The Subcontractor Prequalification Process for GC Estimators

A sub walks in low by $340,000, and the room gets excited. Then someone asks if they’re bonded for the full contract value. Silence. That’s a prequalification failure, and it happened weeks before bid day.

Prequalification isn’t a form you send once. It’s a process with decision points at every stage, from building your database to cutting that final PO. Here’s how to run it without gaps.

Build the database before you need it

The worst time to prequalify a sub is during the bid. You’re already short on time, the drawings are half-issued, and now you’re chasing a COI from a company you found three days ago. The prequalification process starts between pursuits, not during them.

Every sub in your database should have a baseline record: CSI MasterFormat trade codes, bonding capacity, insurance limits, license status, certifications (MBE, WBE, DBE), and a contact history. If you can’t pull a sub’s bonding limit in under 60 seconds, you don’t have a database. You have a contact list.

Set a review cadence. Prequalification data expires. A sub who was bondable at $2M two years ago may be in financial trouble today, or may have grown to $8M. Touch every active sub in your database at least annually, and put new subs through the full intake before they see their first invite.

Different jobs, different bars

Not every job has the same bar. A $900,000 interior renovation has different risk thresholds than a $22M ground-up. Before you send a single invitation, define the minimum qualifications for that specific project: minimum bonding capacity as a percentage of the trade scope, required insurance limits, any owner-mandated certifications, and relevant experience (same building type, same delivery method, similar complexity).

This matters because it turns prequalification into a filter, not a formality. A sub who passes baseline intake might still not qualify for a particular job. The project-level check catches that before bid day, not after you’ve already plugged their number in.

If the owner has prequalification requirements of their own, common on public work, federal projects, and institutional GMP jobs, get those in hand before you finalize your invite list. Finding out the owner requires SBA surety bonding compliance after you’ve already solicited a sub who doesn’t meet it wastes everyone’s time. That’s an avoidable mistake, and there’s no good excuse for making it twice.

The questionnaire goes out before the invite, not after

If a sub is new to your database and they’re going on an invite list, they earn a prequalification questionnaire before the invite. Not after. The questionnaire should cover: years in business, project volume in the last 24 months, current backlog versus capacity, bonding capacity and current surety relationship, references from comparable projects, and any claims, liens, or litigation in the past three years.

A sub with $4M in annual revenue and $6M in current backlog is already overcommitted. A sub who can’t name their surety on the spot probably doesn’t have an active bonding relationship. These aren’t gotcha questions. They’re the minimum information needed to make an informed decision about who gets access to your bid.

For mid-to-large GCs managing dozens of active trades, the questionnaire process doesn’t have to be slow. The goal is a decision, qualified, conditionally qualified, or not yet, before the invite goes out, not a full financial audit.

Bid behavior is qualification data too

Prequalification doesn’t end when you send the invite. It continues through the entire bid phase. Pay attention to how subs respond. Did they acknowledge the invite within 48 hours or four days? Did they ask scope questions that suggest they actually read the drawings? Did they submit with a complete scope narrative or a one-line number with a page of exclusions?

Response speed and bid quality are real data points about how that sub operates. As covered in how labor escalation is changing what goes into a GC bid, the margin for error on today’s bids is thin. A sub who is hard to reach during the bid will be hard to reach during submittals, RFIs, and schedule updates. That’s not an assumption. That’s pattern recognition.

Log all of it, acknowledgment date, questions asked, bid submission time, exclusions included. That record builds over time into something genuinely useful: a ranked understanding of which subs are reliable, which ones sandbag their bids, and which ones you can call at 1:45 on bid day and actually get an answer.

The low number isn’t the decision, it’s the starting point

Bid day produces a number. Prequalification produces the context that tells you whether to trust it. When the low bid comes in $280,000 under the next competitor, the prequalification record is what you reach for first. Is their bonding capacity sufficient for this scope? Have they built anything at this scale? What did their last two bids with you look like during execution?

A low number from a well-qualified sub is a different conversation than a low number from someone you’ve never worked with, who returned the questionnaire incomplete, and who submitted 90 seconds before your deadline. Both hit the same cell in the spreadsheet. Only one of them is a real bid. Treat them the same and you’ll eventually learn that lesson on a job that can’t afford it.

Comms Center tracks every stage of the subcontractor relationship, from initial invite through bid receipt, with a searchable log of every communication in between. Estimators get a live view of which subs have acknowledged, which have gone silent, and which have submitted, across every active pursuit simultaneously. Learn more at commscenter.com.