Preconstruction planning and services field guide

What preconstruction is, and why the early decisions decide the job

Preconstruction is the planning phase that runs before any crew breaks ground. The team prices the design as it develops, checks whether what is drawn can actually be built, looks for savings that do not cost the owner value, plans the schedule and the site, and locks a budget, often a guaranteed maximum price, before construction starts. The work happens on paper, where changing your mind is cheap.

The reason it carries so much weight is timing. Most of what a building will cost is committed by the choices made in design, long before the money is spent. A structural system, a mechanical approach, a floor-to-floor height, the routing of the main electrical distribution: each of those decisions sets a cost that the field then spends a year executing. Move them while they are still lines on a drawing and the change costs a markup and an hour of someone's time. Move them after the gear is bought and the slab is poured and the change costs a change order, a schedule hit, and the trust of the owner.

This guide covers the planning side. Pricing the work in detail, the takeoff and labor units that turn a count into a number, lives in the estimating guide. Building and running the schedule the field works from lives in the scheduling guide. Preconstruction is where both of those disciplines get set up before the job is real.

Why early influence is worth more than field savings

Early in design, the ability to influence cost is high and the cost of a change is low. As the design firms up and construction starts, that flips: influence drops toward zero while the cost of a change climbs. The curve is the whole argument for preconstruction. You spend a modest fee to put a builder in the room while the decisions are still cheap to change.

A worked example makes it concrete. Reroute a main feeder on a schematic drawing and the cost is a redrawn sheet. Reroute the same feeder after the switchgear is released and the conduit is in the slab and you are paying for demolition, new gear, a delay while the long-lead replacement ships, and the ripple into every trade that was sequenced behind it. Same decision. Two orders of magnitude difference in what it costs.

This is also why preconstruction wins negotiated work. An owner who has been burned by change orders on a hard-bid job will pay for a team that finds the problem on paper. The fee is small against a single avoided change on the gear or the structure. The savings are real, but they are invisible, because they are the changes that never happened.

What preconstruction services actually cover

Preconstruction is a set of services, not a single deliverable. The core list is consistent across firms even when the names differ.

Budgeting and estimating at each design milestone, so the price tracks the drawings as they develop. Constructability review, reading the design for whether it can be built and sequenced. Value engineering, finding equal-function alternatives that cost less. Scheduling and phasing, the preliminary timeline and the logistics that go with it. Long-lead identification and procurement planning, catching the equipment that drives the schedule. Risk identification and the contingency that prices the unknowns. Buyout planning, the strategy for breaking the work into bid packages and awarding the subs. And the budget lock at the end, often a guaranteed maximum price with its assumptions written down.

Which of these the contractor delivers, and how formally, depends on the contract and the delivery method. On a large CM-at-risk job the precon team is a staffed department with a fee schedule. On a smaller negotiated job it might be the estimator and the project manager doing the same thinking with less paperwork. The thinking is the constant. The deliverables scale to the project.

Which delivery method preconstruction fits

Preconstruction depends on the builder joining the team early, so it fits the delivery methods that bring the contractor in during design. On construction-manager-at-risk, the CM is hired in design as the owner's advisor, prices the work as it develops, and converts to a guaranteed maximum price as the documents firm up. On design-build, the designer and builder sit under one contract, so the pricing and constructability feedback happen inside the team from the start. On a negotiated job, the owner picks the contractor on relationship and qualifications rather than low bid, which buys the same early involvement.

The method that does not get preconstruction is traditional design-bid-build, the hard bid. There the design is finished first, then put out to bid, and the contractor shows up after every decision is made. Whatever a builder would have caught on paper is now the owner's problem to fix in the field, through change orders.

None of this makes hard bid wrong. It is the right tool when the scope is well defined, the owner wants the lowest price on a complete design, and the project does not carry the kind of risk that early input would have priced down. Match the method to the project, then expect the preconstruction depth that method allows.

Design-stage estimates: pricing a moving target

The estimate is not a single number produced once. It is a series of estimates, each more precise than the last, tracking the design through every stage, concept through complete documents. Early on you are pricing intent with very little drawn, so the estimate is a wide range built on cost models, area, and history. As the design develops, the count gets real, the systems get specified, and the range tightens.

AACE International's estimate classification puts numbers on this. A Class 5 estimate, made at the conceptual stage with the design barely started, can run on the order of minus 30 to plus 50 percent or wider. A Class 3, a budgetary estimate around design development, narrows considerably and is commonly used to set the control budget. A Class 1, built on essentially complete documents, is the tightest, often quoted near plus or minus 10 percent. The exact ranges depend on the project type and the method, so treat the class as a statement of how defined the design was, not a guarantee of accuracy.

The practical sequence on a building job tracks the design phases: a conceptual estimate, a schematic-design estimate, a design-development estimate, and a construction-documents estimate. Each one is the moment to reconcile against the budget and tell the owner the truth about where the number sits. Skip a milestone and the next surprise is bigger.

Budget reconciliation as the design develops

A new estimate is only useful if you compare it to the last one and to the owner's budget. Reconciliation is that comparison: line up the current estimate against the control budget and the prior milestone, then explain every movement. Scope was added here. A system got more expensive there. An allowance was carried for something still undefined. The drift is the story.

The failure is estimating in a vacuum. A team that produces a design-development number without reconciling it to the schematic number and the budget hands the owner a figure with no context, and nobody can tell whether the project is on track or quietly running away. By the time the construction-documents estimate lands, the gap is a crisis instead of a series of decisions.

Track the budget as a living document. Carry the original program budget, the current estimate, the variance, and the reason for it, milestone over milestone. When the number drifts up, that is the moment to value-engineer, cut scope, or get the owner to add funds, while there is still design left to change. Catch it late and the only lever left is cutting quality in the field.

The constructability review

Constructability review is reading the design as a builder, not as a checker of code. The question is not whether the drawing is correct. It is whether the thing drawn can be built, in what order, with what access, by crews who have to physically get to the work. You catch the conflicts on paper, where the fix is a revised detail, instead of in the field, where the fix is a crew standing around while two trades argue over the same six inches of ceiling.

The common findings are mundane and expensive. A pipe and a duct and a cable tray all routed through a corridor that does not have room for all three. A piece of equipment specified larger than the door it has to pass through. A pour sequence that traps a crew with no way out. A main switchboard located where there is no path to bring it in once the walls are up. None of these are design errors in the narrow sense. Every line is correct on its own sheet. They collide because no one read the sheets together as a sequence of real work.

A real constructability review names the conflict, the trades involved, and the means-and-methods question behind it, then routes it back to the design team while the answer is still cheap. The reviewer is the person who has built it before and knows where the drawing and the jobsite part ways.

Value engineering without cutting value

Value engineering is finding a different way to deliver the same function for less money. The Society of American Value Engineers frames value as function divided by cost, so you raise value by holding the function and lowering the cost, not by stripping the function out. Done right, the owner gets the same building for less. Done wrong, it is just cost-cutting wearing a respectable name, and the owner pays for it later in performance, maintenance, or a space that does not do what they needed.

The honest version starts with function. What does this assembly have to do? Then it asks whether a different system, material, or method does the same job for less. A switchgear configuration that fits the same load for less cost. A lighting layout that hits the same footcandles with fewer fixtures. A slab approach that pours faster. Each idea gets priced, its trade-offs named, and the owner decides. That last part is not optional. Value engineering proposes; the owner disposes. The team that quietly swaps a spec to save itself money has not value-engineered anything. It has substituted, and on most contracts that is a problem.

Bring the ideas early, while the design can still absorb them, and log every one: what was proposed, the saving, the trade-off, and what the owner decided. The log is what keeps a VE program honest and keeps the savings from turning into a fight later.

MEP coordination and clash detection

The systems that fight for space are mechanical, electrical, and plumbing, and the fight is worst above the ceiling and in the shafts. Early coordination, increasingly run in a shared model with clash detection, finds where a duct, a pipe, a conduit run, and a sprinkler main all want the same room before any of them is installed. Resolve it in the model and you spare the field the worst kind of rework: tearing out installed work to make room for the trade that came behind it.

This is a lighter touch in preconstruction than it will be once the trades model their shop drawings, but the major routing and the tight spots are worth catching while the design is still moving. The clashes that survive into the field are the ones nobody looked for early.

Long-lead items drive the schedule, so find them first

Some equipment takes so long to arrive that the order date, not the install date, controls the schedule. On electrical work this is the gear: switchgear, switchboards, transformers, generators, and large distribution equipment. As of the 2026 review those lead times are long and volatile. Standard switchgear assemblies have run on the order of 30 to 50 weeks, medium-voltage equipment longer, and large power transformers have stretched past two years in constrained markets. A building cannot be energized until that equipment shows up, so a missed long-lead item can hold the whole project no matter how well the rest is sequenced.

Identify the long-lead items in preconstruction, not after the design is done. Run a long-lead workshop with the owner, the design team, the major trades, and the vendors, and build the real lead times into the schedule. Where it makes sense, pre-purchase the critical equipment early, sometimes under a separate procurement contract, so the clock starts before the rest of the documents are finished.

The long-lead items belong on the preliminary schedule as the controlling dates they are, which is the scheduling discipline this guide does not repeat. The point here is to catch them while there is still time to act. Discover a 40-week piece of gear two months before you needed it installed and there is no recovery that does not cost money or time.

The preliminary schedule and phasing

Preconstruction produces the first real schedule: the milestones, the major sequence, the phasing, and the durations the budget and the logistics plan get built around. It is not the detailed critical-path schedule the field will run from once the job starts. It is the framework that proves the project fits in the time the owner has, flags the long poles, and sets the phasing if the building has to be occupied or turned over in stages.

Phasing is where the early schedule earns its keep. If the owner needs one wing open while another is still under construction, that constraint reshapes the sequence, the temporary services, and the cost, and it is far cheaper to plan for than to discover. The same goes for any milestone with money attached, a tenant move-in or a liquidated-damages date.

The mechanics of the critical path, float, the look-ahead, and how you prove a delay belong in the scheduling guide. In preconstruction the schedule is a planning and feasibility tool: does this design build in this time, in this order, and what has to be ordered or decided first to make that true.

The site logistics plan

The logistics plan is how the site works as a place to build, drawn before anyone is on it. Where do trucks come in and turn around. Where does material lay down without blocking the work. Where does the crane sit and what is its reach. How do temporary power, water, and sanitary get in. How does the site change as the building goes vertical and the laydown you had at grade disappears under the structure.

On a tight urban site this is most of the planning, because there is no room for error. A crane with no pick radius to the far corner, a delivery route that closes the only gate, a laydown area that sits exactly where the loading dock pour goes: each is a problem that costs real money once the trailers are set and the fence is up. Draw the site through its phases, because the logistics that work for the foundation rarely work for the topping-out, and the plan has to move with the building.

For electrical work the temporary-power plan is part of this. Sizing the temporary service, locating the spider boxes, and planning the path from temporary to permanent power belongs in the logistics conversation, not as an afterthought once the trades show up needing power.

The risk register and the contingency that prices it

Every project carries unknowns, and the question is whether you price them on purpose or get surprised by them. A risk register lists them: the soil that might not be what the borings said, the long-lead gear that might slip, the design that is not finished, the permit that might drag, the owner decision that is still open. Each risk gets an owner, a likelihood, an impact, and a plan, and the ones that carry real cost exposure get money set against them in the contingency.

Contingency is the dollar reserve for the unknowns inside the defined scope, not a slush fund for scope the owner adds later. That distinction matters, and it is worth settling in writing: who can draw on the contingency, for what, and up to what amount. On a guaranteed maximum price job the contingency lives inside the number, and how it is controlled and what happens to the unused balance is a negotiation, not a default.

The amount of contingency is not a fixed percentage. It tracks how defined the design is and how much risk the project carries. A well-defined renovation with known conditions needs less than a fast-tracked job priced off half-finished drawings. Price the unknowns you can name, carry contingency for the ones you cannot, and write down what the number assumes.

The guaranteed maximum price

A guaranteed maximum price is the contractor's commitment to deliver the project for no more than a stated number, with the owner paying the actual cost of the work plus a fee up to that cap and any savings below it usually shared or returned per the contract. The thing that makes a GMP hard, and the thing crews get wrong, is that it is usually set before the design is complete. The contractor is guaranteeing a price for work that is not fully drawn.

That gap is bridged with assumptions. To price an undrawn portion, the contractor has to assume how it will be designed and built, and a GMP of any size carries a long list of those assumptions, plus clarifications, exclusions, and allowances for the scope still undefined. The number is only as good as the document that explains what it includes and what it does not.

A GMP fits CM-at-risk and design-build, where the builder has been in the design and can price the risk it is taking. The cap protects the owner; the assumptions and the contingency protect the contractor against the design it has not seen. When the documents finish and the buyout comes in, the actual cost is reconciled against the GMP, savings are settled per the contract, and the assumptions are what the two sides argue from if the scope turns out different than the number assumed.

The assumptions, clarifications, and exclusions behind the number

The number on the cover page is meaningless without the pages behind it that say what it includes. Assumptions state what the contractor took as given where the design was silent. Clarifications resolve what a drawing left ambiguous. Exclusions name what the price does not cover, so nobody assumes it is in there. Allowances carry a dollar figure for scope that is real but not yet defined enough to price.

This document protects both sides, which is exactly why it gets shortchanged. The owner reads it as the contractor hedging. The contractor who skips it is the one who eats the difference when the owner expected something the price never included. Write it down in plain language: the basis of the estimate, the drawings and revisions it was priced from, the assumptions, the clarifications, the exclusions, and the allowances with their amounts.

The test is simple. Six months into construction, when the scope in question comes up, the assumptions page should answer who owns the cost without a fight. If it cannot, it was not specific enough. A GMP, a hard bid, and a negotiated number all live or die on this document. The difference between a clean change and a dispute is usually whether the assumption was written down before the work started.

Allowances for undefined scope

An allowance is a placeholder dollar amount carried in the budget for scope that is real but not yet defined enough to price properly. The light-fixture package the owner has not selected. The finish in a space still being designed. A utility connection whose final routing is not set. You carry a number so the budget is complete, and you name it as an allowance so everyone knows it is a placeholder, not a hard price.

Allowances cut both ways. They keep the budget honest about scope that exists but is not drawn, and they are a common place for the number to drift, because the actual cost lands when the scope is finally defined and it rarely lands under the allowance. Carry allowances for what genuinely cannot be priced yet, write down what each one assumes about quantity and quality, and reconcile them as the design resolves.

The discipline is to retire allowances, not accumulate them. Each one is a decision deferred, and deferred decisions are where budgets quietly grow. As the design firms up, convert allowances to real priced scope and tell the owner whether the actual came in over or under the placeholder. An allowance that rides all the way to construction is a surprise waiting to happen.

Buyout planning and the bid packages

Buyout is the plan for turning the estimate into signed subcontracts. The work gets broken into bid packages by trade and scope, the packages go to qualified subs, and the awards get made against the budget the estimate set. Planning the buyout in preconstruction means deciding how to slice the work, who is qualified to bid each package, and the order to award them, so the long-lead and early-start trades are bought first.

The estimate set the labor and material targets each package is measured against, and that pricing discipline is the estimating guide's subject, not this one. Preconstruction's job is the strategy: clean package scopes with no gaps and no overlaps, a qualified bidder list for each, and a sequence that gets the critical trades under contract in time to perform.

Bringing the major subs in early is part of this. On the trades that drive the cost and the schedule, the electrical contractor, the mechanical contractor, the structural sub, getting them in during design for pricing and constructability buys real trade knowledge the design team does not have. They know the gear lead times, the install sequence, and where the detail will not build. The earlier they are in, the more of that knowledge lands while it can still change the drawing instead of becoming a change order.

Permits, approvals, and utility lead times

Permits and approvals are long poles that sit outside the contractor's control, which is exactly why they have to be identified early. The building permit, the trade permits, special approvals for the use or the site, and the utility coordination for power, gas, and water each run on their own clock, and some of those clocks are long. A utility service upgrade or a new transformer from the power company can take as long as the longest piece of equipment on the job.

Map the permits and approvals the project needs, the submittal and review times for each, and the utility lead times, and put them on the schedule as the milestones they are. The permit that nobody started and the utility that needs a year of notice are not surprises if someone listed them in preconstruction. They become surprises when the design is done and the team learns the approval that gates the start was never filed.

Driving the owner's decisions on a schedule

The owner is part of the critical path even though no contract calls them a trade. Finish selections, scope choices, value-engineering approvals, budget sign-offs: every one is a decision only the owner can make, and a late decision moves the budget and the schedule as surely as a late delivery does. Preconstruction has to drive those decisions, not wait for them.

The tool is a decision log with dates. List the decisions the project needs, who owns each, the date it is needed by to keep the design and procurement moving, and the date it actually gets made. Tie the need-by date to the real driver behind it, the long-lead order that cannot be placed until the gear is selected, the drawing that cannot finish until the layout is approved. An owner who can see that a decision two weeks late pushes a 40-week order two weeks later tends to decide on time.

The failure mode is a team that lets decisions drift because pushing the owner feels impolite. The polite version is the one that costs the owner money. Late decisions blow the budget, and the decision log is how you make the cost of waiting visible before it is spent.

The handoff to the construction team

Everything preconstruction learned has to reach the field, or the planning was an expensive exercise that got filed and forgotten. The handoff is the deliberate transfer of the estimate, the assumptions, the schedule, the logistics plan, the buyout strategy, and the risk register from the precon team to the people who will build the job. On a smaller job those are the same people. On a large one they are not, and the knowledge falls through the gap if nobody owns the transfer.

The classic failure is the assumptions. The precon team priced the GMP around a set of assumptions, the field team never reads them, and the first time anyone opens the assumptions page is during the argument about a change. The field has to know what the number assumed, what was excluded, what the allowances cover, and which risks were live, because they are the ones who will hit those conditions.

Treat the handoff as a real meeting with a real package, not an email with a folder attached. Walk the field team through the budget basis, the schedule logic, the long-lead status, and the open risks. The estimate and the assumptions and the schedule are only worth what reaches the person holding the prints. Lose the knowledge in the handoff and the field re-learns it the expensive way.

Preconstruction as cheap insurance

The preconstruction fee is small against what it prevents. A team spends a fraction of a percent of the project cost on planning, and the return is the changes that never happen, the long-lead gear ordered on time, the constructability conflict caught on paper, and the budget that did not run away because someone reconciled it at every milestone.

The economics are easy to defend and hard to see, because the savings are invisible by nature. You cannot point to the change order you avoided. The owner who skips real preconstruction to save the fee usually pays it back several times over in field changes, schedule slips, and a budget that surprised everyone at the end. The contractor who treats precon as overhead to be minimized produces a GMP full of soft assumptions and gets to argue them later.

Spend the fee where the influence is highest, which is early. A dollar of planning while the design is moving is worth far more than a dollar of problem-solving once the work is in place. That is the case for the discipline, and it is why owners who have been through a bad hard-bid job rarely want to go back.

What to document

Preconstruction produces decisions, and a decision nobody can find later is a decision that gets re-argued. The record is what carries the reasoning from the planning room to the field and answers the question, months out, of why the number was what it was and what it assumed.

Keep the estimates at each milestone, the budget reconciliations between them, the value-engineering log with the owner's decisions, the assumptions and clarifications behind the GMP, the risk register, the long-lead list with order dates, and the owner decision log. Tie them to a field tool so they survive the handoff and stay with the job instead of in someone's inbox. FieldOS is built to hold the estimate basis, the assumptions, the VE log, and the decision record where the field team can actually reach them.

Common mistakes

• Joining the project after the design is finished, when there is no cost left to influence.
• Producing milestone estimates without reconciling them to the budget and the prior number.
• Skipping the constructability review and discovering the conflicts in the field.
• Value engineering that cuts real function, or substituting a spec without the owner's decision.
• Missing a long-lead item, so a 40-week piece of gear holds the whole job.
• Setting a GMP with assumptions and allowances that were never written down.
• Letting owner decisions drift because pushing for them felt impolite.
• Handing off to the field without walking them through the estimate basis and the assumptions.

Field checklist

Standards and references

Preconstruction does not run off a single code the way an installation does. It runs off estimating practice, the delivery-method contract, and a few recognized methodologies, and the right references depend on the contract and the project.

For estimate precision and how it tracks design, AACE International's estimate classification system is the common reference, with its Class 5 through Class 1 framework tied to the level of design definition. For the management of the work and the owner's delivery decision, the Construction Management Association of America publishes guidance on CM-at-risk and the manager's role. For design-build, the Design-Build Institute of America sets the practice and the contract forms. For the design phases the estimates track and the standard agreements behind a GMP, the American Institute of Architects contract documents define the milestones and the terms.

For value engineering, the value methodology from SAVE International, value as function divided by cost, is the recognized framework, and several public agencies run formal VE programs on the same basis. None of these are building codes. They are practice standards and contract frameworks, and which one governs depends on the delivery method you are under and what the contract says. The contract documents and the owner's program control the specifics, so confirm the method and the agreement before leaning on any one of them.

Terms and definitions

Preconstruction carries its own vocabulary, and the same word can mean different things across a contract, an estimate, and a schedule, so it is worth pinning the terms down.

The terms below are the ones that drive the decisions in this guide. Define them the way the contract uses them, because a GMP, an allowance, and a contingency each have a specific meaning that the agreement controls.

Preconstruction

The planning phase before construction, where the team prices the design, checks constructability, value-engineers, schedules, and locks the budget.

GMP

Guaranteed maximum price, the contractor's capped commitment, with the owner paying the actual cost of the work plus a fee up to the cap.

Value engineering

Finding an equal-function, lower-cost alternative in systems, materials, or methods, with the owner deciding, without cutting the value the owner needs.

Constructability

Whether the design can actually be built and sequenced, reviewed on paper before the conflicts reach the field.

Estimate class

The AACE measure of how defined the design was when the estimate was made, Class 5 conceptual through Class 1 on complete documents.

Long-lead

Equipment whose order-to-delivery time is long enough that the order date, not the install date, controls the schedule.

Allowance

A placeholder dollar amount carried for scope that is real but not yet defined enough to price.

Contingency

A dollar reserve for unknowns inside the defined scope, not a fund for scope the owner adds later.

Related tools

Calculators and readiness checks for this work

Compare your options

FAQ

People also ask

• How do you estimate electrical work?
• What is the critical path?
• What is value engineering in construction?
• How do I calculate maximum pulling tension?
• What is selective demolition?