Construction Submittal Log Process

Most submittal logs are filing systems. They should be schedules.

That single confusion causes more procurement damage than anything else in construction administration. A log built as a filing system answers the question “where is this submittal?” A log built as a schedule answers the question “which submittal, if it slips one more week, delays the building?” Those are different documents, and only the second one is worth maintaining.

I have watched a project lose eleven weeks on switchgear because the submittal was approved on time. Approved on time, filed correctly, status current, nobody’s fault by any reasonable accounting. The problem was that “on time” had been defined as sixty days after subcontract execution, which is what the specification said, and nobody had worked backward from the date the switchgear had to be energized through a forty-two week manufacturing lead time to discover that sixty days was already four weeks too late on the day the subcontract was signed.

The log was perfect. The arithmetic was never done.

This article treats the submittal log as what it actually is: the procurement schedule for a building, expressed as documents. Eight stages, from extracting requirements out of the specifications through retiring the log at closeout, with the backward-scheduling calculation that most teams skip and the metrics that tell you whether the process is working before it fails.

What the Submittal Log Actually Is

A submittal log is the register of every document, sample, and data package the contract requires the contractor to submit, with each item’s responsible party, required submission date, review status, disposition, and downstream procurement and installation dependencies.

The last clause is the one that separates a working log from a spreadsheet. A log that records status without recording dependency cannot tell you what a delay costs, which means it cannot support a decision.

Table 1. Submittal Log Compared With Related Registers

The submittal log sits upstream of most of those. Fabrication cannot release until a shop drawing is approved. Delivery cannot be scheduled until fabrication releases. Installation cannot start until delivery arrives. So a submittal date is not an administrative milestone, it is the first domino in a chain that ends at a work activity on the critical path.

Core Terminology

Submittal. Any document, sample, or data package the contract requires the contractor to provide for review or record.

Action submittal. Requires the design team’s review and approval before the work may proceed. Shop drawings, product data, samples.

Informational submittal. Provided for record without approval action. Certificates, test reports, qualification statements.

Submittal register. The complete list of required submittals, extracted from the contract documents. The log is the register plus tracking data.

Ball in court. Which party currently holds an item and owes the next action. The most useful single field in a log.

Review duration. The contractual time allowed for the design team’s review, usually stated in Division 01 as working days.

Cycle time. Actual elapsed time from submission to disposition, which frequently exceeds the contractual review duration.

Required on site. The date material must physically arrive to support installation without delaying the schedule.

Release for fabrication. The point at which an approved submittal permits manufacturing to begin. The real milestone on long-lead equipment.

Float to need. Days between the projected approval date and the latest date approval could occur without delaying installation. This is the field that turns a log into a management tool.

Deviation. Any difference in a submital compared to a specified requirement, whether or not a submitter identifies the difference.

Objectives of the Process

• Protect procurement. Long-lead equipment governs commercial project schedules more often than any trade’s productivity, and submittal approval is the gate on it.

• Make review time visible and accountable. Contractual review durations exist. Actual cycle times routinely exceed them, and without measurement nobody owns the difference.

• Catch deviations before installation. A deviation approved by inattention becomes a compliance problem at inspection or a warranty problem later.

• Reduce resubmittal cycles. Each resubmittal costs two review durations plus handling. The first-pass approval rate is the most improvable number in construction administration.

• Establish a defensible record. Delay claims turn on when things were submitted, when they were returned, and what the disposition said.

• Prevent the closeout crunch. Closeout submittals are specified in the same sections as everything else and get discovered at substantial completion, when the crews that would produce them have demobilized.

Why the Log Governs the Schedule

The mechanism is arithmetic, and it is worth setting out plainly because the conclusion surprises people.

Consider a piece of electrical distribution equipment. Working backward from the date it must be energized: commissioning needs two weeks before that, installation needs three weeks, delivery must precede installation, shipping takes two weeks, manufacturing takes thirty weeks after release for fabrication, and release requires an approved submittal. The design team has fifteen working days for review, which is three weeks, and if the first submission comes back requiring resubmission you spend another three. Add the subcontractor’s preparation time of three weeks after subcontract award.

That chain totals roughly forty-six weeks before the equipment is energized, and that assumes one resubmittal. On a twenty-four month project with energization at month eighteen, the submittal must be in the design team’s hands by about month seven. If the subcontract was awarded at month five and the specification allows sixty days for submission, the arithmetic is already tight before anyone has done anything wrong.

Table 2. Backward Schedule Calculation, Worked Example

Fifty weeks. On many commercial projects that means the electrical subcontract must be awarded before the drywall package has even been bid.

This is the calculation that makes the submittal log a schedule instrument. Run it for every long-lead item during preconstruction and you will discover which packages must be bought out early, which is a buyout strategy decision driven by submittal arithmetic rather than by trade sequence.

Stakeholders and Ownership

Table 3. Submittal Process RACI

R responsible, A accountable, C consulted, I informed.

Two assignments matter more than they appear. The scheduler is accountable for required-on-site dates, because those dates come from the schedule and nowhere else, and a log populated with dates the project engineer guessed is a log built on fiction. And purchasing is accountable for lead times, because lead times change with market conditions and a log carrying last year’s lead times will quietly mislead you.

The Eight Stages

Table 4. Stage Summary

Stage 1: Build the Register From the Specifications

Purpose. Capture every required submittal from the contract documents. Not from last project’s log.

This is where most logs are compromised at birth. A register copied from a similar project carries that project’s requirements, which are similar and not identical, and the differences are precisely the items nobody will look for.

Requirements live in more places than people check:

1. Part 1 submittals articles in every technical section. The obvious source, and still incompletely mined because reviewers stop at the first list.

2. Part 1 quality assurance articles. Installer and manufacturer qualification statements, mockup requirements, preconstruction test reports.

3. Part 2 source quality control. Factory test reports and certifications.

4. Part 3 field quality control. Field test reports, inspection records, startup reports.

5. Part 1 closeout submittal articles. Frequently a separate article from the main submittals list, and the most commonly missed.

6. Division 01 sections. Submittal procedures, quality requirements, closeout procedures, O&M data, demonstration and training, and commissioning each impose submissions across all trades.

7. Drawing general notes. Occasionally impose submittal requirements not repeated in the specifications.

8. Delegated design triggers. Each carries sealed calculations, drawings, and often a deferred submittal package to the authority having jurisdiction.

9. Addenda. Add and modify requirements silently.

A large technical section can list required submissions in four separate articles. A register built from one article per section will be perhaps sixty percent complete, and the missing forty percent is weighted toward closeout and quality items.

Stage 2: Number and Structure the Log

Purpose. Make the log navigable and the fields sufficient for management.

Number by specification section, then sequentially within section, then by revision. Section 23 31 13, third submittal, second revision, will be referred to as 233113-003-R2 in this current convention. Having a consistent formatting designation allows items to be arranged in order with the responsible section readily identified without the need to look up the section, and any gaps in the sequence indicate missing items.

Avoid numbering by date received or by subcontractor. Both make the log unsuitable against the specifications, which is the check that matters.

Table 5. Minimum Log Fields

Eighteen fields sounds like a lot. Six of them are dates and four are derived automatically. The set that most logs lack are required on site, lead time, float to need, and days aged, which are exactly the four that make the log actionable.

Stage 3: Schedule Backward From Need Dates

Purpose. Convert the register into a schedule.

For every item, work the arithmetic in Table 2. Most items are short-lead and the calculation is trivial. The value is in identifying the twenty or thirty items where it is not.

Method:

10. Get the required-on-site date from the schedule for the installation activity the submittal supports. Not from the subcontractor, and not from the specification.

11. Get the lead time after release for fabrication from purchasing or from the subcontractor’s supplier, current as of now.

12. Add fabrication, shipping, and a staging buffer.

13. Add the contractual review duration, and add a second review cycle for anything complex, delegated, or historically prone to rejection.

14. Add contractor intake screening time.

15. Add subcontractor preparation time after award.

16. The result is the latest date the subcontract can be awarded and the latest date the submittal can be submitted.

17. Compare against the actual buyout schedule. Where the required award date precedes the planned award date, you have a procurement problem that must be solved in buyout sequencing, not in submittal administration.

Table 6. Long-Lead Categories Worth Explicit Backward Scheduling

Treat those ranges as prompts to ask rather than as data. Lead times move with market conditions, and the number that matters is the one a supplier will commit to this quarter.

Stage 4: Assign and Issue Requirements

Purpose. Explicitly tell each subcontractor the amount and due date for their obligation.

Issue a subcontractor-specific extract from the log at subcontract execution, listing every submittal that party owes, with the submit-by date rather than the specification’s relative deadline. Attach it to the subcontract where possible, because a submittal schedule that is contractual is enforceable and one that is informational is aspirational.

Two practices worth adopting. Hold a submittal kickoff with each major subcontractor, walking the list and confirming their supplier lead times against your assumptions, because your lead time data is second-hand and theirs is not. And state the review duration you will hold the design team to, so the subcontractor understands the cycle they are working within.

Stage 5: Intake Screening

Purpose. Reject incomplete submissions before they consume a review cycle.

This is the highest-leverage stage in the process, and it is also the most commonly missed stage. People often skip this stage because it feels like effort that’s been done before. This is not the case. A submittal forwarded incomplete comes back rejected, and that costs a full review duration plus handling on both sides. Screening it out costs an hour.

Screen for:

• Every item the specification’s submittal article requires is present, checked against the register entry

• Product data covers the specified product, not a similar one

• Shop drawings show the actual project conditions, not catalog details

• Dimensions reflect field conditions where field verification was required

• Deviations from the specification are identified by the submitter, and where they are not, flag them yourself

• The subcontractor has reviewed and stamped it, which is a contractual requirement in most general conditions and is routinely ignored

• Referenced items are attached

• Delegated design submissions carry the seal of an engineer licensed in the project jurisdiction

• The submittal is legible, complete, and correctly identified

The deviation check deserves emphasis. Most general conditions require the submitter to identify deviations in writing. Most submitters do not. A deviation that reaches the design team unflagged and gets approved has arguably been accepted, and untangling that later is expensive. Catching it at intake is a five-minute comparison against the specification.

Table 7. Intake Screening Outcomes

The third row contains an important convention. A submittal returned by the contractor at intake should not consume a revision number, because revision numbers should record design team review cycles. Conflating the two makes the first-pass approval rate unmeasurable, and that metric is the main lever on review efficiency.

Stage 6: Transmit and Manage the Review Cycle

Purpose. Keep items moving and make delay visible while it is still recoverable.

Transmit with a defined review duration stated on the transmittal, referencing the Division 01 requirement. Then manage two things: aging and priority.

Aging. Report days in current position, every week, for every open item. An item that has been with the design team for eighteen working days against a fifteen-day requirement is a contractual issue, not a relationship issue, and raising it at eighteen days is easier than raising it at forty.

Priority. Not every item deserves equal attention. Sort by float to need and work the top of the list. A submittal with four days of float matters more than one with ninety, regardless of which arrived first.

Where the design team is a bottleneck, the useful intervention is data rather than complaint. A monthly report showing average cycle time against contractual duration, by discipline, changes behavior in a way that meeting discussion does not. Where a consultant is consistently over, that is a conversation for the owner, because the owner holds that contract.

Table 8. Escalation Thresholds

Stage 7: Disposition, Resubmittal, and Release

Purpose. Convert a returned submittal into either released work or a controlled resubmittal.

Disposition codes vary by contract and the exact wording matters, because it determines whether work may proceed and who carries the risk if it does.

Table 9. Typical Disposition Codes and What They Permit

Two operational cautions.

The second row is where risk quietly transfers. “Approved as noted” means the contractor is now responsible for ensuring every note is incorporated, and if a note conflicts with the specification or with another trade’s work, proceeding without raising it accepts the conflict. Read the notes rather than filing the approval.

The last row is a trap. A conditional approval that requires a response but permits work to proceed creates a situation where the work is built and the condition is unresolved. Treat conditional approvals as requiring closure before release.

Resubmittal management. Track revisions per item and treat the third revision as a process failure requiring intervention instead of another iteration. There are four probable causes for repeated rejection. Either the subcontractor is not comprehending the requirement, the requirement specification is unclear, the reviewer is following a standard the document does not contain, or there is a substitution attempt that has not been declared. Each has a different remedy and none of them is a fourth submission.

Release and distribution. On approval, two things must happen. Release for fabrication, communicated to the subcontractor and recorded with a date, because that date starts the lead time clock everyone will later ask about. And distribution to the field, superintendent, and any affected trade, with a record. Field crews building from a superseded submittal is a real and recurring failure, and the cause is almost always distribution rather than approval.

Stage 8: Closeout Submittals and Retirement

Purpose. Collect what is owed at the end, before the people who can produce it leave.

Closeout submittals are specified in the same sections as everything else and they are the most commonly missing register entries. Warranties, operation and maintenance data, record drawings, training records, attic stock inventories, certificates, final test reports, and commissioning documentation.

The failure pattern is consistent. These items are identified at substantial completion, requested from subcontractors whose crews demobilized weeks earlier, and produced badly and late under retention pressure. The remedy is to schedule them like any other submittal, with submit-by dates derived from substantial completion rather than discovered at it, and to make specific closeout items conditions of specific payment applications rather than of final payment alone.

Start collecting at roughly seventy-five percent completion. Report outstanding closeout submittals monthly from that point, by subcontractor, in the same meeting where payment is discussed.

Required Documentation

Table 10. Documents Required to Run the Process

Technology Integration

Table 11. Submittal Log Approach Comparison

The distinction that matters is the first column. Workflow platforms are good at moving submittals through a process and largely indifferent to whether the register they are moving is complete. A platform with excellent notifications tracking a register missing forty percent of its closeout requirements will run smoothly to a bad outcome.

AI-Assisted Opportunities

Two parts of this process are genuinely constrained by document volume, and both are the parts that fail quietly.

The first is Stage 1. Extracting every submittal requirement means reading four articles in every section of a project manual that may run 1,200 pages, plus Division 01, plus delegated design triggers, plus addenda. Nobody does that exhaustively inside a buyout window, so registers get built from templates and the differences go unnoticed.

The second is Stage 5. Screening a submittal against the specification requires comparing what was submitted against what the section demands, item by item, including identifying deviations the submitter did not flag. Done properly on a large submittal that is an hour of careful reading, which is why it gets done quickly or not at all.

Platforms built to interrogate construction documents address both. iFieldSmart AI, for instance, indexes the full project manual alongside the drawings, extracts submittal requirements across every section and article with citations back to the source, and compares submitted documents against the governing specification requirements to flag missing or non-compliant content. That converts register construction from a sampling exercise into a complete one and gives intake screening a baseline to check against.

Table 12. AI Use Case Matrix for Submittal Management

Implementation

18. Audit your current log against the specifications on a live project. Pick three technical sections, extract every submittal requirement from all four article types, and compare against the log. The gap is usually instructive and usually concentrated in closeout and quality items.

19. Add the four missing fields: required on site, lead time, submit-by date, float to need. This is the change that converts the log from a filing system into a schedule.

20. Run the backward calculation for long-lead items during preconstruction, and let the output drive buyout sequencing rather than the reverse.

21. Institute intake screening with a written deficiency return. Do not let contractor returns consume revision numbers.

22. Report aging weekly and cycle time monthly, by discipline. Data changes review behavior; meetings do not.

23. Start closeout collection at seventy-five percent completion and tie specific items to specific payment applications.

24. Measure first-pass approval rate and treat third revisions as process failures requiring a meeting.

25. At project completion, review which items caused schedule impact and why, and fix the register template or the process stage responsible.

Table 13. Implementation Roadmap

Best Practices

Table 14. Submittal Log Best Practices

Common Mistakes

Table 15. Common Submittal Log Failures

How This Plays Out on Real Projects

Class A office tower, 340,000 square feet. Curtain wall drove everything. The backward calculation showed that unitized glazing needed release for fabrication eleven weeks earlier than the buyout schedule contemplated, because the enclosure milestone gated every interior trade. The team moved the glazing award ahead of three other packages purely on submittal arithmetic. That is Stage 3 doing what it exists for.

Regional hospital expansion, 190 beds, occupied campus. Deviation detection at intake. A submitted air handling unit differed from the specified unit in filtration configuration, and the deviation was not identified by the submitter. On healthcare work filtration is a licensing matter rather than a preference, and an approved deviation would have surfaced during infection control commissioning with occupancy at stake. Five minutes of comparison at intake against a section the engineer had already read.

Hyperscale data center, 60 MW. Long-lead electrical equipment, where the arithmetic is unforgiving. Switchgear and transformers carried lead times that put required award dates before the design was complete enough to buy from, which meant the team had to procure against an early release package with a defined change mechanism. That is a strategy decision, and it only becomes visible if somebody runs Stage 3 during preconstruction rather than starting the log at mobilization.

Advanced manufacturing facility, food grade. Process equipment vendor submittals arriving after the building submittals were closed. Because vendor documents were known to be late, the log carried the dependent items as held with the dependency recorded rather than as overdue, and the review capacity was reserved. Treating a known-late dependency as a scheduled event rather than a failure is the difference between managing it and reacting to it.

University science building, 11 floors. Resubmittal spirals on laboratory casework. Three items reached a fourth revision, which triggered the intervention meeting and revealed that the reviewer was applying a standard the specification did not state. Two of the three closed on the next cycle after the requirement was clarified by RFI. Repeated rejection is a signal about the requirement, not only about the submitter.

Multifamily podium, 240 units. Closeout submittals, the residential pattern. Warranties, O&M data, and attic stock across twenty-two subcontractors, most of whom demobilize progressively as buildings turn over. Starting collection at seventy-five percent and tying specific items to specific payment applications converted a retention fight into a routine collection.

Highway interchange, design-bid-build. Certified test reports and material certifications, which dominate infrastructure submittal logs. The register requirements sat in measurement and payment sections as much as in technical sections, and missing a certification meant unpaid work rather than delayed work. Register completeness matters differently when payment is tied to documentation.

Historic warehouse conversion to hotel, 140 keys. Mockups and preconstruction testing. Three sections required mockups approved before fabrication release, which inserted an additional approval gate ahead of the normal cycle and extended the effective lead time on custom millwork and window assemblies. Mockup requirements live in Part 1 quality assurance articles, which is one of the four article types a template-based register never captures.

Frequently Asked Questions

How early should the submittal log be built?

Register extraction should happen during preconstruction, alongside buyout planning, because the backward calculation for long-lead items has to influence buyout sequencing to be useful. Waiting until mobilization means the log documents procurement decisions already made. On negotiated and CMAR work, start at the point you have a substantially complete project manual.

How many submittals should a commercial project have?

Calculating submittals for a project with about 200 sections can be roughly estimated between 300 and 600 submittal items including closeout submissions. One can expect 1.5 to 3 submittals per specification section. Quality assurance and closeout submissions almost definitely weigh in on the missing submittal items in a log with 150 items.

What is the difference between an action and an informational submittal?

An action submittal requires the design team’s review and approval before the work may proceed. An informational submittal is provided for record without approval action. The distinction matters for routing and urgency, and it matters commercially, because withholding work pending approval of an informational submittal wastes time while proceeding without approval of an action submittal transfers risk to you.

Who is responsible when the design team exceeds the review duration?

Contractually, the design team, and by extension the owner who holds that agreement. Practically, the contractor bears the schedule consequence unless the delay was documented contemporaneously. That is why written notice on the day the duration expires matters more than it feels like it does at the time. A cycle time report by discipline is the single most effective tool for changing review behavior.

How do we reduce resubmittals?

Intake screening first, which eliminates the avoidable rejections. Then measure first-pass approval rate by subcontractor and by reviewing discipline. A low rate concentrated in one subcontractor is a capability issue addressed by a kickoff meeting. A low rate concentrated in one reviewing discipline is usually either an ambiguous specification or a reviewer applying an unstated standard, and both are addressed by clarifying the requirement rather than by resubmitting.

Should the contractor review submittals before forwarding them?

Yes, and most general conditions require it, including a contractor stamp. The review is for completeness, coordination with other trades, dimensional consistency with field conditions, and identification of deviations. It is not a technical design review, and framing it as one creates exposure. Screening is about whether the submission is complete and whether it departs from the specification, both of which the contractor is positioned to judge.

How should deviations and substitutions be handled?

They are different things procedurally. A deviation is a departure within an otherwise specified product and must be identified in writing at submission. A substitution is a different product, and Division 01 sets a procedure with a deadline that has usually passed by the time construction starts. Most substitution rejections are procedural rather than technical. Where a subcontractor submits a substitution disguised as a product data submittal, return it and require the substitution procedure.

What does “approved as noted” actually obligate us to do?

Incorporate every note, and take responsibility for having done so. The risk sits in notes that conflict with the specification, with another trade’s approved submittal, or with a field condition. Proceeding without raising such a conflict accepts it. The practical habit is to read the notes as instructions requiring verification rather than as an approval requiring filing.

How do we keep field crews from building off superseded submittals?

Distribution with a record, and a single controlled location for current approved submittals. This failure is nearly always a distribution problem rather than an approval problem: the correct version exists and the crew has an older one. Reviewing the current submittal status with each trade before it mobilizes closes most of it.

When should closeout submittals be collected?

Start at roughly seventy-five percent completion and report outstanding items monthly by subcontractor from that point. Tie specific closeout items to specific payment applications rather than to final payment alone, because final payment leverage arrives after the people who could produce the documents have gone. Warranties, O&M data, training records, and attic stock inventories all take longer to assemble than anyone expects.

How does the log interact with the RFI log?

Bidirectionally, and the interaction causes real delays. A submittal may be held pending an RFI answer, and an RFI may arise from submittal review. Record the dependency explicitly in both logs and escalate the dependency rather than the dependent item. Circular waiting, where a submittal waits on an RFI that waits on a submittal, is common and only visible if both logs carry the cross-reference.

What metrics actually indicate the process is healthy?

Four. First-pass approval rate, which measures submission and requirement quality. Average cycle time against contractual duration by discipline, which measures review performance. Count of items with float to need under fifteen working days, which measures schedule risk. And closeout submittals outstanding as a percentage of the register from seventy-five percent completion onward, which predicts the retention fight. Aging and open item counts are useful operationally but say little about whether the process is working.

Expert Recommendations

For the project engineer: extract the register from this project’s specifications and mine all four article types per section. The extraction you do in week one determines whether the log can be trusted in month fifteen, and closeout requirements are the ones that will be missing.

For the project manager: run the backward calculation on long-lead items before buyout sequencing is fixed, and let it change the sequence. Three to six packages on a typical commercial project need earlier award than trade logic suggests, and submittal arithmetic is the only thing that reveals which.

For the scheduler: own the required-on-site dates and push back when the log carries dates that did not come from the schedule. A log populated with estimates cannot compute float, and float is the whole point.

For purchasing: refresh lead times per project and per quarter. Lead times on electrical distribution, transformers, and generators have moved substantially and repeatedly, and a log carrying last year’s figures will mislead confidently.

For the superintendent: review current submittal status with each trade before mobilization, and insist on a single controlled location for approved submittals. Crews building from superseded documents is a distribution failure and it is preventable.

For the project executive: ask for the cycle time report by discipline monthly and take it to the owner where a consultant is consistently over the contractual duration. That is an owner conversation, and having it early with data is far more effective than having it late with a claim.

For the owner or owner’s representative: ask your contractor for the count of items with float to need under fifteen days. It is one number, it is a direct measure of procurement risk on your project, and a contractor who cannot produce it does not have a scheduled log.

Where This Leaves You

The submittal log is the least respected document in construction administration and one of the most consequential. It gets treated as clerical work, assigned to the most junior person available, and maintained as a status record. Meanwhile it governs when fabrication releases, which governs when material arrives, which governs when work starts.

Two changes convert it. Build the register from the actual project manual rather than a template, mining all four places requirements hide. And add the four scheduling fields so the log can tell you what a delay costs rather than only that a delay occurred.

Neither requires software or headcount. Both can be done on the project you are running now, and the backward calculation on your long-lead items will probably tell you something uncomfortable this week rather than in month fifteen.