Electrical estimating, takeoff, and bidding field guide

What an electrical estimate is, and why labor decides it

An electrical estimate is a priced prediction of what it costs you to build the job, and it has four parts. The takeoff is the count: every device, fitting, length of conduit, and foot of wire on the plans. The material cost is that count priced from a supplier quote. The labor is that count converted to hours through labor units, the hours it takes to install each item. Overhead and profit go on top. Add them and you have the bid.

Material price you can get on the phone, and your competitors get nearly the same number from the same distributors. Labor is where bids separate. The hours are an estimate, the material is a quote, and the variable you can get badly wrong is the hours. Miss the count, miss the labor unit, or miss the conditions that slow the crew down, and the material being right to the penny will not save you.

So treat electrical estimating as a labor-hour problem with a material line attached. The count feeds both, but the count times the right labor unit is the number that wins the job at a price you can build, or loses it because you bid hours you cannot hit. Sizing the conductors and the service for that job is a separate calculation, handled in the load calculation and conductor ampacity guides; this one is about pricing the work, not sizing it.

What is a labor unit, and how does the labor-unit method work?

A labor unit is the number of labor hours it takes to install one item under normal conditions. A duplex receptacle carries a labor unit. So does a length of EMT, a wire termination, a fixture, a panel. Multiply the count of an item by its labor unit and you get the labor hours for that item. Sum every line and you have the total labor hours for the job. That is the labor-unit method, and it is the defining idea in electrical estimating.

The published source most of the trade leans on is the NECA Manual of Labor Units, the MLU, which has been the industry reference since the 1920s. It lists labor units for thousands of items, and it states what the unit includes: material handling, studying the drawings, layout and measuring, the install itself, and normal non-productive time. It does not include supervision, and it does not include your specific job conditions. The MLU is a starting point, not a verdict.

Labor units are expressed in hours per unit, often per hundred feet for raceway and wire and per each for devices and gear. The exact hours depend on the edition and the item, so pull them from the current MLU or your own database rather than memory, and treat any single number as the published baseline you then adjust.

The core arithmetic is one line repeated: count times labor unit equals labor hours. Eighty receptacles at the receptacle labor unit gives the receptacle hours. Twelve hundred feet of 3/4 in EMT at the conduit labor unit per hundred feet gives the conduit hours. Do that for every line and the total labor hours fall out of the bottom. Those total hours are the most important number in the estimate, because they drive the largest and least predictable cost, and they also set the crew size and the schedule, since hours divided by a crew and a workday give a duration.

The hours then feed the rest of the bid. They get multiplied by the burdened labor rate to become a labor dollar figure, the material gets added, and the whole thing rolls up in the recap. But it starts here, with the count and the labor unit, and everything downstream inherits whatever error you let into this step.

The takeoff: counting what the plans show

The takeoff is the quantity count, and it is the foundation the whole estimate sits on. You go through the drawings and count every receptacle, switch, fixture, panel, disconnect, and piece of gear, then measure the linear feet of conduit and the wire that fills it. Miss a count and you have not bid the work, no matter how good your labor units are. There is no labor unit for a device you never counted.

How you count has changed more than what you count. The old way was a set of colored highlighters and a roll wheel on paper prints, marking each symbol as you tallied it so nothing got counted twice or skipped. Digital takeoff replaced the highlighters with on-screen counting tools that drop a stamp on each symbol and keep a running tally, and measure conduit by tracing the routed path on the PDF instead of rolling it by hand.

Either way, the discipline is the same. Count by system and by area so you can check yourself, mark what you have counted so you do not double it, and reconcile the device count against the panel schedules and the fixture count against the lighting schedule. The wire takeoff comes off the conduit takeoff plus the homeruns, not off a guess. A clean takeoff is slower than people want and it is the cheapest insurance in the estimate.

Adjusting labor units for the conditions

Published labor units assume normal conditions, and almost no job is entirely normal. The MLU itself prints three columns for many items, a normal column and progressively harder ones, because the same device takes more hours to install fifteen feet up than at chest height, more in a congested ceiling than in open framing, and more in an occupied retrofit than in new construction. Adjusting the labor units for the real conditions is the difference between an estimator and a calculator.

The conditions that move the number are height, congestion, the building being occupied or existing, working off lifts or scaffold, off-hours work, weather exposure, and how clean the access is. Each one is a productivity hit. The trade handles them with a difficulty factor, a percentage you add to the base hours for a system or an area to reflect how much slower the crew will actually run.

Be honest in both directions. Padding every line for difficulty prices you out of the job; pricing every line at the normal column on a retrofit in a live hospital loses money on every hour. The skill is matching the factor to the real condition, area by area, and the only way to calibrate it is to compare estimated hours against the hours the crew actually burned on jobs like it. That feedback loop is the difference between guessing at the factor and knowing it.

The material takeoff and pricing

The material side is the same count, priced. Every item you counted for labor also has a material cost: the device, the box, the plate, the connector, the conduit, the wire, the gear. The count times the unit material price gives you the material dollars, and the count comes straight off the same takeoff that fed the labor.

Get the price from a current supplier quote, not from last year's number. Wire and steel move with the commodity markets, and copper in particular can swing enough between bid and buyout to erase a thin margin. For a real bid you send the takeoff to your distributors and get a project quote with a hold period, and you note when that price expires. Stale pricing is one of the quiet ways an estimate goes underwater.

Two material habits save jobs. First, account for waste and the small stuff, the straps, the wire nuts, the fittings, the box of screws, which never show on the plans but show up on the invoice. Second, separate the commodity material you buy in bulk from the quoted gear you buy once, because they price and lead differently. The big switchgear is not a line you mark up like a box of receptacles.

Estimating by assemblies

An assembly bundles all the pieces of a repeated install into one unit so you count it once instead of seven times. A standard receptacle assembly is the box, the device, the plate, the connectors, the ring, the wire whip, and the labor unit for each, rolled into a single line with a combined material cost and a combined labor unit. Count the receptacle, and the assembly pulls in everything that goes with it.

Assemblies speed the takeoff and cut the most common estimating error, the forgotten component. When the plate or the connector lives inside the assembly, you cannot leave it off, because counting the device counts the whole kit. On a job with hundreds of identical devices, that consistency is worth more than the time it saves.

Build your assemblies to match how your crews actually install, not a generic template. A receptacle in a metal-stud commercial wall is a different assembly than one in a poured wall or surface raceway, with different boxes, supports, and hours. Estimating software ships with assembly libraries you can adopt and then edit, and the edit is the point: the assembly is only as good as the material list and the labor units inside it.

Switchgear, panels, and long-lead equipment

The major equipment, switchgear, switchboards, panelboards, transformers, large disconnects, and motor control, gets handled differently from commodity material. You do not price it off a unit cost in the database. You send the one-line and the specs to the manufacturers or their reps and you get a quote for that exact assembly, because the gear is engineered to the job and the price depends on the configuration.

Two things make gear its own line in the estimate. The dollars are large enough that a single quote can swing the bid, so you want it from the source and you want it firm. And the lead time is long, often the longest single item on the schedule, which means it has to be released early or it drives the whole job late. Estimators flag long-lead gear at bid time so the project team orders it the day the contract is signed, not the day they need it.

Carry the gear quote with its lead time, its hold date, and what it includes, because gear quotes vary in scope. One vendor includes the integral surge device and the shipping splits, another does not, and the cheaper number is sometimes the one missing scope you then have to buy separately. Read what the quote covers before you let the low number into the bid.

What is the burdened labor rate?

The burdened labor rate is the true hourly cost of an electrician on the job, the base wage plus the labor burden. The burden is everything you pay on top of the wage: payroll taxes, workers compensation and liability insurance, unemployment, and benefits like health, pension, and vacation. Multiply your total labor hours by the burdened rate, not the bare wage, or you have underpriced every hour in the estimate.

The burden is a real percentage on top of the wage, and it is large enough that ignoring it sinks a bid. Industry references commonly put the all-in burden somewhere in the range of roughly 35 to 55 percent over base wage, but yours is yours: it depends on your workers comp experience modifier, your benefit package, your state, and union versus open shop. Calculate your actual burden from your own payroll and insurance numbers and update it yearly.

Blend the rate to the crew you will actually run. A job built by a foreman, two journeymen, and an apprentice has a different average burdened rate than one run entirely by journeymen. Estimate the crew mix, weight the rate to match, and the labor dollars track what you will really spend rather than a single rate that fits no actual crew.

Direct job expenses

Direct job expenses, the DJE line, are job costs that are not material and not installation labor but are spent on this specific job. They get their own line because folding them into material or overhead hides them, and hidden costs are unbilled costs.

The usual DJE list is permits and inspection fees, equipment rental like lifts and trenchers, temporary power and temporary lighting, the job trailer and its utilities, dumpsters and cleanup, consumables and small tools, fuel and travel, testing and commissioning by outside firms, and any bonds the job requires. On a prevailing-wage or out-of-town job, add per diem and travel time. None of these are in your labor units, and most are not in your material takeoff.

DJE is where careless estimates leak. The permit is obvious; the three months of scissor lift rental, the temporary power that runs the whole project, and the final torque-and-test report from a third party are the ones that get forgotten and then come straight out of profit. Build a standard DJE checklist and run every estimate against it, because the line you forget is the line you pay for yourself.

How do you add overhead and profit?

Overhead and profit go on after you have the job cost. Overhead is the cost of running the company that is not chargeable to any one job: the office, the estimators, the trucks and the shop, insurance, software, and the owner who is not turning a wrench. Profit is what is left for the company after every cost is paid. You recover overhead and earn profit by adding them to the cost as a markup.

Overhead is usually carried as a percentage of cost, set so that the year's jobs together cover the year's office expense. If your annual overhead runs around 15 percent of your cost of work, every estimate has to carry at least that much just to break even before a dollar of profit. Industry references commonly land overhead in the low-to-high teens as a percentage of revenue, but the only number that matters is the one your own books produce, so calculate it from your actual indirect costs and your actual volume.

Profit is a separate add on top of overhead, and it is a decision, not a leftover. Set it by how badly you want the work, how much risk the job carries, and how busy you are. A risky, fast, unfamiliar job should carry more profit than a repeat job for a known client. The mistake is treating profit as whatever happens to be left at the end, because what is left at the end of a job estimated without profit is usually nothing.

Markup versus margin

Markup and margin describe the same dollars from opposite ends, and confusing them quietly costs contractors money on every job. Markup is the percentage you add to your cost to get the price. Margin is that same profit as a percentage of the price you charge. They are not equal, and margin is always the smaller number.

The arithmetic catches people. A 25 percent markup on cost does not give you a 25 percent margin; it gives you a 20 percent margin. Mark up cost by 25 percent and the markup is a fifth of the final price, which is 20 percent of revenue. If you set prices by markup but talk about them as margin, your real profit is lower than you think on every bid, all year.

Pick one language and run the math in it. Most estimating software lets you enter the number either way; what it cannot do is know which one your target is. If the goal is to clear a 20 percent margin, mark the cost up by 25 percent, not 20. Get this backwards across a year of bids and the gap between the margin you planned and the margin you booked is the gap that surprises owners at tax time.

The recap: turning the estimate into a bid number

The recap is the summary page where every cost rolls up into the bid. It is the last thing you build and the first thing you check, because it is where the labor hours, the material, the direct job expenses, the sub quotes, and the overhead and profit all land in one place and turn into a single price.

The structure is consistent: total the labor hours and multiply by the burdened rate for labor dollars, add total material, add direct job expenses, add subcontractor and gear quotes, sum that to job cost, then apply overhead and profit to reach the bid. Lay it out so each component is visible on its own line, because a recap that hides the labor-to-material ratio hides the thing most likely to be wrong.

Sanity-check the recap before it goes out. Does the labor as a share of the total look right for this kind of work? Do the total hours match a crew and a duration that fit the schedule? Is every quote in the number current and in scope? The recap is the last place to catch an error cheaply. After this, the next chance to find it is on the job, where it costs real money. When the number is right, it becomes the quote and then the contract, and a CRM that carries it from the bid into the job record is what holds the field to the price you bid.

Lump sum, unit price, T and M, or cost-plus?

How you bid changes how the risk sits, and you pick the type to fit how well the scope is defined. Lump sum is one fixed price for a defined scope, and it puts the risk of the quantities on you; if your count is short, the shortfall is yours. It is the standard for hard-bid work off complete plans, and it rewards a clean takeoff.

Unit price bids a rate per unit, so much per receptacle, per foot of trench, per fixture, and the owner pays for the quantity actually installed. It fits work where the scope is known but the quantity is not, like site lighting or underground where the final count moves. The risk shifts to quantity verification, so the unit has to carry its full share of overhead and profit because there is no lump-sum cushion to hide a thin unit.

Time and materials bills the actual labor hours at an agreed rate plus material at an agreed markup, and it fits work that cannot be defined enough to price, like troubleshooting, demolition into the unknown, and most change work. Cost-plus is the negotiated cousin, your cost plus an agreed fee, common on fast-track jobs where the design is not finished. The rule across all four: the less defined the scope, the more the contract should pay you for what actually happens rather than what was drawn.

Reading the plans and the specs

You estimate from the whole document set, not the drawings alone. The plans tell you what and where; the specifications tell you the quality, the method, and the products, and the specs routinely drive cost the drawings never show. A note in Division 26 that calls for a specific manufacturer, a higher short-circuit rating, an arc-flash study, or hospital-grade devices can move the bid more than a dozen extra receptacles.

Read the spec before you trust the symbols. The plan shows a receptacle; the spec decides whether it is a standard device or a tamper-resistant, hospital-grade, or isolated-ground unit at several times the cost and a different labor unit. The plan shows conduit; the spec decides EMT, rigid, PVC-coated rigid, or MC cable, each with its own material price and install hours. Take off the quantity from the plan, price it to the spec.

Watch the general conditions and Division 1 too, because that is where the schedule, the working hours, the cleanup, the safety program, and the closeout requirements live, and those are DJE and labor-condition costs. The load and conductor sizing the spec drives belongs in the load calculation and conductor ampacity work; here, the point is that the spec, not just the one-line, sets the material and the method you are pricing.

Scope, exclusions, and clarifications

The scope statement is what protects the bid after you win it. It says, in writing, what the price includes and, just as important, what it does not. Without it, every gap in the drawings becomes an argument you are positioned to lose, because the owner heard one number and assumed it covered everything.

Write the exclusions plainly. Common ones on electrical bids are fire alarm, low-voltage and data, trenching and backfill, concrete and housekeeping pads, painting, patching, cutting of structure, temporary power if not in the documents, and anything in another division that the plans imply but do not assign. Add clarifications for the assumptions you priced on, the working hours you assumed, the gear manufacturer you carried, and the quantities you used where the documents were unclear.

Exclusions are not a way to hide a low number; a wall of exclusions on a clear scope reads as a contractor trying to weasel, and good owners discount it. The honest version names the real boundaries of an electrical scope and the genuine ambiguities in the documents. That list is the cheapest change-order protection you have, because a scope you wrote down at bid time is a scope you can point to when the question comes up six months in.

Addenda and bid day

Between issue and bid, the design changes, and those changes come as addenda. An addendum can add a floor of lighting, change the service size, swap the gear manufacturer, or answer an RFI in a way that moves your count. Acknowledge every addendum and fold each one into the takeoff, because a bid that misses the last addendum is a bid on the wrong job, and on hard-bid public work it can be thrown out.

Bid day is its own discipline. Sub and supplier quotes arrive late, sometimes minutes before the deadline, because everyone protects their number until the last moment. You plug the real quotes into the recap as they land, swap out the placeholders you carried, and recheck the bottom line each time, because a single gear quote coming in high can change the whole number late.

Have a process for the last hour and stick to it. Lock the takeoff early so only pricing changes at the end, keep a clean list of which quotes are in and which are still placeholders, and leave time to read the final recap once with a clear head before it goes out. The errors that hurt most are the bid-day arithmetic slips, the transposed quote, the sub left out, the addendum not added, made in the rush nobody planned for.

Pricing change orders and extras

A change order is an estimate too, just a small one against work already underway, and it follows the same rules: count the added work, apply labor units adjusted for the conditions, price the material, add the direct costs, and put overhead and profit on top. The piece people drop is the impact, the lost productivity when the change disrupts a flow that was already moving.

Price changes the same way you price the base job, with one addition. A change that stops the crew, sends them back into a finished area, or pushes work out of sequence costs more per unit than the same work in the original flow, and the labor unit should carry that hit. The receptacle added to a wall that is now closed and painted is not the receptacle labor unit from the open-rough estimate.

Most disruptive changes are billed time and materials, because they cannot be defined cleanly enough to lump-sum. Track the hours and material against the change as the work happens, not from memory afterward, and keep that record tied to the original bid. Good job records turn a disputed extra into a documented one, and keeping the bid, the change, and the actual hours in one place is what makes that record defensible instead of a he-said argument.

Estimating software and digital takeoff

Modern electrical estimating runs on dedicated software, the Accubid and McCormick families and their peers, and the reason is the labor-unit database. The software holds tens of thousands of items with material prices and labor units, lets you count on-screen, and rolls the count through the assemblies and the labor units into a recap automatically. It does the arithmetic that used to take days, so the estimator spends time on judgment instead of multiplication.

Digital takeoff is the front end of that. You load the PDF plans, count symbols with on-screen tools that keep a live tally, and trace conduit runs to measure footage, and every count flows straight into the estimate database instead of getting transcribed off a paper tally. The error of re-keying a hand count disappears, and the takeoff is faster and auditable.

What the software does not do is know your crews or your conditions. It ships with generic labor units and list-price material, and an estimate left on the defaults is an estimate built on someone else's productivity and yesterday's prices. The value is in tuning it: your labor units, your assemblies, current material pricing, and your burden and overhead. Once the bid is won, the estimate has to hand off to the people building it, so the quote, the contract, and the scope have to land in the field record and the job gets run against the number that was bid. FieldOS is built to carry that handoff.

Tuning labor units to your own crews

The best labor units in your estimate are not the published ones; they are the ones you proved on your own jobs. Published units are an industry average under normal conditions. Your crews, your tooling, your prefab shop, and your typical job conditions are not the average, and the only way to know your real productivity is to compare the hours you estimated against the hours you actually burned.

That feedback loop is the difference between an estimate that drifts and one that sharpens. After each job, the as-built labor hours by system get set next to the estimated hours, and the variance tells you which labor units were optimistic and which had room. Do that across enough jobs and you build a labor-unit database calibrated to your company, which is the single biggest edge a contractor can have over a competitor estimating off the book defaults.

This only works if the field actually captures the hours by task, and that is where job-cost feedback lives or dies. Time that is logged to the job in a system like FieldOS, broken down by area and system rather than dumped as a daily total, is what lets the estimator close the loop. An estimate without job-cost feedback is a guess that never learns; with it, every job makes the next bid more accurate.

Residential, commercial, and industrial estimates

The method is the same across sectors; the weighting and the detail change. Residential work is high-volume, repetitive, and often estimated by the device count and the square foot, with assemblies doing most of the heavy lifting because the installs repeat so consistently. The risk is in volume and speed, and a small labor-unit error multiplies across hundreds of identical devices.

Commercial work is the heart of plan-and-spec estimating: a defined drawing set, a real specification, gear and feeders to quote, and conditions that swing with height and occupancy. This is where the full takeoff, the assemblies, the gear quotes, and the condition factors all carry real weight, and where most hard-bid competition lives.

Industrial work, plants, process, and heavy power, pushes the detail further. Conduit is rigid and the runs are large, the terminations and equipment connections carry heavy labor, instrumentation and controls add scope, and the conditions are often the hardest the labor units have to account for. The labor unit per termination on a large feeder in a plant is a serious number, and the estimate lives or dies on the connections and the gear, not the device count that drives a commercial fit-out.

The data center and large-project estimate

Large projects, data centers above all, change the estimate less in method than in scale and in where the cost concentrates. The device-and-fixture count that dominates a commercial fit-out is a small fraction here. The money is in the power distribution: the switchgear, the large feeders, the busway, the UPS and battery systems, the generators and paralleling gear, and the terminations that connect them all.

Two things dominate a data center estimate. Gear quotes and lead times drive both the price and the schedule, and the long-lead equipment, the switchgear and the gensets, can carry lead times measured in many months that have to be released the day the contract signs. And the labor concentrates in large-conductor terminations, cable pulls, and the redundant power paths that the reliability tier demands, so the labor unit per termination and per foot of large feeder is where the hours pile up.

Estimate these with the redundancy in the count. A facility built for concurrent maintainability carries duplicate paths and gear, so the takeoff counts the redundant runs as real work, not as a drawing artifact. The feeder sizing and the heat that drives conductor selection on these long, heavy runs is load-calculation and ampacity work; for the bid, the point is that the count is small but each counted item carries large material and large hours, so an error on a feeder or a gear line moves the number hard.

What to document in the estimate

An estimate you cannot reconstruct is an estimate you cannot defend, learn from, or hand off. The bid number is the output; the record is what makes it usable after the job is won. Six months in, when the field is over hours or a change is disputed, the estimate file is the only thing that says what was and was not priced, and at what assumption.

Keep the takeoff quantities by system and area, the labor units and any condition factors applied, the material pricing with its source and hold date, the gear and sub quotes with their scope and lead times, the direct job expense list, the burdened rate and crew mix, the overhead and profit applied, and the written scope and exclusions. When the job closes, add the as-built hours next to the estimated hours so the next bid inherits what this one taught you.

Common mistakes

• Wrong or missing labor units, or leaving the published units on the defaults instead of your own data.
• Not adjusting labor for height, congestion, occupancy, or retrofit conditions.
• A short or skipped count, where work never made it into the takeoff at all.
• Pricing material off stale numbers instead of a current supplier quote with a hold date.
• Forgetting direct job expenses, permits, rental, temp power, testing, and travel.
• Pricing labor at the bare wage instead of the burdened rate.
• Carrying no overhead or no profit, or confusing markup with margin so the booked margin comes in low.
• Sending a bid with no written scope and exclusions to protect it.
• Bidding off incomplete plans, a missed addendum, or without seeing the site conditions.
• Plugging gear and sub quotes into the recap without reading what they include.

Field and bid checklist

Standards and references

The labor-unit reference most of the trade uses is the NECA Manual of Labor Units, published by the National Electrical Contractors Association and revised on a cycle, with multiple columns for normal and harder conditions. Treat its numbers as a calibrated starting point, not a fixed truth, and tune them to your own job-cost history. Cost-data references such as the RSMeans electrical data give another published baseline for material and labor, again to be adjusted to your market.

The estimating method itself, takeoff, labor units, material, direct job expenses, the recap, and overhead and profit, is standard practice across electrical contracting and is taught through NECA and contractor associations rather than codified in one document. The software families, Accubid and McCormick among them, encode that method in their workflows and ship labor-unit and assembly libraries built on the published references.

Treat every cost, labor unit, burden percentage, overhead rate, and markup in this guide as method, not as a quoted figure to copy. The real numbers come from your own payroll, insurance, and job-cost records, from current supplier and gear quotes, and from the market you bid into. Material prices move with the commodity markets, labor units vary with the edition and your crews, and burden and overhead are specific to your company. The product designs and pricing benchmarks the trade publishes change, so verify against the current editions and your own books before you commit a number to a bid.

Units and terms

Estimating carries its own vocabulary, and the same idea shows up under several names across a bid set, a labor manual, and a piece of software.

Labor units are hours per unit, often hours per hundred feet for raceway and wire and hours per each for devices and gear. The takeoff is the quantity count. The recap is the summary that totals the bid. DJE is direct job expenses. The burdened rate is the wage plus burden. O and P is overhead and profit. Markup is a percentage of cost; margin is a percentage of price, and they are not equal.

Takeoff

The quantity count of every device, fitting, fixture, and length of conduit and wire from the plans

Labor unit

The labor hours to install one item under normal conditions, from the NECA MLU or your own data

Assembly

A bundle of the material and labor units for a repeated install, counted as one line

Burdened rate

The base wage plus labor burden (taxes, insurance, benefits), the true cost of an hour

DJE

Direct job expenses: permits, rental, temp power, consumables, testing, and travel for the job

Recap

The summary that rolls labor, material, DJE, and quotes into job cost, then adds overhead and profit

Markup vs margin

Markup is profit as a percent of cost; margin is profit as a percent of price; margin is always smaller

Related tools

Calculators and readiness checks for this work

Compare your options

FAQ

People also ask

• What is an electrical load calculation?
• What is conductor ampacity?
• How do you estimate a landscaping job?
• How do you estimate a plumbing job?
• How do you estimate an HVAC job?