Ask a homeowner what worried them most during a renovation and price is rarely the answer. It is duration — living in a construction site and being told "a couple more weeks" four times. That makes the schedule a product feature, and every day over costs $180 to $450 with no revenue attached.
In this article
- The date you gave is a product feature
- Building a schedule that can survive contact
- Where the days actually go
- Selections are a schedule problem, not a taste problem
- A project in numbers
- Running the schedule week to week
- Communicating a delay so it does not become a dispute
- What the contract should say about time
- Five mistakes
- The five numbers
- Frequently asked questions
The date you gave is a product feature
Ask a homeowner what they were most anxious about during a renovation and price is rarely the answer. It is duration — living in a construction site, not knowing when it ends, and being told "a couple more weeks" four times in a row.
That makes the schedule a product feature rather than an internal document, and it explains a pattern most builders have noticed without naming: the company that finishes in fourteen weeks having promised twelve gets worse reviews than the company that finishes in sixteen having promised sixteen. The absolute duration matters less than whether the promise held.
It also has a direct financial consequence. Every day a job runs long costs general conditions — supervision, temporary power and facilities, dumpsters, insurance, equipment on site — commonly $180 to $450 a day on residential work, spent with no revenue attached. On a job that runs nine days over, that is $1,600 to $4,000 straight out of the margin, before any of the downstream effects.
Not the one caused by weather or a supplier. It is the one caused by a client selection that arrived late — because it is the only category of delay that is entirely preventable and the one builders most often absorb without pricing.
Building a schedule that can survive contact
Most residential schedules are a list of trades in the order they normally happen. That is a sequence, not a schedule, and it cannot tell you what a slip costs.
A usable schedule has four properties.
Named milestones with dates, not just durations. Demolition complete, foundation inspected, dry-in, rough-ins passed, drywall complete, cabinets set, punch list, final inspection.
Dependencies made explicit. Which tasks cannot start until another finishes. This is what turns a slip into a calculable number rather than a feeling.
Lead times for long-lead items, counted backwards from installation. Windows, cabinets, tile, specialty fixtures and appliances routinely run six to sixteen weeks, and each one has a date by which it must be ordered — which is usually a selection date for the client.
Buffer at the dependency points, not spread evenly across the job. A day of float before an inspection or a trade handoff is worth five days of float somewhere nothing depends on.
Then publish it. A schedule the client can see turns you from a source of anxiety into the person managing it, and it makes every subsequent conversation about a specific date rather than a general worry.
Keep reading — free
The rest of this article is worth the 20 seconds
Tell us where to send it and the full piece unlocks right here, along with everything else on this blog.
Rather just talk? Message us on WhatsApp.
Where the days actually go
Across residential projects the same six causes account for nearly all overrun, in roughly this order.
Client selections made late. The largest and the most preventable. Tile chosen in week six that needed to be ordered in week two.
Change orders. Every change adds days. Unrecorded, those days become "the builder was slow."
Trade availability. A sub booked without enough notice, or booked and then not confirmed.
Inspections. Scheduling lead time plus the possibility of a failure and a re-inspection. In some jurisdictions this alone is a week per inspection.
Concealed conditions. Rot, rock, undersized service, a slab that is not where the drawings say. Genuine, and manageable if the contract addresses it.
Weather and materials. Real, largely outside your control, and the only category most builders actually warn clients about.
The instructive part of that list is that the first three — more than half of typical overrun — are entirely inside the builder's process.
The other reason to keep the first three visible is that they are the ones a client can help with. Nobody can accelerate an inspection queue, but a homeowner who understands that a tile decision has a deadline attached to a completion date behaves very differently from one who was simply asked to pick a tile at some point.
Selections are a schedule problem, not a taste problem
The single highest-return change most residential builders can make is moving selections before construction starts.
Give the client a selection schedule at contract signing: every item, the allowance, and the date by which the decision must be made, counted backwards from its installation date. Not a wish list — a document with deadlines, signed alongside the contract.
Then state the consequence plainly and in advance: "a selection made after its deadline may add days to completion and may change the price." Written before anything goes wrong, this is a policy. Introduced in week nine, it is an excuse.
Where the client is genuinely undecided, help them decide rather than waiting. A showroom appointment booked by you, a shortlist of three, and a default option you will order if no decision arrives by the date. Most late selections are not disagreements; they are people avoiding a decision they find overwhelming, and removing options is a kindness that also protects your schedule.
A project in numbers
A $286,000 addition, sold on a 14-week schedule:
What happened:
• Tile and plumbing fixtures selected in week 7, needed by week 4: +8 days
• Three change orders, none with days recorded: +6 days
• Framing inspection failed once, re-inspection lead time: +4 days
• Drywall sub mobilized late: +4 days
• Weather: +3 days
• Delivered in 19 weeks, 5 weeks late
What it cost:
• General conditions at $310/day for 25 working days: $7,750
• Owner's additional supervision, 46 hours at $65: $2,990
• Next project start pushed 3 weeks, delaying its first draw: financing cost $1,400
• Client withheld final payment 5 weeks pending punch list: financing cost $900
• Total: $13,040, against a job budgeted for $71,500 of gross profit
Eighteen of the 25 days — the selections, the unrecorded change days and the late sub — were inside the builder's control. Written into change orders and a selection schedule, most of the remainder would at least have been attributed, which is the difference between a schedule that slipped and a builder who was late.
Running the schedule week to week
A schedule built at contract signing and never touched is a document. A schedule updated every Friday is a management tool, and the difference is about twenty minutes a week.
Update actual dates against planned, per milestone. Not a feeling about whether the job is on track — the specific number of days ahead or behind at each dependency point. A job that is four days behind at dry-in and nobody has said so out loud is a job that will surprise everyone in month three.
Look forward three weeks, not backwards. The useful question on Friday is not what slipped, it is what has to be ordered, confirmed or decided in the next twenty-one days for the following month to happen. Long-lead items, inspections, trade confirmations and client selections all live in that window.
Confirm every trade twice. One week out and again the day before mobilization. Most missed dates are discovered on the morning of, when nothing can be done, rather than on the Thursday before, when the whole week could be resequenced.
Recalculate the completion date every week, honestly. Builders tend to carry a private hope that the lost days will be made up later. They rarely are, and a completion date that quietly drifts for six weeks before being announced is the version clients experience as dishonesty rather than delay.
Communicating a delay so it does not become a dispute
Clients forgive delays. They do not forgive discovering one.
Tell them the moment you know, not when it becomes visible. A builder who says on Tuesday that cabinets are running two weeks late is managing the project. The same information discovered by the client on the day of expected installation is a builder who was hiding something.
Give the new date, not a range. "About two more weeks" is the phrase that destroys credibility, because it will be said again.
Say what caused it, without assigning blame to the client even when it is theirs. "The tile selection came through on the 14th, so the order lands the 28th and installation moves to the week of the 5th" is factual and does not require anyone to be at fault.
Send a weekly update whether or not anything changed. Three lines: what got done, what is next, whether the completion date moved. It costs five minutes and it eliminates most of the calls you would otherwise receive.
Record every schedule change against a change order. A completion date that moved through eleven documented changes is a project. The same date moved with nothing written is a broken promise.
The weekly update is also where a builder builds the credibility they will need later. A client who has received nineteen honest updates believes the twentieth when it carries bad news. A client who has received silence and then an apology has no basis to believe anything.
What the contract should say about time
Most residential contracts handle money carefully and time casually, which is backwards relative to where the disputes actually happen.
Four clauses are worth having. A substantial completion definition — when the work is usable for its intended purpose, which is what triggers final payment, as distinct from every punch item being closed. An excusable delay clause covering weather, concealed conditions, client-caused delay, permit and inspection timelines, and supply disruption, with the mechanism for extending the date. A change order clause that states days as well as dollars, so the completion date adjusts automatically as changes are approved. And a client-caused delay provision, which is uncomfortable to include and is the one that prevents late selections from becoming your fault.
Where liquidated damages appear in a contract handed to you — more common on commercial and some custom residential work — read them carefully and price them. A daily penalty you have not accounted for can consume the entire margin of a job that finishes only slightly late.
Five mistakes
1. Promising the schedule you hope for. Finishing in sixteen weeks having promised sixteen beats fourteen having promised twelve.
2. Letting selections happen during construction. The largest single source of delay, and entirely preventable with a signed selection schedule.
3. Change orders with no days on them. Every unrecorded change converts into "the builder was slow."
4. Announcing a delay when it becomes visible. The client's complaint is almost never the delay; it is finding out about it themselves.
5. Spreading buffer evenly. Float only helps where something depends on something else.
The five numbers
Promised versus actual completion, in days, on every job. The number that predicts your reviews.
Days added by signed change orders versus total slip. The proportion you can attribute is the proportion you can defend.
Delay days by cause. Selections, changes, trades, inspections, conditions, weather. It tells you which process to fix first.
General conditions cost per day, per job. Without it, a delay has no price and gets treated as free.
Days from substantial completion to final payment. Where schedule problems turn into cash problems.
A reliable schedule is a sales argument
Nothing you say about craftsmanship builds as much confidence as named milestones with dates. Send us your project mix and close rate and we will show you where the proposal is losing.
Frequently asked questions
What causes most construction delays on residential projects?
Six causes account for nearly all of it, and the largest is client selections made late — tile chosen in week six that needed ordering in week two. Then change orders, each of which adds days that become "the builder was slow" when nobody records them. Then trade availability, usually a sub booked without enough notice or booked and never confirmed. Then inspections, where scheduling lead time plus the possibility of a failure and re-inspection can cost a week each in some jurisdictions. Then concealed conditions like rot, rock or undersized service. Then weather and materials. The instructive part is that the first three, more than half of typical overrun, sit entirely inside the builder's own process.
How do you stop client selections from delaying a job?
Move them before construction starts. Give the client a selection schedule at contract signing listing every item, its allowance and the date the decision must be made, counted backwards from the installation date — a signed document with deadlines, not a wish list. State the consequence plainly and in advance: a selection made after its deadline may add days and may change the price. Written before anything goes wrong that is a policy; introduced in week nine it is an excuse. Where a client is genuinely undecided, help them decide rather than waiting: book the showroom appointment yourself, offer a shortlist of three, and name a default you will order if no decision arrives, because most late selections are avoidance rather than disagreement.
How should a builder tell a client the job is running late?
The moment you know, not when it becomes visible — a builder who says on Tuesday that cabinets are two weeks late is managing the project, while the same information discovered by the client on installation day is a builder who was hiding something. Give a specific new date rather than a range, because "about two more weeks" destroys credibility the second time it is said. State the cause factually without assigning blame even when it is the client's: "the tile selection came through on the 14th, so the order lands the 28th and installation moves to the week of the 5th." Send a three-line weekly update whether or not anything changed. And record every schedule change against a change order, so the date moved through documented decisions rather than broken promises.