The RFI: Construction's Cheapest Insurance Policy

What an RFI Is For

A Request for Information is a formal written question about the construction documents. A structural drawing shows a beam where the mechanical drawing shows a duct run. A specification calls for a product that is no longer manufactured. An existing wall is 90mm off where the as-builts said it would be. In each case the contractor cannot proceed without a decision, and the decision belongs to someone else.

The RFI is how that question gets asked in a way that counts. An informal call to the architect may well produce the same answer faster, but it produces no record, and a verbal instruction that later turns out to have added three weeks to the schedule is worth approximately nothing in a claim.

That evidentiary function, not the clarification itself, is why the RFI process exists in the form it does. The document is a question. The log is the asset.

What Belongs in One

The quality of an RFI determines the speed of its answer, and speed is the thing that costs money. A well-formed RFI does four things: it states the specific conflict, it cites the exact drawings and specification sections in tension, it explains the schedule or cost impact of not knowing, and it proposes a resolution.

That last item is the one most often left out and the one that most reliably shortens the cycle. An RFI that asks "please advise" requires the architect to originate a solution. An RFI that says "we propose routing the duct below the beam per detail 4/A-502, which maintains the required clearance — please confirm" requires them only to agree or correct. One of those gets answered this week.

• →The specific question, in one sentence, answerable yes or no where possible.
• →Drawing numbers, detail callouts and specification sections that conflict.
• →Photographs of the actual field condition where one exists.
• →The proposed resolution, with the reasoning behind it.
• →The date by which an answer is needed to avoid schedule impact, and what happens if it is missed.

The Response Clock Is the Real Story

Most contracts set a response period for RFIs. The number varies, and the contract governs, but the operational point is the same regardless of what it says: the elapsed days between submission and response are a tracked quantity, and they are the raw material of any later delay claim.

This is where the RFI log stops being administrative overhead and starts being a financial instrument. A project with forty open RFIs averaging twenty-two days against a ten-day contractual window has a documented, quantified impact. The same project with forty phone calls has a feeling.

Why RFIs Get Expensive

An RFI costs almost nothing to raise and a great deal to leave open. When an answer does not arrive, a crew does one of two things, and both are costly. They stop, which is a documented but real schedule hit. Or they proceed on an assumption, which is faster right up until the assumption is wrong and the work is torn out at somebody's expense.

The second failure mode is the one that hurts, because the cost lands long after the decision that caused it, and by then the question of who authorised the assumption is a matter of recollection rather than record.

The mitigation is unglamorous. Log every RFI with its submission date. Track elapsed days against the contractual window. Escalate on a defined schedule rather than when someone remembers. Put the proposed resolution in the original submission so that the fastest path for the responder is also the path you want. None of that is clever, and all of it is cheaper than a tear-out.