurement caveats stated, a dated operating chronology, and an explicit list of what is missing and how it limits conclusions. That record is yours whether or not you build with us, and it is what makes competing proposals comparable later.

Because we file interconnection applications in Dominion, Duke, Santee Cooper and cooperative territory regularly, we also know which data each provider will and will not release, and how long it takes. That saves weeks on portfolio work — see mapping multi-meter and multi-site portfolios if you are screening more than one building.

Next in the sequence: the feasibility gate, demand-charge mechanics and ROI modeling. Our commercial solar service covers all of it, in Charleston, Columbia and other South Carolina cities. Request a commercial assessment to begin.

Frequently asked questions

Are 12 months of utility bills enough for a commercial solar assessment?

They can provide one complete seasonal cycle and are a useful starting point, but the number alone does not establish readiness. Additional history may be important when operations, tenants, meters, rates, or billed-demand history changed. One year may also contain a shutdown or other abnormal condition. The assessment scope should state why the selected period is representative and disclose what is missing.

Do we need interval data if we have monthly bills?

It depends on the decision. Monthly bills can document billed energy, demand fields, account identity, rate, and cost components. Interval data add the time pattern of measured use and can be important when load alignment, time-varying rates, or demand behavior matters. If interval data are unavailable, label any simulated profile and limit the conclusions accordingly; do not present it as measured site data.

Is a screenshot of the utility portal acceptable?

It can support a specific observation, but it may omit source fields, time-zone information, meter identity, data quality flags, and exportable rows. Keep the screenshot if it adds context, then request the original bill or data export where available. Record exactly what the screenshot can and cannot establish.

Why do interval totals not match the bill?

Possible reasons include different date boundaries, units, time conventions, missing intervals, meter multipliers, estimated reads, account consolidation, a meter exchange, adjustments, or import/export/net-channel differences. Investigate and log the difference rather than forcing agreement. The acceptable treatment depends on the source and the model question.

What if the property has multiple meters or tenants?

Create a meter hierarchy before aggregating anything. Identify which utility meter supplies each building or tenant, which submeters sit behind another meter, who holds each account, and which loads belong to the proposed project boundary. This prevents both omissions and double counting. Utility and legal review may still be needed to determine how a future project can relate to those accounts.

What data are needed when the site already has solar?

Collect the existing system’s production records, utility import/export/net records, meter topology, monitoring coverage, and known gaps. Utility net use may not equal gross facility load. Any reconstruction must align the correct energy flows, meters, timestamps, and units and document the method.

Should the raw utility file be cleaned before it is shared with the analyst?

Preserve the raw file and create a separate working copy. Document every rename, conversion, timestamp treatment, aggregation, exclusion, and join. That makes corrections possible and prevents a polished spreadsheet from concealing how the source changed.

Does a complete data package guarantee an accurate solar model?

No. A controlled package improves traceability and exposes limitations, but model results also depend on the question, model choice, tariff interpretation, system assumptions, site conditions, and future operations. The analyst should identify which inputs and assumptions control each conclusion and should not present estimates as guaranteed outcomes.

Sources and methodology

• DOE FEMP: Evaluating Your Utility Rate Options
• DOE FEMP: Energy Management Information System Capabilities
• NLR REopt: Load Profile Input
• NLR REopt: Rate Inputs
• NLR SAM: Residential and Commercial Models
• DOE: Green Button
• EPA ENERGY STAR: Commercial Buildings and Onsite Renewable Energy
• Dominion Energy South Carolina: Rates and Tariffs
• Dominion Energy South Carolina: Large Business Services
• Santee Cooper: Current Rates