Subcontractor prequalification is the GC's first line of defense against project risk — bonding capacity, insurance coverage, safety record, financial stability, references, and operational capacity all checked before the sub bids, not after they've been awarded. Done well, it filters out the firms most likely to default mid-project, deliver substandard work, or generate liability exposure. Done as a paperwork ritual, it produces a binder full of stale forms and zero actual screening. This is a working framework for mid-market GCs on what prequalification should actually evaluate, the recurring failure modes, and how AOS handles the lifecycle.
If you're the GC's preconstruction lead, contract admin, or operations manager and you've inherited a prequal process that everyone fills out but nobody really uses, you're not alone. Most mid-market GCs have prequal forms; very few have prequal workflows that actually inform award decisions. This post is on the difference.
What prequalification actually is
Subcontractor prequalification is the GC's process for assessing whether a sub has the technical, financial, and operational capacity to perform a specific scope of work on a specific project (or on the GC's projects generally). It typically happens once when the sub is onboarded into the GC's bidder pool, with annual refresh cycles, plus project-specific checks before sub award on larger or higher-risk projects.
The standard prequal evaluation covers seven categories:
- Bonding capacity. Can the sub bond the project? What's their single-job and aggregate bonding limit per their surety? What's their bonding history?
- Insurance coverage. Do they carry the required policy limits (CGL, WC, Auto, Umbrella, Professional, Pollution as applicable)? Do they have the endorsement capability the project requires? (Covered in depth in our COI compliance guide.)
- Safety record. What's their EMR (Experience Modification Rate)? OSHA recordables? Lost-time injury rate? Any willful violations or repeat citations?
- Financial stability. Annual revenue, working capital, debt structure, profitability trend. Audited financials or reviewed CPA-prepared statements.
- References. Prior project references from GCs and owners. Both successful and unsuccessful project history.
- Operational capacity. Active backlog, available staff for this project's timeline, equipment fleet for self-perform work, geographic coverage.
- Compliance and certifications. DBE / MBE / WBE certifications, prevailing-wage compliance history (covered in our prevailing-wage post), union affiliation, trade-specific certifications (asbestos, refrigerants, scaffolding, etc.).
That's a lot of data. Done as a one-time intake with annual refresh, it's a manageable workflow. Done as project-by-project re-collection from scratch, it's a nightmare for both the GC and the sub.
The four prequal philosophies (and what's wrong with each)
1. The paperwork ritual. The sub fills out a 30-page form once. The form goes in a binder. The binder sits on a shelf. When a project comes up, the GC checks if the binder exists, not what's in it. This is most common at smaller GCs and produces almost no actual risk screening.
2. The annual refresh. The sub updates the form yearly. The GC's contract admin tracks who's current vs. expired. There's a real list of prequalified subs. But the depth of evaluation is shallow — mostly checking that documents exist, not that the numbers tell a useful story.
3. The project-specific check. Before each bid invitation, the GC re-verifies the sub's prequal status, specifically for the project's size, scope, and risk profile. A sub prequalified for $2M jobs gets a different review when bidding a $20M scope. This is what most enterprise GCs do; it's labor-intensive but actually screens.
4. The third-party prequal. The GC uses a service like ISN, Avetta, or Triax to outsource prequal collection. Subs pay to participate. Coverage is broad but depth varies; the GC still has to do project-specific overlay for higher-risk awards.
The right answer for most mid-market GCs is a hybrid: third-party prequal for general bidder-pool maintenance, project-specific overlay for higher-risk awards, with the GC maintaining one structured record per sub that's actively used in award decisions — not a binder on a shelf.
What good prequal evaluation actually looks like, per category
Bonding. Single-job limit and aggregate limit per the surety's letter (not the sub's claim). Check the surety's rating (A.M. Best A- or better is the typical bar). Check whether the sub has bonded jobs comparable to your project size. A sub with $5M single-job capacity isn't appropriate for your $8M scope.
Insurance. Limits per your contract requirements. Endorsement capability for AI / Waiver / Primary & Non-Contributory. Carrier rating (A.M. Best A- or better). Loss history if available. (See COI compliance guide for the endorsement and tracking details.)
Safety. EMR below 1.0 is the typical bar (industry average is 1.0; below indicates a better-than-average safety record). OSHA Form 300 logs for the last 3 years. Willful violation history. For project-specific awards, also check whether the sub's safety program matches your jobsite requirements (PPE specs, lockout-tagout, fall protection per the project's risk level).
Financial. Audited or CPA-reviewed financials (audited is stronger than reviewed; both are stronger than internal). Working-capital ratio above 1.0. Debt-to-equity within industry norms. Revenue trend (declining revenue is a yellow flag worth investigating; not a disqualifier on its own). Most useful financial check: does the sub's revenue support taking on your project without overextending? A $20M revenue sub bidding a $4M scope is fine; a $5M revenue sub bidding the same scope is a concentration risk.
References. Three to five recent project references, with at least one from a GC similar to yours in size and scope mix. Talk to the references; don't just collect names. The most useful question to ask: "Would you hire them again, and if so, for what kind of scope?" The answer tells you what the references actually think.
Operational capacity. Current backlog (a sub with $5M backlog has different available capacity than one with $25M). Staff allocation for your project's timeline. For self-perform work, equipment fleet appropriate to the scope. Geographic logistics if the project is more than 90 minutes from the sub's yard.
Certifications and compliance. Specific to the project's requirements — DBE participation goals, union signatory status, prevailing-wage compliance history, trade certifications. Most projects don't require all of these, but missing the ones that ARE required is an immediate disqualifier.
The recurring prequal failure modes
1. Annual refresh that nobody actually reviews. The sub updates their forms each January. The GC's contract admin marks them current. Nobody actually compares this year's EMR to last year's, or notices that working capital dropped 40%. The data is collected but not evaluated.
2. Project-specific prequal skipped on "trusted" subs. The sub has worked with the GC for ten years. Annual prequal is current. The GC awards a $30M scope to a sub whose bonding capacity is $20M. Mid-project, the sub overextends and starts having cash-flow problems. The award decision didn't actually check fit-to-scope.
3. Reference checks reduced to "do you have any?" The sub provides three references. The GC's contract admin collects the names but doesn't call. The references are real but stale, or strategically chosen to avoid the unsuccessful jobs.
4. Financial review delegated to whoever can read a balance sheet. Most mid-market GC contract admins aren't financial analysts. The sub's audited financials get filed without actual analysis. Working capital, debt ratios, profitability trends — the analytic capability isn't there.
5. Safety record taken at face value. The sub reports EMR 0.85. The GC accepts. The sub's actual EMR is 1.4 (the 0.85 was a calculation error or a misrepresentation). Workers' Comp data is verifiable through the sub's insurance carrier — most GCs don't verify.
6. Bonding capacity confused with bonding willingness. The sub's surety has authorized $20M aggregate. The sub has $15M bonded on other jobs. The sub bids your $8M project. They can't actually bond it — available capacity is $5M — but the prequal form doesn't ask "how much do you have available right now."
7. Prequal data not refreshed when something material changes. The sub's principal departs mid-year. The sub's surety relationship changes. The sub takes on a problematic project that affects financial stability. None of this surfaces until next January's refresh, by which point the GC has awarded multiple jobs to a sub whose risk profile shifted six months ago.
What good prequal workflow looks like
The principles:
One structured record per sub. Every data point that prequal evaluates is a structured field, not a free-text PDF. Bonding limits are numbers; EMR is a number; insurance limits are numbers. Comparison across subs, comparison across years, and threshold-based filtering all become possible.
Tiered evaluation depth. Lightweight annual refresh for all subs; deeper project-specific evaluation for higher-risk awards; deepest evaluation (third-party reference calls, financial analyst review) for the largest awards.
Change-event triggers. If a sub's EMR jumps, their financials deteriorate materially, their key principals depart, or their bonding capacity drops — the GC's prequal system should surface it. Annual refresh isn't enough; event-driven refresh is.
Reference-call workflow. Structured questions, recorded answers, distributed to the bid-decision team. Not "did the contract admin call them?" but "what did the GC reference say specifically about the sub's communication, schedule reliability, change-order behavior, and closeout cooperation?"
Financial analysis with thresholds. Working-capital ratio, debt-to-equity, revenue trend, project-concentration risk — computed with thresholds flagged. A sub whose working capital dropped to 0.8 from 1.4 last year gets a yellow flag automatically.
Integration with award decision. The prequal status (and any flags) surface at the moment of bid-leveling and award. The PM making the award sees "this sub has yellow flag on financial; below your bonding-capacity threshold for this scope" before signing the subcontract.
How AOS handles subcontractor prequalification
In AOS, prequal is a structured per-sub record that drives the award workflow. Specifically:
- One structured prequal record per sub, with all data points as fields (not PDFs). Bonding limits, insurance limits, EMR, financial ratios, reference scores, certifications, equipment fleet, geographic coverage.
- Annual refresh workflow with automated reminders to the sub. Sub updates their data once; flows to every active GC client on AOS (cross-tenant), reducing the sub's prequal-update burden across the GC network.
- Change-event triggers. EMR change above threshold, insurance lapse, financial deterioration beyond configured limits — all surface as flags to the GC's preconstruction team.
- Project-specific overlay. Before bid invitation on higher-risk projects, the GC PM can require deeper-tier evaluation: third-party reference calls, financial-analyst review, project-specific safety-program review.
- Reference-call workflow. Structured questions; recorded answers; distributed to the bid-decision team; permanently attached to the sub's prequal record.
- Financial analysis with thresholds. Working-capital ratio, debt-to-equity, revenue trend, project-concentration risk — computed from uploaded financials with configurable threshold flags.
- Integration with the award workflow. At bid-leveling and award, the PM sees the sub's prequal status with any active flags. Award against a flagged sub requires acknowledgment.
- Cross-tenant prequal sharing. When subs are on AOS, their prequal data flows to all active GC clients with one update on the sub's side — reducing the prequal-update burden that today eats so much sub office time.
- Audit trail. Every change to a sub's prequal record is timestamped and attributable. If a sub default ever generates litigation, the GC can produce the full prequal history at any moment in time.
For the connected COI compliance workflow, see the COI compliance guide. For why GCs should care about their subs' tech stack more broadly (and why connected prequal is part of that), see why GCs should care about their subs' tech stack. The GC COO role page and GC PM role page walk through the workflow from each seat.
The prequal readiness checklist
- Is your prequal data structured (fields you can sort, compare, threshold), or is it PDFs in folders?
- Do you have tiered evaluation depth, with deeper review for higher-risk awards?
- Do change events (EMR change, financial deterioration, principal departure) trigger prequal refresh, or only the calendar?
- Are reference checks producing structured answers that the award team actually sees?
- Is financial analysis happening with thresholds, or just being filed?
- At the moment of bid-leveling and award, does the PM see the sub's prequal status with any active flags?
- If a sub default generates litigation, can you produce the full prequal history at any point in time?
If most answers are "in PDFs and binders, with workarounds," that's the workflow a connected platform replaces.
If you'd like to see it
We're in private-beta design-partner mode for mid-market subs and GCs. If you'd like to walk through how AOS handles subcontractor prequalification on your real bidder pool — particularly if you're tightening prequal in response to an actual loss or close call — apply to the beta. The commercial GC overview covers the platform broadly.