On a cost-plus or GMP project, an estimating error is a problem. On a lump-sum contract, it's a loss. The price is fixed. If your scope is wrong, you eat the difference.
That's the deal. And in 2026, with bid timelines compressed and document sets ballooning, the conditions that produce estimating errors are getting worse — not better.
For six of the last nine years, "errors and omissions in contract documents" has ranked as the top cause of construction disputes globally, according to the Arcadis 2025 Global Construction Disputes Report. The average U.S. dispute value hit $60.1M. Most of that starts with a scope gap nobody caught before bid day.
This article breaks down where estimating errors come from, what they actually cost, and what the best pre-con teams do differently.
The numbers are concrete. FMI's Construction Disconnected report puts annual U.S. rework costs from miscommunication and bad project data at $31 billion. Twenty-six percent of that traces back to communication breakdowns. Another 22% comes from bad project data at bid time.
At the project level, change orders on commercial work average 8–14% of total project cost, according to Navigant research republished by the AIA. On projects with weak scope definition, that number climbs above 25%.
Those aren't abstract percentages. Here's what they look like in practice:
These aren't edge cases. They're the kind of gaps that show up in Provision's Scope Gap Playbook, drawn from interviews with 200+ general contractors across North America. The pattern is consistent: the error was knowable before bid day. It just wasn't caught.
This is the single most-cited anti-pattern in GC pre-construction. Writing "scope of work: as per plans and specifications" doesn't define scope. It defers the argument.
When a dispute comes, that language doesn't protect you. Owners and subs both read it in their favor. The GC gets squeezed in the middle.
Specific document references — sheet numbers, spec sections, revision dates — are what protect your position. Generic incorporation language is not a scope definition. It's a placeholder that invites conflict.
Every project has a different document set. Pulling last quarter's concrete scope for this quarter's job works until it doesn't. When it fails, it fails expensively.
The estimating teams that avoid this treat templates as a floor, not a ceiling. They start with a standard checklist and build upward from the actual drawings — not downward from a boilerplate.
Spec-first estimating misses conflicts. The spec says one thing. The drawing says another. The addendum changes both. If your estimator reads in the wrong order, they're scoping to a document set that no longer exists.
Drawings-first is Habit 1 in the Eight Habits framework from the Scope Gap Playbook's chapter on trade-specific gaps. It forces estimators to reconcile what's actually shown before they price what's written.
Pre-bid scope review — when it happens at all — often happens in the last hour before submission. By then, the number is set. Nobody's catching a $45K stone-depth conflict at that point. They're just checking math.
A pre-issue scope review checkpoint — before the scope goes to the sub for pricing — is where the real catches happen. That's Habit 8. It requires time in the schedule, which means it has to be planned before the bid starts, not added at the end.
Sub relationships matter. But the dynamic has shifted. 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."
Subs are protecting their margins. So should you. Scope clarity at buyout isn't bureaucracy — it's the mechanism that keeps a $400K gap from becoming a field dispute six months into the job.
Listing exclusions tells the sub what you're not paying for. But it doesn't tell them what you are paying for. That's the difference between an exclusion and a clarification.
A Pre-Construction Lead at a Top-ENR Canadian GC described what proper scope clarity looks like:
"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.'"
That level of specificity is hard to achieve manually when you're managing 10 trades across a 2,000-page document set. But it's the only standard that holds up when money is on the line.
Some scope gaps are trade-specific and recurring. Knowing where they cluster helps estimating teams focus their review time. The following are the most common, based on GC interviews across 200+ firms.
For a full breakdown by trade, see Chapter 5 of the Scope Gap Playbook.
Most standard contracts include language that makes the GC responsible for work that is "reasonably inferable" from the contract documents — even if it isn't explicitly shown or specified.
That's a wide net. And owners and their lawyers know how to use it.
One Senior PM at a Canadian ICI GC described how this expectation has expanded:
"Our construction management clients expect us to find the scope gaps in the design too now. They expect us to be designers and engineers."
The practical answer is document cross-referencing at bid time. Every trade scope needs to be checked against the drawings, the spec, and the addenda — not just the last bid that looked similar. That's where most "readily inferable" disputes originate: a conflict that was visible in the documents but not caught during estimating.
The Scope Gap Playbook's chapter on subcontract language covers how leading GCs write around this language — and how they use clarification clauses to limit exposure before the contract is signed.
The firms with the tightest margins on lump-sum work share a consistent set of habits. These aren't revolutionary processes. They're disciplined execution of fundamentals that most teams know but few enforce consistently.
These are the Eight Habits from the Scope Gap Playbook. They're not a checklist to review once. They're a process to enforce on every pursuit.
Manual review is the baseline. AI tools built for pre-construction are how leading GC teams scale it.
The gap isn't effort. Most estimating teams work hard. The gap is coverage — 2,000-page document sets reviewed in 30 hours, by one or two people, under bid-day pressure. That's where things get missed.
Scope Agent generates complete scope-of-work packages from construction documents in under 60 minutes. It replaces 30–40 hours of manual extraction per bid. That's not a speed improvement — it's a capacity improvement. Teams that were turning away pursuits because of bandwidth can now compete on more work with the same headcount.
Risk Review runs a structured risk checklist against your project documents with 99.5% accuracy on pre-built checklists. It surfaces the clauses that shift risk to the GC — "readily inferable" language, owner-favorable indemnity provisions, missing scope callouts — before you're locked into a fixed price.
Chat Agent lets your estimators search a full project set — drawings, specs, contracts, addenda — and get cited answers in under 20 seconds. When a sub calls with a question at 4:45 PM on bid day, your team can answer it from the actual document, not from memory.
Provision has reviewed over $100 billion in project value and processed more than 66,000 documents. The accuracy benchmark matters: 95% verified accuracy across real project documents. That's the number that earns trust with a Chief Estimator who's been burned by generic AI tools that hallucinate spec sections.
To see how a mid-market GC team uses these tools in a real pre-construction workflow, read the Cleveland Construction case study.
One Senior PM at a Toronto mid-market developer put it plainly:
"If we could catch three scope gaps or three missed items on every scope of work, then this thing pays for itself."
That math is easy to run. Three gaps at an average of $100K–$300K each is $300K–$900K per project recovered. On lump-sum work, that's the difference between a profitable job and one you regret winning.
The tools exist. The habits are known. The firms protecting margin in 2026 are the ones enforcing both — consistently, on every pursuit, not just the big ones.
If you want to see how Provision fits into a GC pre-construction workflow, or explore the Scope Agent in more detail, the demo takes 30 minutes and uses your actual document set.
An estimating error on a lump-sum contract is any gap between the scope you priced and the scope you're required to deliver. Because the contract price is fixed, the GC absorbs the cost of the error — not the owner. Common errors include missed trade scopes, drawing-spec conflicts, and "readily inferable" work that wasn't included in the estimate.
On a GMP contract, cost overruns can sometimes be absorbed through contingency or reconciled with the owner. On a lump-sum job, the price is fixed at award. Any scope gap, missed trade, or omitted item comes directly out of the GC's margin. There's no mechanism to recover the cost unless the owner issued a formal change directive.
The most common errors are: using "as per plans and specs" instead of specific document references, copying scope from a previous similar job, missing trade-specific gaps in MEP and envelope work, reviewing scope five minutes before bid submission, and relying on sub relationships rather than written scope clarity at buyout.
Individual scope gaps range widely. Real examples from GC interviews include a $45K stone-depth mismatch, a $200K wood-flooring gap on a luxury condo, a $300K lead-lined glass omission on a hospital project, and a $400K missed roof cover board on a $50M build. The FMI Construction Disconnected report puts total annual U.S. rework costs at $31 billion.
Most standard construction contracts require the GC to complete work that is "reasonably inferable" from the contract documents, even if it's not explicitly shown. This means scope gaps visible in the drawings — but not priced — can become the GC's obligation. Catching these conflicts during estimating, before award, is the only reliable way to manage the exposure.
Yes, but only if the tool is built for construction pre-construction workflows. Generic AI tools like ChatGPT don't understand construction document structure and can't produce bid-ready scope packages. Purpose-built tools like Provision's Scope Agent generate complete scope packages from full project sets in under 60 minutes, and Risk Review surfaces contract clauses that shift risk to the GC — before you're locked into a fixed price.
They follow a consistent set of habits: reviewing drawings before specs, using specific document references in scope packages, front-loading sub buyout conversations, tailoring scopes to individual trade sophistication, and running a formal pre-issue review checkpoint before any scope goes out. These aren't new ideas — they're just enforced consistently on every pursuit, not only the large ones.
See how Scope Agent finds missing trades and spec conflicts before bid day.
See Scope AgentMore Articles
