Most GCs treat change order disputes as a field problem. They're not. By the time a sub is billing extras on site, the gap was already there — buried in a scope sheet written at 11 PM on bid day.
According to the 2024 Arcadis Global Construction Disputes Report, the average construction dispute value in North America reached $60.1M. Errors and omissions in contract documents have repeatedly ranked as the number-one dispute cause in North America. That's not a project execution problem. That's a pre-construction problem.
The firms with the tightest margins don't just manage disputes better. They prevent them earlier — in the scope sheet, before the subcontract is ever issued.
A scope gap is any work that's clearly shown on the drawings or called out in the specs but not explicitly assigned in the subcontract scope of work. When that gap hits the field, somebody pays for it.
Sometimes the GC absorbs it under "readily inferable" language. Sometimes the sub bills it as an extra. Either way, someone's margin takes the hit — and someone ends up in a dispute.
The examples are consistent across the industry. A $300K lead-lined glass package omitted from a hospital imaging suite scope was absorbed by the GC under a "readily inferable" clause. A $400K roof cover board on a $50M project went unscoped and was only recovered through a relationship concession from the sub. A $200K wood-flooring scope gap on a luxury condo project wasn't caught until buyout.
These aren't freak occurrences. They're what happens when the scope sheet is written from memory and boilerplate instead of from the drawings.
For a deeper look at where these gaps form trade by trade, see the trade-specific scope gaps chapter in Provision's Scope Gap Playbook — built from 200+ interviews with GC estimators and pre-construction leads.
Before you can fix the problem, you have to name it. These are the habits that show up most often in the scope sheets that end up in dispute.
This is the most-cited anti-pattern across GC pre-construction teams. It feels like coverage. It isn't. When a sub prices to "plans and specs" and you scope to "plans and specs," nobody owns the gap — and the dispute writes itself.
Pulling last year's hospital scope for this year's hospital feels efficient. It's not. Different drawings, different specs, different subs. The gaps from the last project don't carry forward — but the omissions in the boilerplate do.
A junior estimator under time pressure will default to whatever was in the last bid folder. That's not a competence problem — it's a process problem. If there's no drawings-first discipline built into the workflow, boilerplate wins.
Scope sheets reviewed at the last minute get a gut-check, not a real review. Gaps that would have been caught with an hour of structured review get missed. Then they show up in the field.
As one Estimating Manager at a Canadian ICI GC put it: "We have less subs who just kind of a gentleman's agreement… they've become more quick to clarify that we're not including that one piece of scope." The environment has changed. Subs are more aggressive about exclusions. GC scope sheets need to match that precision.
The firms that consistently hold margin in preconstruction share a small set of habits. These aren't new concepts — but they're applied consistently, not just when there's extra time.
Pull up the drawings first. Walk every sheet relevant to the trade before you touch a scope template. The scope sheet should describe what's on the drawings — not the other way around. Boilerplate is a floor, not a ceiling.
Don't write "as per contract documents." Write "Structural drawings S2.1 through S4.3, Specification Section 03300." Specific references create accountability. Generic incorporation creates gaps.
When a pre-construction lead at a Top-ENR Canadian GC described the standard of specificity required, they used this framing: "It's descriptive — bread, put it on a plate, use the open jar… You have to get to that level of detail or else they'll just be like, 'you didn't tell us that.'"
Don't wait until post-award to align on scope. The buyout conversation should start during the bid — before the scope sheet is final. Pre-bid alignment with key subs surfaces gaps when there's still time to price them.
A scope template is a starting point. Every project has unique conditions — site constraints, trade overlaps, spec requirements that don't exist in your standard template. Add those items explicitly. Don't assume the template covers it.
The pre-bid walk isn't a courtesy — it's a scoping exercise. Walk it with a markup set and flag anything the drawings don't show clearly. Existing conditions, access issues, and trade interface points all belong in the scope sheet.
A national mechanical contractor reads scope differently than a regional specialty sub. The more sophisticated the sub, the more likely they'll hold you to exact language. Tailor your scope detail to match. Don't send a junior sub the same tight scope you'd send a major mechanical contractor — and vice versa.
Exclusion lists say what you're not doing. Clarification lists say what you are doing — and at what standard. Both are necessary. Exclusions without clarifications leave grey area that turns into change orders.
Before the subcontract goes out, a second set of eyes reviews the scope sheet against the drawings. Not a full re-estimate — a structured check for gaps, contradictions, and missing trade assignments. This checkpoint catches what the original author missed under time pressure.
For the full framework behind these habits, including how they apply to specific trade packages, read the subcontract language and scope chapter of the Scope Gap Playbook.
Here's the real constraint. On a mid-size ICI project, a GC's pre-construction team is working through a 2,000-page project manual, 300+ drawing sheets, multiple addenda, and several subcontract templates — simultaneously, across multiple pursuits.
A 2018 study by PlanGrid and FMI found that miscommunication and bad project data cost the U.S. construction industry an estimated $31.3 billion in rework. Communication breakdowns and bad project data together accounted for 48% of all rework. The document volume problem is a direct driver of that number.
As a Director of Pre-Construction at a Mid-Market Southeast GC described it: "Pre-con is working in the scope sheet world and project management is working in the scopes of work." The handoff between those two worlds — between what was priced and what was contracted — is where disputes are born.
This is the gap that makes consistent execution of the eight habits so difficult. The habits require time and attention. The bid calendar doesn't give you either.
Preventing change order disputes doesn't require AI. It requires discipline and process. But the document volume problem is real, and it gets worse as bid count increases.
That's where purpose-built tools add value — not by replacing the estimator's judgment, but by doing the document-reading work faster and more completely than a manual review under time pressure.
Provision's Scope Agent reads the full project set — drawings, specs, and contracts together — and generates a complete scope-of-work package in under 60 minutes. It extracts scope items at 97% match accuracy (validated against human estimator benchmarks, where the human baseline on the same exercise was 91.3%). That's not a marketing claim — it's an internal validation against a dataset of real GC bids.
For GCs managing high bid volume, Scope Agent addresses the core constraint: there aren't enough hours in the bid calendar to do a drawings-first review on every trade for every pursuit. The tool does the extraction. The estimator reviews, adjusts, and applies judgment. That's the division of labor that makes consistent habits executable.
Provision has processed over 100,000 construction documents and reviewed $100 billion in project value. Estimators using it report getting through pursuits 2x faster without reducing scope quality.
For contract risk identification — the "readily inferable" language, liquidated damages clauses, and indemnity provisions that set the stage for disputes — Risk Review flags every risk cited to the exact clause, section, and page. Teams using it report an 80% reduction in contract and spec review time.
If your team needs to search across drawings, specs, addenda, and RFIs without reading every page, Chat Agent returns cited answers in under 20 seconds — pinpointing the exact document, section, and page number for each response.
Preventing change order disputes requires one structural fix above all others: close the gap between the scope sheet and the subcontract.
What was priced in the bid and what's in the contract have to match. When they don't, the field team is left negotiating scope that should have been settled at the desk. As one Senior PM at a Toronto mid-market developer put it: "If we could catch three scope gaps or three missed items on every scope of work, then this thing pays for itself."
That's the standard. Not perfection. Three items per scope of work, caught before bid day instead of discovered on site.
Consistent habits, enforced by process — and where needed, supported by tools built for the volume — is how the best preconstruction teams get there.
See the full framework in the Scope Gap Playbook, including trade-specific gap patterns, operator-cited examples, and the contract language that creates or closes disputes.
Errors and omissions in contract documents have repeatedly ranked as the top cause of construction disputes in North America, according to the Arcadis Global Construction Disputes Report. In practice, this means scope gaps — work shown on drawings but not explicitly assigned in the subcontract scope of work.
A scope gap creates ambiguity about who owns a piece of work. When that ambiguity hits the field, the sub bills it as an extra. The GC disputes it. Without a clear paper trail back to the scope sheet and drawings, the GC has limited leverage. The dispute is often resolved at the GC's expense — or in arbitration.
"Readily inferable" language requires a contractor to perform work that a reasonable professional would infer from the contract documents — even if it's not explicitly listed. GCs often use this as a backstop. But as the $300K lead-lined glass example shows, courts and arbitrators don't always agree on what's "inferable." Explicit scope is always safer than implied scope.
Prevention starts at scope sheet creation — before the subcontract is issued, and ideally before the bid goes out. The pre-issue scope review checkpoint (checking the scope sheet against drawings before sending) is the single highest-leverage step. Catching a gap at that stage costs nothing. Catching it in the field can cost hundreds of thousands of dollars.
Purpose-built AI tools like Provision's Scope Agent read the full project set — drawings, specs, and contracts — and extract scope items across all trades. This reduces the manual document-reading time that creates gaps under bid-day pressure. The tool generates a scope-of-work package in under 60 minutes at 97% match accuracy, giving estimators a structured starting point for review rather than a blank page or recycled boilerplate.
An exclusion states what a trade does not include. A clarification states what it does include — and at what standard. Exclusions are necessary but insufficient on their own. Without matching clarifications, the grey area between "not included" and "someone else's work" creates exactly the disputes GCs are trying to prevent. Both belong in every scope of work.
The core habit is drawings-first scoping — but that requires time, which is the constraint on high-volume bid calendars. The firms that manage it consistently use a combination of structured scope templates (as a floor, not a ceiling), pre-issue review checkpoints, and tools that accelerate document reading. Scope Agent is built specifically for this constraint — it handles the extraction work so estimators can focus on review and judgment rather than reading every sheet manually.
Scope Agent reads your full project set and builds a complete scope package in under 60 minutes.
See Scope AgentMore Articles
