Construction Estimate Template: What Every Column Is For

What does a construction estimate template actually need to contain?

Six sheets. All of them can be built in a common spreadsheet program without add-ins — the value is in the structure, not in the file. If a template you downloaded is missing one of these, you will end up adding it by hand on the first real job.

Cover and division summary. Project name, project location and market, drawing set date and revision number, estimator, estimate date, the level of design the estimate was priced from (conceptual, schematic, design development, or construction documents) with the accuracy range that carries, the basis of the schedule assumed, and a roll-up of cost by CSI division. Stating the estimate class on the cover is what stops a budget number being quoted back to you as a bid. Write it so it stands on its own, because it may be the only page a reviewer reads closely. It should pull its numbers from the detail sheets by formula, never by retyping.

Detail sheets. One row per priced item, grouped by division. This is where quantity, unit, rates, and extensions live. Whether you use one long sheet with a division column or a separate tab per division is a preference; one sheet filtered by division is easier to audit and easier to sort.

Rate table. Every unit cost, wage rate, burden percentage, and equipment rate in one place, referenced by the detail sheet. This is what lets you reprice a whole job for a different market or a different year without touching a single takeoff quantity.

General conditions sheet. Division 01 costs priced as their own line items — supervision, temporary power, dumpsters, sanitary, site cleanup, small tools, safety. Most of these are driven by schedule duration, not by scope quantity, so they do not belong as a percentage on the summary page.

Bid summary. The markup stack: subtotal of direct cost, general conditions, escalation to the construction midpoint, contingency, overhead, profit, bond, uncovered insurance, and tax. Each as its own visible line with its own basis — and each stated once. Workers' compensation and general liability that are already inside the burdened crew rate must not reappear here; the summary insurance line covers only what the burden does not, typically builder's risk, umbrella, subguard or project-specific policies. Note on the sheet which policies sit in burden and which sit in the stack.

Assumptions, exclusions, and clarifications. The page a scope dispute gets argued from. It travels with the estimate to the client, every time.

Why organize the estimate by CSI division?

Because on North American commercial work everyone else already is. Specification books are written in MasterFormat order, subcontractors bid in their own division, and architects and owners review by trade section. An estimate organized any other way forces a translation step every time it meets another document, and translation steps are where scope gets dropped.

The divisions you will use on most vertical work are 01 General Requirements, 02 Existing Conditions, 03 Concrete, 04 Masonry, 05 Metals, 06 Wood, Plastics and Composites, 07 Thermal and Moisture Protection, 08 Openings, 09 Finishes, 10 Specialties, 11 Equipment, 12 Furnishings, 13 Special Construction, 14 Conveying Equipment, 21 Fire Suppression, 22 Plumbing, 23 HVAC, 26 Electrical, 27 Communications, 28 Electronic Safety and Security, 31 Earthwork, 32 Exterior Improvements, and 33 Utilities. Division 14 in particular is not optional on anything with an elevator, and conveying equipment carries a procurement lead that has to be confirmed with the supplier early. Keep all of them visible on the summary even when they carry zero cost. A division showing a deliberate zero reads differently from a division that is simply absent, both to a reviewer and to you three weeks later.

The second reason is comparison. Once several closed-out jobs are coded the same way, your own division-level history becomes the sanity check on every new estimate — and for the work you repeat, that history is the more reliable input.

What does each column in the detail sheet do?

Left to right, this is the anatomy of a single priced row. Each column exists because something breaks when it is missing.

Why split labor, material, and equipment instead of using one installed price?

A single "installed cost per SF" is faster to type and nearly useless afterward. The split costs you one row of setup and pays for itself on the first change.

Burden only rides on labor. Payroll taxes and workers' compensation are rated on labor dollars, and liability is often allocated to labor as well — check how your own carrier rates each. If labor is fused into a blended rate, you cannot apply burden correctly and you cannot show anyone how you did.

Tax generally rides on material. Sales and use tax treatment varies by state and by contract type, and the split is rarely uniform — but it is not always material-only. Several states assess a gross receipts or transaction privilege tax on the full contract amount, labor included, and some treat the contractor as the taxable consumer of materials instead. Confirm the treatment for the project's jurisdiction and contract form before you set the basis, and state on the bid summary which one you carried. A blended rate forces you to guess at the split later.

Markup is not uniform. If your company carries a different markup on self-performed work than on subcontracted work — or a different one again on material-only supply — that policy cannot be applied to a number that has already been blended.

Change orders are argued in hours. When a change disrupts the sequence, the claim is about productivity — hours per unit, not dollars per SF. If your base estimate never recorded hours, you have no baseline to demonstrate the impact against.

The schedule comes out of the labor column. Total man-hours divided by realistic crew size gives an activity duration and a crew-loading check — the input a scheduler needs, not a schedule. Real durations still have to absorb work-area availability, trade sequencing, cure and inspection holds, mobilization, and learning curve, and the critical path frequently runs through procurement and approvals rather than installation hours. Use the calculation to test whether a proposed duration is physically achievable with the crews you can field. A blended estimate produces a price and no schedule information at all.

The three escalate differently. Material prices move with commodity markets and supplier quotes; labor moves with local wage agreements and availability; equipment moves with rental markets and fuel. Blended, you cannot tell which one moved when the job goes sideways.

How should the waste factor be handled?

Keep it in its own column, applied to material, and leave the neat quantity untouched. The moment waste is folded into the takeoff number or into the unit rate, two things become impossible: verifying the takeoff against an independent measurement, and adjusting waste when the conditions change.

Waste is a function of the material and the geometry, not a house constant. Sheet goods cut into a plan full of angles waste more than the same sheet goods in a rectangular warehouse. Small pours, long narrow slabs, and pump runs waste concrete differently than a large flat placement. Rolled and continuous materials waste less in offcuts, but laps are a separate addition and must be calculated rather than absorbed into a waste percentage. Membrane roofing and underlayment carry side and end laps set by the manufacturer's requirements; reinforcing steel carries lap splices, hooks, and bends set by the drawings and the code, and in congested elements the splice allowance is a substantial fraction of the tonnage. Compute laps and splices into the neat quantity from the detail, then apply a waste factor on top for offcuts and damage. Lengths ordered in fixed stock sizes waste according to how badly the design lengths divide into stock lengths, which is a real calculation, not a percentage guess.

One point that is often missed: for manufactured and cut-to-fit materials, waste affects the ordered quantity while labor is driven by the in-place quantity. That rule does not extend to bulk earthwork. Bank, loose, and compacted measure are three different quantities of the same soil, converted by swell and shrink factors specific to the material, and the conversion drives haul volume, truck counts, and placement hours — not just what you buy. Carry those conversions as their own columns with the measure stated, never as a waste percentage, and never assume the factor without a geotech report or your own history on that material. Handling and cutting the waste does consume labor, but that belongs in the productivity rate, not in a second application of the waste factor to the labor column. Applying waste to both sides is a quiet way to inflate an estimate without anyone being able to see where.

What actually belongs inside a unit rate?

Define it once, on the rate table, and write the definition down on the sheet. A unit rate is a container, and estimates go wrong when two people fill it differently — one estimator's "$/SF installed" includes equipment and the other's does not, and the two rows are silently incomparable.

The more durable construction is to store productivity and price separately: labor hours per unit in one column, fully burdened crew cost per hour in the rate table. Productivity is a property of the work and the crew and is reasonably stable; the wage rate is a property of the market and the year and changes constantly. Keeping them apart means a new market or a new wage agreement is a one-cell edit, not a re-estimate.

Where the rates themselves come from is the harder question, and the honest answer is that published cost data is a starting point and not an answer. It is built on regional averages and assumed conditions that may not match your crews, your access, your shift, or your supply chain. Use it to fill gaps and to sanity-check, then replace each line with your own closed-out cost as soon as you have one. An estimating workbook whose rate table is still entirely purchased data after several completed jobs is a workbook that is not being fed.

Where do general conditions, markup, and bond go?

General conditions belong on their own priced sheet, built from the schedule. Supervision, project management, temporary power and water, trailer, sanitary facilities, and safety are time-dependent — they cost what they cost per month, times the number of months. Others in Division 01 are quantity-driven and should be priced that way: dumpsters by the number of pulls, which tracks demolition volume and packaging waste rather than calendar time; final cleanup by square foot of finished area; temporary protection by the area actually protected. Price each general conditions line on the basis that actually drives it, and show the basis in its own column. Carrying them as a flat percentage of direct cost means a job that runs long carries the same general conditions as a job that finishes early, which is exactly backwards.

After that comes the markup stack, and the order of operations is a policy decision your workbook should state explicitly rather than leave implied in a formula. Contingency, overhead, profit, bond, insurance, and tax can each be applied to different bases, and applying them in a different sequence produces a different bid price from the same direct cost. Bond premium is normally calculated on the final contract amount, so it is applied after the other markups, not alongside them — bonding an un-marked-up number under-recovers the premium unless you correct for it elsewhere. Take the rate from your surety's current rate schedule rather than a house percentage: premiums run on a sliding scale per thousand of contract value that steps down as the contract grows, and the rate varies with your surety credit, the contract type, and whether performance and payment bonds are quoted together. Because the premium is itself part of the contract amount, gross the calculation up if the contract sum is large enough for the difference to matter.

Contingency deserves a line of its own and a written basis. It is coverage for the parts of the scope the documents do not yet resolve, and it should shrink as the design does. A contingency that stays at the same percentage from schematic through issued-for-construction is not a contingency, it is a hidden margin — which is fine as a business decision but dangerous as an unlabeled one, because it disappears the first time someone asks you to sharpen the number.

What belongs on the assumptions and exclusions sheet?

This page takes little time to write and can settle arguments that would otherwise drag on. It should go out with the estimate every time, not on request.

• The drawing set: sheet list, issue date, and revision number the takeoff was performed against
• Specification sections used, and any section not yet issued
• Scope explicitly excluded — permits and permit fees, testing and special inspection, hazardous material abatement, utility company charges and impact fees, off-hours and premium-time work, winter conditions and temporary heat, rock excavation, unsuitable or unstable soils and any import required, dewatering, differing or unforeseen subsurface conditions, escalation beyond the stated quote validity dates, liquidated damages, and the builder's risk deductible. On anything with site work, the subsurface exclusions are the ones that get argued; an estimate silent on rock and dewatering will be read as having carried them.
• Allowances carried, with the basis for each amount
• Alternates priced and whether they are included in or excluded from the base number
• Quantities taken from schedules or counts versus quantities assumed where the documents are incomplete
• Quote validity dates for any priced material subject to market movement
• Assumed working hours, access, laydown, and site availability
• Assumed schedule duration, since the general conditions depend on it
• Sales tax treatment and who is responsible for it

How do you keep the workbook from breaking?

Spreadsheet estimates fail quietly. These habits are what separate a workbook that survives ten jobs from one that gets rebuilt every time.

• No typed numbers inside formulas — every rate comes from the rate table by reference
• One rate table, one wage and burden definition, one place to change a price
• Protect calculated columns so no one types over an extension; an overwritten formula leaves no trace and does not surface until the total is already wrong
• Use structured tables or named ranges instead of raw cell addresses that break when a row is inserted
• Keep units in their own column and never mix units within a single item type
• Add a check column: cost per square foot of building area, or hours per unit, compared against your history — any row that reads wrong by an order of magnitude shows up here
• Total the sheet in two directions and confirm they agree
• Keep a locked copy of the estimate as issued, and do revisions in a new file with a version number
• Never reuse last job's file without clearing quantities first; leftover rows from a previous project are a recurring and expensive failure

Where does a spreadsheet stop being the right tool?

A workbook is excellent at arithmetic and at nothing else. It does not link a quantity back to the marked-up drawing it came from, so verification means re-measuring rather than clicking a number and seeing the highlighted area. It does not tell you when a drawing revision invalidates a row. Co-authoring exists, but it is fragile on a large linked workbook, so two estimators on one job usually still means dividing the sheets and reconciling by hand. And it will return a wrong answer with total confidence when a formula gets clipped, because nothing in a spreadsheet knows what a reasonable number looks like.

That is the case for takeoff software with a linked digitizer and, at larger volumes, for a real estimating database. It is not, by itself, a case for outsourcing. If you build the same few building types, in one market, with your own crews, your historical unit costs for that work are likely to be more reliable than an outside firm's, and a workbook fed by your own closed-out jobs is a strong tool. In that situation outsourcing the estimate usually gains you little.

Outsourced takeoff and estimating earns its cost in narrower situations: bid volume that exceeds the hours you have, a scope or building type you have never priced, a market you are entering, or a period where the alternative is not bidding at all. Even then, what comes back is quantities and rates in a structure like the one above — and the rate table still needs your numbers in it before the total means anything about your company. A takeoff can be done by someone else. A cost is yours.