A typical residential project generates 40-90 RFIs. The average time-to-resolution on those is 6.4 calendar days, and roughly a third of them block a downstream trade while they sit open. If even a quarter of those RFIs are sloppy enough to require a follow-up before the architect or engineer can respond, the GC has just handed back two to three weeks of float across the project.
The five mistakes below are the ones we see repeated almost universally. None of them require any tooling to fix. They require a checklist and a small amount of discipline at the moment of authorship.
Mistake #1: A vague subject line
The subject of the RFI is the only thing the architect reads on their phone notification at the airport. "Question about the stairs" will be ignored. "Confirm tread depth at U-stair between L1 and L2 — A-203 detail 4 shows 10 in, A-101 plan shows 11 in" will be answered in a coffee shop in three minutes.
Fix: Every subject is a question, names the location, and cites the sheet. Three components, every time, no exceptions.
Schedule impact when fixed: 1-2 days saved per project. Avoiding three follow-up cycles on five RFIs is real time.
Mistake #2: Missing context
A good RFI gives the recipient everything they need to answer it without opening their own file. That means a clip of the sheet in question with the area circled, a photo of the field condition if relevant, and a one-paragraph statement of what the GC currently intends to do absent the answer.
The third item is the one that gets skipped. Architects answer RFIs faster when the question implicitly tells them the cost of not answering. "If we don't hear back by Thursday we are planning to frame at 10 in tread depth per A-203" produces an answer on Wednesday. "Please clarify" produces an answer on the architect's normal schedule, which is whenever they get to it.
The best RFI you can write is one where the recipient can answer "yes that's right" and you proceed.
Fix:Attach the sheet snippet, the field photo if applicable, and a one-sentence statement of the GC's working assumption.
Schedule impact when fixed:2-3 days saved per project. Eliminating the architect's "clarify what you are asking" round-trip is the biggest single lever in this list.
Mistake #3: No priority designation
Treating every RFI as the same priority means the architect processes them in arrival order. That is not what you want. You want the RFI that is blocking your framer answered before the RFI asking what stain color to use on the deck rail.
Use three priorities. Resist the urge to invent more.
Critical.Blocking a trade currently on site or about to mobilize within 48 hours. Architect's expected turnaround: 24 hours.
Normal. Affects work scheduled in the next 1-3 weeks. Expected turnaround: 5 days.
Fix: Set the priority before you send. Critical RFIs get a phone call follow-up the same day.
Schedule impact when fixed:1-3 days saved per project. Critical RFIs get the architect's focused attention and clear faster.
Mistake #4: No deadline on the RFI itself
An RFI with no "needed by" date is not actually a request. It is a note. The architect cannot prioritize their queue against your other RFIs if none of them have dates.
The deadline you set should be the date the answer needs to be in hand to keep the schedule. Not a comfortable buffer past that, not the date you would like it. The date it has to be answered.
Fix:Every RFI has a "needed by" date. The date appears in the subject line if it is critical: "[NEED BY THU 5/14] Confirm tread depth..." The PM maintains a tracker of overdue RFIs and follows up by phone the morning after each one goes overdue.
Schedule impact when fixed: 2-4 days saved per project. The follow-up discipline is what actually closes the gap — the deadline by itself is just a target.
Mistake #5: No audit trail
The fifth mistake is what kills you not on schedule, but on disputes. An RFI answered by text message, by phone call, or by "the architect said it was fine on the walk-through" is not an RFI. It is hearsay.
Three months later when the homeowner's lawyer asks why you installed a 10-inch tread instead of 11 inches as drawn, you need to be able to point at a dated, written record of the architect confirming the change. Not a screenshot of a text thread. A signed RFI response on architect letterhead, with a date, attached to the project file.
Fix:Every RFI is written. Every response is written. Verbal confirmations get followed up in writing ("Confirming our conversation today — you approved a 10-inch tread per RFI 047") and a written reply is filed. No exceptions, especially for the architect you have known for 15 years and especially when it is a small change. Small changes are where disputes start.
Schedule impact when fixed: Minimal on schedule — but on disputes, this is the single biggest source of GC losses at close-out, and the savings are routinely $5,000-$30,000 per project.
Putting the five together
A GC who fixes all five of these on an average 30-RFI project recovers 6-12 schedule days, eliminates the most common sources of close-out dispute, and shifts the working relationship with their design team from reactive to proactive. The cost is roughly three minutes per RFI of additional authoring discipline.
Three minutes per RFI for 30 RFIs is 90 minutes of total authoring overhead. The return is on the order of $8,000 in recovered field labor and avoided rework on a typical project. No tool, just the checklist.
How Buildra fits
Buildra's RFI module enforces subject, priority, deadline, and assignee at authorship time — you literally cannot submit an RFI that is missing any of the five. Auto-generated sheet citations pull the relevant clip from your plan set into the RFI body. Every response is logged with a timestamp and attached to the project file for the audit trail. The discipline is built into the tool so it does not depend on your team remembering.
Rework, missed details, and undocumented changes silently destroy GC margins. The root cause is information debt — and five systematic fixes can stop it.
Undocumented decisions are dispute fuel. Categories, approvers, and what to record — plus how voice-to-decision drops the friction of capturing each one.
Track who, when, and why for every change. An anonymized story of a $40K dispute won with documented decisions — and how voice capture multiplies the effect.