Four Document Classes One Reconciliation Engine and What That Architecture Means for Underwriting Division 8 Bids

To reconcile a door schedule with floor plans and specs at the same time, a Division 8 reconciliation engine must also read door and frame elevations. Its purpose is to assemble each opening’s requirements and expose contradictions before pricing. A Division 8 bid is underwritten by the reliability of that reconciliation.

Why a Division 8 bid is structurally different from every other trade scope

Most trades price from a governing quantity survey, linear measure, or material schedule. Division 8 has a different starting condition. The information needed to price a single opening is distributed across the door schedule, door and frame elevations, partition schedules and floor plans, and the hardware specification. None is complete on its own, so the estimator must establish agreement across documents before treating an opening as a priced assembly.

That distribution changes what constitutes a complete takeoff. An opening can be identified correctly while its requirements remain unresolved, because the record in one document depends on information held in another. The door schedule is usually the first document opened, but it primarily establishes what openings exist. It is not a reliable standalone statement of everything those openings require. Treating it as one embeds missing information in the price as an unstated assumption.

What each of the four document classes actually contributes to an opening's price

The door schedule establishes opening identity through the door mark, size, type, and fire label. Its apparent completeness can be misleading. Handing, leaf count, and wall type are often left blank, so even a filled-in schedule row may leave out parts of the assembly. Door and frame elevations supply the leaf count and pair status that the schedule may omit. A pair is one opening with two leaves, and the elevation shows both leaves drawn.

The partition schedule and floor plans establish the wall conditions that drive frame selection. Those conditions include the stud depth, total wall thickness, fire rating, and STC. The frame depends on both the wall and the door. What estimators call jamb depth or throat size depends on stud and wall information held in these documents and absent from the door schedule.

Section 08 71 00 supplies the hardware requirements through groups such as HW-1, HW-2, and HW-3. Each set names the brand, model, function code, and finish for every piece on the opening, while the specification also governs approved manufacturers, substitution rules, and finishes. These requirements cannot be inferred from the opening’s dimensions. F&K Estimatings’ scope description says that leaving out panic hardware, continuous hinges, and magnetic hold-opens can cause major budget overruns. Those items live in the specification.

SMA Estimating’s takeoff formula defines the per-opening assembly cost through the panel, frame, hardware set, glazing, sealant, and installation labor. Completing that assembly requires all four document classes. The schedule establishes the panel, the elevation establishes leaf count, the partition information determines the frame requirements, and the specification supplies the hardware set.

How conflicts between the four documents create bid risk that doesn't appear in the opening count

The most financially dangerous conflicts can leave the opening count unchanged. A door may be assigned to hardware group 107.3 in the schedule and group 107.4 in the specification. Both references describe the same opening, but they cannot both govern its hardware assignment. A review that checks only totals has no reason to flag the discrepancy. Both values must become visible before the estimator can decide which document governs.

A second conflict is an assignment to a hardware set that does not exist in the specification. The opening is counted, but its hardware is absent. Because hardware typically represents a large share of the assembly cost, the resulting bid is structurally short.

Plan-to-schedule conflicts create a related risk at the opening-identity level. Doors can appear on plans but not in the schedule, appear in the schedule but not on plans, or appear repeatedly on plans against a single schedule entry. Each document can still look internally consistent. SMA’s verified-count formula is Plan Count + Schedule-Only Units − Duplicates − Exclusions.

These conflicts locate the risk at the reconciliation layer. An undetected hardware discrepancy becomes an RFI, a resubmittal, or a cost overrun during the submittal phase after the contract is awarded, when the hardware distributor must produce a detailed schedule listing every item for every door by group.

Why sequential, document-by-document review cannot eliminate that class of conflict

Sequential review can be disciplined without making cross-document contradictions reliably visible. SMA’s workflow proceeds through reading schedules, setting scale, counting one type at a time, coloring as the count progresses, marking matchlines, and reconciling against the schedule. Each step has a valid purpose. Yet the sequence still moves attention from one document to another, rather than holding all four document classes against the same opening while it is being assembled.

The limitation belongs to the process. A large hospital package can have many separate schedules across many pages, so the estimator has more information to keep track of between passes. While counting, the estimator must keep schedule assignments, plan locations, elevation details, and specification requirements linked in working memory. More than eight in ten bid packages have inconsistencies from the architect that the estimator must sort out.

A sequential process can catch individual contradictions, but it cannot eliminate the class of risk simply through more careful counting. Most of its effort goes into extracting quantities, while the simultaneous comparison receives little of that time. Steel Krafts’ takeoff practice places a licensed AHC at this junction to flag ambiguous openings so that quantity and compliance questions are resolved together. That response adds expertise where the process needs judgment, rather than treating the discrepancy as a failure of diligence.

How does a reconciliation engine check door schedules against plans, elevations, and specs at once?

A Division 8 reconciliation engine changes the unit of review from the individual document to the opening described across documents. While reading a drawing, it also holds the specification, elevations, plans, and schedule in view. The estimator gets the information for each opening brought together with an organized list of conflicts, saving the work of cross-referencing separate extracts by hand. This is the operative distinction from symbol counting. Recognizing a door on a plan does not establish whether its scheduled hardware exists or agrees with the specification.

The same pass supplies fields that the architect left blank. Handing comes from the floor plan, leaf count from the elevation, and wall type from the partition schedule. Pair detection makes the difference concrete. The engine reads the two leaves in the elevation and reports leaf count separately from opening count, so both quantities are available before pricing. What otherwise requires a deliberate additional pass becomes part of reconciling the opening.

Hardware disagreements remain decisions for the estimator. Where the schedule assigns one set and the specification assigns another, the output lists both values for the estimator to review. The engine places the contradiction in the estimator’s decision list, where the estimator can decide which document governs before the number is issued.

That output changes the allocation of estimating time. Roughly ninety percent of the package consists of straightforward openings that can be assembled where the documents agree. The rest goes into a short, organized list for the estimator to judge. The estimator’s work starts with the unresolved conditions rather than with the task of remembering where those conditions might be hidden.

Where institutional jobs raise the reconciliation requirement beyond what the project spec alone supplies

Institutional work can add an owner-standard layer above the four project document classes. Correct agreement within the project set is then insufficient if the result conflicts with the owner’s governing requirements. The University of Michigan’s Architecture, Engineering and Construction design guidelines show that a September 2017 revision to Section 087100 changed office lockset part numbers for classroom use. A distributor quoting a campus project needs to check a listed product’s approval against the applicable owner standard as well as the project specification.

Michigan State University’s CP25054 specification sets out which work falls under separate contracts. It states that new cores for door hardware will be completed by MSU Key Shop. That provision changes the scope being priced. The reconciliation must distinguish the hardware requirements from the work assigned to the owner’s operation, or an otherwise complete opening assembly can include work outside the bidder’s scope.

Western Carolina University’s 2025 construction documents make hardware-schedule format a condition of review. They require the Sequence of Format for the Hardware Schedule published by the Door and Hardware Institute, with nonconforming schedules returned for correction before checking. USC Health Sciences Center specifications also require the distributor to have staff with the stated DHI credentials, including an Architectural Hardware Consultant.

Michigan DTMB’s statewide door hardware specification adds a compliance constraint for fire-rated openings. It requires compliance with NFPA 80 and hardware tested and listed by UL for the required door types and sizes, consistent with door and frame labels. Together, these examples establish distinct reconciliation obligations involving product approval, scope, schedule format, personnel, and compliance. They do not establish identical requirements for every institution.

Where an owner standard governs, hardware sets must be authored against it and checked for conflicts with the project specification. Authority runs downward from that governing standard. Where the standard defines approved manufacturers, an unlisted product requires the applicable substitution process rather than an assumption that project-level “or equal” language grants permission. The additional layer changes what counts as a resolved opening.

What the reconciliation architecture means for bid confidence, quote validity, and volume

A defensible number requires resolved scope. All four document classes must have been compared, conflicting requirements addressed, and pair counts separated from opening counts before the takeoff can support that number. The practical gain is less estimating time spent reconstructing relationships that the output can present together, leaving more time for the decisions that determine price and qualification.

The volume argument follows from the same constraint. If reconciliation is the bottleneck, compressing it reduces total time per bid. A firm able to bid in under an hour per package can pursue more opportunities without adding estimating headcount. Distributors and smaller operators can also get the checks across documents that are needed to catch hardware-set conflicts. Confidence becomes less dependent on the size of the estimating team.

Three Excel files cover schedules, counts, and hardware, with an issues report that lists every discrepancy and a marked-up PDF. The issues report becomes the basis for written clarifications to the architect. Those clarifications set out the bid’s qualifications and protect margin after award by making unresolved requirements explicit before they are incorporated into the quoted number.

A recap identifies the key system, access control requirements, and approved manufacturer list before the full takeoff runs, allowing an estimator to rule out an unattractive package before opening a plan sheet. That is a bid-board decision.