The Real Cost of RFIs: What the Research Says About Design and Documentation Gaps

RFIs are necessary, but they are not free. We examine major RFI datasets, what design and documentation gaps contribute, and why the best RFI is often the one resolved before construction.

IN SHORT • A landmark Navigant analysis of 1,362 projects reported about 796 RFIs per project and estimated roughly eight hours and $1,080 of review/response effort per RFI in 2013 dollars. • That dataset equated to about 9.9 RFIs per $1 million of construction cost, but project type, size and delivery method materially affect RFI rates. • Peer-reviewed research has connected errors, omissions and redundant engineering documentation with significant RFI burden. • Recent RFI literature treats RFIs as necessary clarification tools but also as non-value-adding transactions that can introduce cost and schedule risk. • The strategic goal is not zero RFIs. It is fewer avoidable RFIs caused by questions that could have been resolved before work starts.

RFIs are one of the most normal documents in construction.

They are also one of the most misunderstood.

A Request for Information is not automatically evidence of a bad design. Contractors need clarification for legitimate reasons: unforeseen conditions, product substitutions, scope questions, sequencing constraints and owner decisions can all require formal communication.

But RFIs are still transactions.

Someone has to identify the question, gather evidence, write it, route it, review it, coordinate with consultants, respond, communicate the answer and sometimes revise the documents.

That consumes time.

And when the RFI exists because the contract documents were incomplete or inconsistent, the project is paying to resolve a question that may have been discoverable earlier.

The most cited RFI dataset is still useful — with caveats

Navigant Construction Forum analyzed RFI information from 1,362 projects representing roughly 1.1 million RFIs.

The projects ranged widely in size and sector. The analysis reported an average of approximately 796 RFIs per project, or 9.9 RFIs per $1 million of construction cost.

Navigant also estimated approximately eight hours of administrative and technical effort to review and respond to an average RFI. Using 2012–2013 labor-cost assumptions, it estimated a cost of $1,080 per RFI.

Multiplying that by the average RFI volume produced an estimated review-and-response cost of about $859,680 per average project.

Those numbers are widely repeated, but they should not be treated as a current universal price tag.

Labor rates have changed. Technology has changed. Project types differ. A routine dimension clarification is not comparable to an RFI requiring several consultants and redesign.

The value of the dataset is not the exact dollar figure.

It is the scale of the coordination process.

Hundreds of RFIs multiplied by multiple reviewers create a substantial management workload even before field impacts are considered.

RFI cost has at least four layers

The direct designer response is only one part of the cost.

1. Creation cost

The contractor or trade has to investigate the condition, locate the relevant drawings, capture screenshots, describe the problem and route the question.

2. Review cost

Architects, engineers, consultants, construction managers and owners may all spend time interpreting the question and coordinating an answer.

3. Waiting cost

If the RFI affects work in progress, the schedule cost can exceed the administrative cost.

Crews may be resequenced. Procurement may pause. A hold point may block follow-on work.

4. Change cost

The answer may require a drawing revision, material change, price change, re-fabrication or rework.

This is why two RFIs with identical administrative effort can have radically different total impacts.

What research says about documentation-driven RFIs

A peer-reviewed study of an electrical and instrumentation engineering contract examined 42 documents and identified 449 errors and omissions, plus substantial information redundancy.

The researchers found that RFIs were required as a result and estimated the indirect cost of raising those RFIs to the contractor at approximately 9% of the E&I contract value.

That is a case study, not a universal benchmark.

But it illustrates an important mechanism: poorly coordinated documentation creates clarification work.

A 2024 systematic literature review of RFI research similarly described RFIs as essential clarification documents but also as non-value-adding tasks that can create schedule and cost risks. The review highlighted the opportunity to use digital methods, text mining and natural-language processing to improve RFI management.

A newer dataset asks a more interesting question: where do RFIs come from?

In August 2026, Flikt published a root-cause analysis of 827 RFIs from three multifamily projects. The company reported that about 60% were driven by issues in the construction documents: missing information, conflicting information or instructions that could not be acted on as written.

This is a vendor-published study based on its own project data, so it should not be generalized to the whole industry.

But it is useful because it shifts the conversation from “How many RFIs did we have?” to “Why did we have them?”

That is exactly the question project teams should ask internally.

The difference between a useful RFI and an avoidable RFI

A useful RFI resolves a genuine uncertainty that could not reasonably have been resolved earlier.

For example:

·       Unexpected subsurface condition

·       Owner-driven change

·       Product unavailability

·       Existing-condition discovery

·       Field condition that differs from record documents

An avoidable RFI originates from information that was already present — or should have been present — in the design package.

Examples:

·       Wrong detail reference

·       Missing dimension

·       Conflicting dimensions

·       Plan/schedule mismatch

·       Drawing/specification contradiction

·       Trade coordination conflict

·       Revision mismatch

The goal of preconstruction is not to eliminate formal communication.

It is to move preventable clarification earlier, when it is cheaper and less disruptive.

Why document gaps survive into construction

If the answer is “the team should have caught it,” that is not much of an improvement strategy.

Projects produce too much information for perfect manual review.

Consider what a coordinator may need to evaluate:

·       Architectural plans

·       Structural plans

·       Mechanical plans

·       Plumbing plans

·       Electrical plans

·       Fire protection

·       Schedules

·       Details

·       Specifications

·       Addenda

·       Revised bulletins

·       Shop drawings

The potential relationships between those sources grow quickly.

A missing callout may be obvious only when someone follows the reference.

A conflict may become visible only when two disciplines are compared.

A specification mismatch may require searching a separate document set.

This is why some RFIs survive even strong design reviews.

RFI volume is a lagging indicator

By the time a project has 500 RFIs, the underlying coordination conditions have already happened.

A more proactive set of metrics would include:

·       Number of cross-document inconsistencies found before construction

·       Number of broken references

·       Number of revision-dependent conflicts

·       Number of missing details resolved pre-award

·       Number of drawing/specification discrepancies identified preconstruction

·       Percentage of RFIs categorized as document-driven

·       Average RFI turnaround time by cause

·       RFIs that affected critical-path work

These metrics help teams distinguish unavoidable questions from process failures.

The cost of slow answers

The financial effect of an RFI is highly dependent on where the question sits in the schedule.

An RFI about a finish color months before procurement is very different from a structural opening RFI raised while a trade is mobilized to install.

That means RFI prioritization should account for:

Decision deadline

When does the team actually need the answer?

Work dependency

Which activities cannot proceed without it?

Procurement dependency

Does the answer affect equipment, fabricated material or long-lead items?

Design dependency

Will the answer trigger changes in other disciplines?

Commercial dependency

Could the response alter scope or price?

Treating every RFI as the same administrative object hides its real risk.

Better RFIs start with better evidence

When an RFI is required, the quality of the question matters.

A weak RFI says:

“Please clarify duct conflict.”

A stronger RFI identifies:

·       Drawing number

·       Detail/section

·       Location/grid

·       Conflicting information

·       Relevant specification

·       Screenshot/markup

·       Exact question

·       Required response date

·       Known downstream impact

Better evidence can reduce the time designers spend reconstructing the problem.

This is also where AI-assisted drawing review can be useful. If a potential issue is detected with its affected sheets and evidence already attached, the team begins the RFI process with context rather than a blank form.

AI should not generate RFIs automatically without review

There is an obvious temptation to turn every detected issue into an RFI.

That would be a mistake.

Many issues can be:

·       False positives

·       Internal coordination matters

·       Already resolved elsewhere

·       Contractor means-and-methods decisions

·       Informational only

A good workflow is:

Potential issue → evidence → human review → accept/reject → RFI only when external clarification is actually required.

Otherwise technology can increase RFI volume rather than reduce it.

The takeaway

RFIs are not inherently bad.

They are a necessary part of construction communication.

But they are not free, either.

The research shows that RFI processing can consume thousands of hours on large projects, and documentation errors and omissions can create significant RFI burden.

The best target is therefore not “zero RFIs.”

It is fewer avoidable RFIs, better evidence when an RFI is needed, and earlier resolution of document-driven questions before they become field delays.

Sources and further reading

1. American Society of Concrete Contractors — Request for Information (RFI) and Rework. Summarizes Navigant’s analysis of 1,362 projects, average RFI volume, time and cost. https://ascconline.org/Home/News/articleType/ArticleView/articleId/121/Request-for-Information-RFI-and-Rework
2. Navigant Construction Forum — Impact & Control of RFIs on Construction Projects. Original 2013 research perspective; dataset of 1,362 projects and 1.1 million RFIs (archived copies circulate online). https://www.coursesidekick.com/management/26981443
3. Love et al. — Assessing the impact of RFIs in electrical and instrumentation engineering contracts. Peer-reviewed case study linking errors/omissions/redundancy to RFI burden. https://research.polyu.edu.hk/en/publications/assessing-the-impact-of-rfis-in-electrical-and-instrumentation-en
4. Construction Innovation — Towards digital approach for managing RFI in construction projects. 2024 systematic literature review of RFI processes, risks and digital opportunities. https://www.sciencedirect.com/org/science/article/pii/S1471417524000058
5. Flikt — Where RFIs Actually Come From. 2026 root-cause study of 827 RFIs from three multifamily projects; useful emerging dataset with vendor-study limitations. https://flikt.ai/rfi-root-cause-study/