Every job has them. A footing hits an old foundation nobody knew was there. The owner walks the model home and decides the great room needs a taller ceiling. An RFI comes back with a detail that changes half the framing on the east elevation. Scope changes aren't a sign something went wrong — they're the normal weather of a construction project. What separates a job that absorbs them from one that bleeds out through them is whether you catch each change early, price it before you build it, and thread the new work into your schedule without knocking everything else off the rails.
That last part is where most teams lose the money. The change gets approved, the check clears, and then nobody updates the plan. The extra work shows up as a surprise on the wall three weeks later, colliding with trades that were already sequenced. Below is how a change actually moves through a well-run job, and where good scheduling and subcontractor software earns its keep — and, just as important, where it doesn't replace judgment.
Catch the change before the crew builds it
The most expensive change order is the one you discover after the work is in place. Once concrete is poured or drywall is hung, you've lost your leverage and you're paying to demo before you pay to rebuild. The whole game is catching the variation before a crew touches it.
Two habits do most of the work here. First, teach your foremen that "this doesn't match the plans" is a stop-and-flag moment, not a "we'll figure it out" moment. A framer who notices the header schedule doesn't match the opening should be able to snap a photo and raise a flag in thirty seconds — because if it takes ten minutes of paperwork, they'll just build it and move on, and you'll eat it. Low friction is the whole point.
Second, and this is the one people skip: review your look-ahead against the actual contract scope. When you're building the weekly work plan and something's on the wall that you can't tie back to a line in the contract or an approved change order, that's your early-warning system going off. Work that isn't in the contract shouldn't be scheduled to run until it's been processed as a change. On a look-ahead board this jumps out — an activity with no home in the original scope is a question you want to ask on Monday, not the following Thursday when the crew's already halfway done.
Document it while the evidence is still standing
Field conditions have a short shelf life. The rock you hit, the rotted subfloor, the utility that wasn't where the as-builts said — all of it gets covered up fast. Photograph it before you touch it, and photograph it with something for scale. A picture of a differing site condition, timestamped and geotagged, is worth more in a dispute than three pages of narrative written after the fact.
Build the habit of one clean change file per change: the request, the field photos, the relevant RFI or ASI, the pricing breakdown, and every email that touched it. Six months later when someone's arguing about whether the ceiling height change was ever authorized, you want to open one folder and have the whole story in order, not go digging through a chain of forwarded messages. The value of software here isn't magic — it's that it forces the discipline of keeping the trail attached to the work while memory is fresh.
Assess the schedule impact — not just the dollars
Here's the mistake that sinks otherwise-good change management: pricing the change and forgetting the time. Everybody remembers to ask "what does it cost?" Far fewer stop to ask "what does it do to my sequence?" A $4,000 change that adds a week to your critical path can cost you a lot more than $4,000 in extended general conditions, liquidated damages, and the cascade of trades that now start late.
Before you approve, run it through the schedule. Where does this work land, what does it push, and does it touch anything on the critical path? This is exactly what a look-ahead view is good for — you can see the change work dropped in next to the original scope and watch what it collides with. If the added electrical rough-in forces the insulation and drywall to slide, you want to know that before you commit, so you can either price the delay into the change or find a way to sequence around it.
A rule of thumb worth keeping: any change that touches a trade already on your three-week look-ahead deserves a hard look at the handoffs. It's rarely the change work itself that hurts you — it's the buffer it eats between trades. Frame-to-rough-in normally wants a day or two of slack for cleanup and inspection; a change that consumes that buffer turns a comfortable sequence into a stack-up where the plumber and the electrician are tripping over each other in the same wall on the same day.
Route approvals so nobody wonders where it stands
Two failure modes live in the approval step, and they're opposites. One is the bottleneck: every change, big or small, waits on one person's signature, and work stalls. The other is the free-for-all: changes get built on a verbal "yeah, go ahead" and there's no paper when it's time to get paid.
The fix is a tiered routing rule everyone understands. Small-dollar field changes might need only the super's sign-off; anything above a threshold routes to the PM; anything that touches the owner's budget or the completion date routes up the chain. The specific numbers matter less than the fact that they're written down and consistent. And the single most valuable thing an approval workflow gives you is visibility — anyone should be able to look and see exactly where a change sits: submitted, priced, in review, approved, or rejected. When status is visible, you stop losing an hour a day to "hey, did that change ever get approved?"
Tie your schedule to that status. An activity that can't start until a change is approved should show that dependency plainly, so a crew doesn't roll up to work that isn't cleared yet. The worst version of this is a subcontractor mobilizing a crew for extra work that never got a signed change order — now you're arguing about whether you owe them for a day of standby.
Integrate approved work into the plan — this is the step everyone skips
Approval is not the finish line. A change that's been signed but never folded into the schedule is a landmine. The moment a change is approved, the new activities need to go onto the wall: added tasks created, existing durations adjusted, and dependencies rewired so the trade flow reflects what's actually going to happen. If the change adds a bank of can lights, that's not just an electrical line item — it's a hold on the drywall in that area and possibly a second inspection.
Resource planning has to catch up too. Extra scope needs crews, and those crews were already committed somewhere. If you're adding two days of masonry, that mason is now not somewhere else — plan it against your real labor availability, not against wishful thinking. A schedule that treats change-order work as if it materializes out of thin air is how you end up over-committed and chasing your tail. When your weekly work plan carries the change work right alongside the original scope, the field sees one coherent plan instead of "the schedule, plus a pile of extras somebody mentioned."
Push the change down to the sub — and keep the two clear
Changes to your contract with the owner usually flow downhill to your agreements with the trades, and the two don't always move at the same speed. A common trap: the owner approves a change, you tell the sub to proceed, and the sub's change order is still sitting unsigned. Do not schedule a trade partner's extra work until their change order is executed. Unapproved work is unapproved work whether it's you or your sub carrying the risk.
Give the field a way to tell original scope from change work at a glance. When a foreman is filling out daily reports, knowing that today's activity is change-order work — and which change order — keeps your cost coding clean and your billing defensible. Blur that line and you'll spend the next pay period untangling which hours belong to base contract and which belong to extras.
Track cost and margin by change, not just in a lump
Keep original scope and change-order work in separate buckets from day one. Lumping them together hides the story your numbers are trying to tell you. Track approved changes against the original contract value as a running total, and you'll always know how much the job has grown — a project that's up 18% in scope is a different animal than the one you bid, and everyone from the owner to the bank should know that number.
Then watch margin by change. Some extras are far more profitable than others; some you priced tight to keep the owner happy and you're barely breaking even. Over a few jobs, the pattern tells you where you're leaving money on the table in your change pricing and where you're competitive. That's not accounting for accounting's sake — it's how you get sharper on the next estimate.
Don't let pending changes rot
Every open, unapproved change is exposure — work you may have to perform, or work you've already performed and haven't been paid for. Keep pending changes visible and, honestly, watch their age. A change that's been "under review" for forty-five days isn't under review; it's stuck, and it needs a phone call, not another week of aging quietly in a list. Old pending changes are where cash flow goes to die.
It's also worth modeling the "what if these don't get approved" case on anything material. If a chunk of pending work is holding up downstream trades, you need a plan for both outcomes rather than assuming it'll all clear.
Read the trend — your changes are trying to tell you something
One change is an event. A rising rate of changes is a diagnosis. When the change volume on a job keeps climbing, it usually means one of a few things: the design was incomplete when you started, the scope was written loosely, or the field conditions were badly understood going in. Sort your changes by source — owner request, design error or omission, differing field condition, coordination miss — and the pattern points you at the real problem.
Look across several jobs and the systemic issues surface. If design-omission changes keep hitting you with the same architect, that's a preconstruction conversation. If field-condition changes cluster on renovation work, maybe your site investigation needs more teeth before you bid. This is the kind of insight that only shows up when the change data has been captured consistently, which is really the argument for running all of this through one system instead of a drawer full of paper.
The bottom line
Scope changes will never stop coming, and you wouldn't want a project rigid enough to refuse them — flexibility is part of the service. The job is to make each change move through a clean path: caught early, documented while the evidence is fresh, assessed for time as well as money, approved through a routing everyone can see, and — the step that saves you — actually integrated into the look-ahead so the field builds it in the right order alongside everything else.
Tools like LookAheadWall help most on that integration seam: the point where an approved change has to become real activities on a weekly plan, sequenced against the trades you've already committed. But the software only pays off if the discipline is there — flag it before you build it, price the time, get the signature, then plan the work. Do that consistently and your change orders stop being the thing that derails the job and start being just another part of the schedule you already know how to run.