How to Vet Subcontractors Before They Cost You a Job

Sound Familiar?

Your subcontractor vetting process has gaps if:

• You've handed a new sub a significant scope based primarily on a referral from someone you trust, without verifying insurance, calling references, or discussing payment terms
• You've had a subcontractor show up without current general liability or workers' comp coverage, and you found out when you asked for the certificate mid-project
• A sub started work and their interpretation of the scope was materially different from yours — and neither of you had written scope language to resolve the disagreement
• You've absorbed back-charges or repair costs from a sub's faulty work because your agreement didn't define warranty obligations or workmanship standards
• You've had a sub disappear mid-project — not returning calls, missing scheduled days — after being paid for work not yet completed
• You've avoided firing a clearly underperforming sub because you weren't sure how to do it cleanly without blowing up the project timeline

What We Found

The 5-Point Vetting Framework: What to Check Before You Commit

Most builders vet subcontractors the same way they vet materials suppliers: they check price and availability, and if someone comes through a referral, they assume everything else is fine. That assumption is the source of most sub-related problems I see in the $1.5M–$15M range. The referral tells you the sub has done at least one acceptable job in one specific context. It tells you almost nothing about their operational profile — how they document, how they handle problems, what their cash flow situation looks like, and what happens when something goes wrong on your job.

The five-point framework gives you systematic visibility into each of these dimensions before you sign anything.

Point 1: Insurance — Verify the Certificate, Not the Conversation

Every subcontractor working on your jobs needs current general liability insurance and workers' compensation coverage. This is not optional and not negotiable — and a sub who tells you they're covered without being able to produce a certificate within 24 hours is telling you something important about how they operate.

The verification process: request a Certificate of Insurance (COI) naming your business as an additional insured. Review it personally or have your insurance agent review it. Check three things: the coverage limits (minimum $1M per occurrence general liability in most markets), the policy expiration date (certificates issued in January on a February policy expiration mean you'll be uninsured for most of the project), and whether workers' comp is actually listed (not all states require it for sole proprietors, but you need to know the exposure either way).

Do not accept a COI that hasn't been issued directly from the insurance carrier or broker. Do not assume a sub's coverage has been renewed just because it was valid last year. Check it before every new project engagement — policy lapses happen, and the sub may not tell you.

The liability you're protecting against isn't theoretical. A sub's employee gets hurt on your site without workers' comp coverage — that injury claim routes to your policy. A sub causes property damage during their scope — without adequate general liability on their end, that claim comes to you. Insurance verification is the fastest risk check you can run, and the failure cost of skipping it is enormous relative to the 10 minutes it takes to do it right.

Point 2: References — Ask the Questions That Actually Reveal Something

Most builders either skip reference calls entirely or run them as a formality — a five-minute conversation that confirms the sub exists and did work. References are only useful if you ask questions that can produce a negative answer.

The four reference questions that matter: (1) "Did they complete the scope as written, or were there significant deviations that required negotiation?" — this surfaces scope creep and communication style. (2) "How did they handle problems that came up mid-project?" — every project has problems; the answer reveals whether the sub is a partner or a problem-generator under pressure. (3) "Did they maintain a clean worksite and coordinate respectfully with other trades?" — bad sub-to-sub dynamics create schedule compression and quality problems across the whole project. (4) "Would you use them again on a project of similar complexity?" — this is the real signal. A reference who hesitates, qualifies, or suddenly remembers they're busy is telling you something without saying it directly.

Call three references minimum. If a sub can't provide three, ask why. If all three are from the same GC or the same project type, probe for references from comparable work to what you're hiring them for. A mechanical sub who has good references on light commercial work may have a completely different operational profile on a custom residential project with a demanding owner.

Point 3: Scope Clarity — Written Before the First Nail

The most expensive sub-related disputes I've seen weren't about quality or timeline — they were about what the scope actually included. The sub thought they were doing rough-in only. The GC expected a finished installation. Neither of them had anything in writing that resolved the disagreement clearly, so the dispute became a negotiation under pressure, mid-project, with a client watching.

Before any sub starts work, produce a written scope of work that describes: exactly what they're providing (labor, materials, or both), what's explicitly excluded, what adjacent work is expected to be complete before they start, and what condition they're expected to leave the work area in. If you use a PM platform like JobTread or Buildertrend, generate the scope from your estimate line items — don't reconstruct it manually for each sub.

A detailed scope document protects both parties. It gives the sub clear parameters so they can price accurately. It gives you a documented basis for back-charges if work isn't completed as specified. And it eliminates the "I didn't know that was included" conversation at closeout — which is where real margin leaks happen on sub relationships.

Point 4: Payment Terms — Align Before You're Dependent

Sub payment terms affect your cash flow directly and the sub's performance indirectly. A sub in a cash flow bind will slow down on your job to take on more work elsewhere. A sub who expects to be paid faster than your draw schedule allows will create friction on every payment cycle.

Set payment terms in writing before the first invoice: pay schedule tied to completion milestones (not calendar dates), retainage percentage if you hold it, and what documentation is required to release payment (invoice, lien waiver, completion photos). The lien waiver requirement is particularly important — conditional lien waivers on each payment and an unconditional final lien waiver at project close protect you from mechanics' liens even when a sub has paid their own suppliers.

Builders who skip the payment terms conversation until the first invoice creates friction, then negotiate under pressure when the sub is already on-site and partially through the scope. That's the worst possible negotiating position. Front-load the conversation and document what you agreed to.

Point 5: Warranty — Define It Before You Need It

Workmanship warranties from subcontractors are often either absent from subcontract agreements entirely, or present in language so vague it's unenforceable. "All work guaranteed for one year" tells you almost nothing: one year from what date, covering what defects, with what response time, and at whose cost?

Minimum warranty language for subcontract agreements: one-year minimum warranty on workmanship from substantial completion date (not project start), defects to be corrected within 30 days of written notice at no additional charge to the GC, and materials warranty passed through from manufacturer where applicable. For mechanical, electrical, and plumbing scopes, require the sub to register any manufacturer warranties in the homeowner's name at completion.

If a sub resists written warranty terms, that tells you something about how confident they are in their own work quality. A good sub has no problem putting warranty terms in writing because they don't expect callbacks. Resistance to this conversation is a yellow flag worth taking seriously.

How to Structure a Trial Project — De-Risk Before You Commit

Even if a subcontractor clears all five vetting points cleanly, that doesn't mean you hand them a $120,000 mechanical scope on your highest-visibility project. Vetting tells you they meet baseline requirements. It doesn't tell you how they perform under pressure, how they communicate when something goes wrong, or how their field crew behaves on your specific type of work.

The trial project structure answers those questions without betting a major project on an unknown quantity.

What Makes a Good Trial Scope

A trial project should be real work — not artificially small or low-stakes, but not the full depth of what you might eventually give the sub. Ideal trial scopes are: self-contained (a discrete phase of a project that can be evaluated on its own completion), meaningful enough that the sub's best people show up (not the crew they deploy when they don't care), and representative of the work type you'll use them for regularly.

Examples that work well: a drywall scope on a bathroom addition before you put them on a full custom home; a finish carpentry scope on a single room before you commit a whole-house trim package; a mechanical rough-in on a smaller remodel before a new construction project with a demanding timeline.

What to Evaluate on the Trial

You're watching five things on a trial project: (1) Schedule discipline — do they show up when they said they would, do they hit their milestone dates? (2) Communication quality — when issues come up, do they surface them to you proactively, or do you find out from the client? (3) Site behavior — do they maintain a clean workspace, coordinate professionally with other trades, and handle the client interactions they're exposed to appropriately? (4) Scope adherence — did they complete exactly what was in the written scope, or were there shortcuts, omissions, or "I didn't know that was included" conversations? (5) Invoice accuracy — was the invoice correct, was the documentation complete, and did they provide the lien waiver without being asked?

A sub who clears all five on a trial project has earned a larger scope. A sub who misses one or two items on a trial project is showing you their operational baseline — and the question is whether you want to bet bigger on that baseline or not.

The Conversation After the Trial

After the trial scope closes, have a direct debrief with the sub. Tell them specifically what went well and specifically what didn't. A sub who responds to direct feedback professionally — acknowledges the issue, explains what they'd do differently, asks what they need to change — is worth investing in. A sub who becomes defensive, makes excuses, or dismisses the feedback is telling you exactly how they'll behave when you give them a bigger scope and something goes sideways.

This conversation is uncomfortable if you're not used to having it. Have it anyway. The cost of not having it is considerably higher than the 20 minutes of awkwardness it requires.

When to Fire a Sub — and How to Do It Without Destroying the Project

The most common sub management mistake I see isn't bad vetting. It's keeping a bad sub too long because the project is already underway and the replacement math feels impossible. Builders carry subs who are clearly underperforming for weeks — sometimes for entire projects — because the alternative feels worse than the problem. It almost never is.

The Non-Negotiable Triggers

Some situations require immediate action regardless of where the project stands:

• Insurance lapses: A sub whose coverage expires mid-project is not operating on your job site another day until the certificate is renewed. This is not a negotiation. The liability exposure is immediate and your own insurer will tell you the same thing.
• Abandonment pattern: If a sub has missed three consecutive scheduled days without contact, they have effectively abandoned the scope. Document the absences in writing, send written notice of default, and begin identifying a replacement. Do not wait for a fourth missed day.
• Safety violations: A sub who repeatedly violates site safety requirements — no hard hats, improper scaffolding, OSHA violations — gets one documented written warning. A second violation ends the relationship. The liability on a job site injury flows to you as the GC regardless of whose employee it was.
• Repeated material defects: If the same sub has produced defective work that required rework on two separate occasions — and the second occurrence happened after a documented conversation about the first — they've demonstrated their quality ceiling. The cost of a third occurrence is higher than the cost of replacement.

The Early Warning Signs That Aren't Immediate Triggers (But Should Be Acted On)

Not every problem requires termination. But some patterns, if ignored, always escalate: chronically late invoices with incorrect documentation (signals cash flow and operational problems); increasing reliance on the GC to manage their own material procurement (the sub's bandwidth is stretched too thin); slow or defensive responses to quality concerns (predicts how warranty callbacks will go); and staff turnover on your specific project mid-scope (you're getting their B crew).

When you see these patterns, have the direct conversation with the sub's principal — not the project foreman — and document it. "We've noticed X and Y on this project. What's the plan to correct it, and by when?" Get a commitment in writing. If the commitment isn't met within two weeks, make the replacement decision. The patterns don't resolve themselves, and every week you wait costs you more.

How to Execute the Replacement Without Destroying the Schedule

Firing a sub mid-project is operationally hard, but it's manageable if you follow a specific sequence: (1) document the performance issues and any prior written notice you've given — this protects you from a claim that the termination was arbitrary; (2) identify a replacement candidate before you pull the trigger — having a replacement lined up in advance is what makes it executable instead of chaotic; (3) conduct a scope audit before the new sub starts — document exactly what's complete, what's incomplete, and what requires rework, so the new sub's starting point is unambiguous; (4) notify the terminated sub in writing with the scope of the deficiency and any back-charges you intend to apply.

The builder who has done this successfully once does it much better the second time. The first time is hard because it feels like failure. It's not. Holding onto a failing sub relationship because the exit is uncomfortable is what actually costs you — in schedule, in quality, in client relationship, and in your own time managing a situation that should have been resolved weeks earlier.

If your current sub vetting process is informal or inconsistent, the free checklist gives you a structured starting point — five criteria, in sequence, before any new sub gets a scope. The builders who use a systematic vetting process report fewer mid-project surprises, lower rework costs, and significantly less time spent managing sub relationships that should have been screened out before they started.

Frequently Asked Questions