How I’d actually use this on a flip
I size the construction loan against this number, not the median. The lender wants a finish date; the project wants a buffer; the calc gives me both. On a Chicago-metro gut rehab at 2000 sqft, typical permit, moderate complexity, GC-managed — the median horizon is roughly eight months, and the buffered ceiling adds another five to seven weeks for inspection re-dos and the cabinet maker who blocked another GC’s order ahead of mine. I always quote the lender the buffered ceiling and explain the 15-20% cushion in the same sentence as the loan term.
The owner-builder toggle is the one I run twice on every deal. If the calc says nine months GC-managed and eleven months owner-builder, the question I ask myself isn’t “can I save the GC markup” — it’s “is the construction loan term long enough to absorb the 1.26x slowdown and still service rate-and-term?” The Census 2024 anchor is exactly that: 12.1 months contractor versus 15.2 months owner-builder for new construction. The calc applies the same ratio to the rehab durations because the math behind the slowdown is coordination overhead, not slower trades.
Permit jurisdiction is the third pass. A 1920s Chicago bungalow in a fast-jurisdiction suburb clears plans in two weeks and breaks ground inside a month. The same property in a slow Cook-County jurisdiction with a structural-review department reviewing every joist size loses six weeks of construction-loan term to plan revisions before demo starts. The calc surfaces the difference; the pre-offer walk-through is when I check which jurisdiction the property sits in.
Where the schedule actually breaks
Five places the timeline lies to operators who don’t know to look — exactly where the calc’s critical path and delay-risk notes are pointed:
- Permit revisions.The first plan submission rarely clears. The second often doesn’t either. Strict jurisdictions queue revisions in two-week increments; on a schedule that already counts permits at five weeks, two revisions is ten weeks of construction-loan time before a single line of demo paint comes off. If the deal pencils only with a one-revision assumption, the deal doesn’t pencil.
- Rough-in inspection slip.All three rough trades — electrical, plumbing, HVAC — converge on a single inspection gate. If any one of them misses the gate, the entire project sits while that trade reschedules. Plan a one-week reinspection buffer; that’s where projects pick up their first real slip. The calc surfaces this on the rough-inspection delay-risk note.
- Cabinet + counter lead time.The two-week counter-fab window after templating isn’t a phase you can compress; it’s a supplier lead time you eat in calendar days regardless of how fast your install crew moves. Order cabinets the day permits clear or you’re paying for the lead time at the wrong end of the project — when you’re trying to call the final inspection.
- Drywall finish quality vs schedule pace.Drywall is the longest single phase on a residential rehab. Hang + tape + 5-coat finish + sand on 2000 sqft is four weeks at full crew. Owner-builder pace stretches that to five or six because the best mudders don’t run their best crews on owner-managed jobs. If the calc’s OB total feels wrong, it’s usually drywall pulling the slowdown.
- Final inspection forgotten items.Final failure on a forgotten smoke detector or an unrated dryer vent costs a two-week reinspection slot — and the punch-list phase can’t close out until the certificate of occupancy lands. Walk the punch list with the inspector’s checklist a week before you call. The calc hard-codes a three-day inspection block; the slip is what the buffer is for.
Methodology
Every day count on this page traces to one of three layers: Census 2024 single-family build-duration anchors, SiteworkMath phase calibrations from flipper experience + AIA contract cadences, and standard CPM forward + backward pass for critical-path identification. The full per-cluster sourcing tier in methodology spells out which sources back which kinds of claims. The output is a planning horizon, not a contractor schedule — the schedule caveat is part of the engine output for a reason.
Show the formulas
- Per-phase duration:
baseDays × sqftFactor × permitFactor × obFactor × complexityFactorwith each factor applied per the phase’s response setting (full / partial / none). - Sqft scaling (sublinear):
(sqft / 2000) ^ 0.7. Doubling sqft adds about 62%, not 100% — crews flex with project size. - Permit jurisdiction multiplier: fast 0.4× (small Chicago suburbs) / typical 1.0× (most Chicago metro) / slow 2.5× (Cook County strict). Applies fully to permit review; partially to design + final inspection.
- Owner-builder multiplier: 1.26× on coordination phases — anchored on Census 2024 single-family build durations (15.2mo OB / 12.1mo contractor). Inspections and design are unaffected.
- Complexity multiplier: simple 0.85× / moderate 1.0× / complex 1.18×.
- Critical path:forward pass establishes earliest start + finish from predecessors; backward pass from successors establishes latest start + finish; phases with float < 0.5 day are critical.
- Buffer: 15% on simple / 17.5% on moderate / 20% on complex — applied to the critical-path total to produce the bufferedTotalDays ceiling.
- Engine logic:
lib/sitework/timeline/timeline.tswith phase catalog inphases.tsand types intypes.ts. Tested intimeline.test.ts.
Frequently asked
What does this construction timeline calculator estimate?
A planning-range schedule horizon — the calendar-day total an operator uses to size the construction loan term and decide whether the project's pace pencils before booking the GC. The 20-phase catalog runs design → permits → demo (or site-prep for new-build) → structure → envelope → MEP rough-in → rough inspection → insulation → drywall → finishes → final inspection → punch list, with predecessors that run a forward + backward CPM pass to identify the critical path. Output is the median + a buffered ceiling — not a contractor schedule.
Why is owner-builder slower than contractor-managed?
Per the US Census 2024 Survey of Construction, single-family contractor-built homes average 12.1 months from start to completion vs 15.2 months for owner-built — a 1.26x slowdown. The slowdown isn't because OB does the work slower (subs run the same pace); it's coordination overhead. A single owner running 8 trades has gaps in the schedule that a GC's superintendent fills automatically. The calculator applies the 1.26x to coordination-heavy phases (rough trades, finishes, drywall) and skips it on inspection wait times and design phases that don't depend on management style.
How does the permit jurisdiction tier change the schedule?
Permits are calendar risk, not just fees. The fast / typical / slow tiers correspond to ~0.4× / 1.0× / 2.5× on the permit review phase: fast jurisdictions (small Chicago suburbs, contractor-friendly municipal review) clear plans in 1-2 weeks; typical (most Chicago metro) is the 5-week baseline; slow (Cook County strict, structural review departments) routinely take 10-12 weeks plus a third revision cycle. Strict-jurisdiction risk doesn't just delay the project — it eats construction-loan term while the calendar burns. The methodology dropdown shows the per-phase multiplier.
What's the critical path and why does the Gantt outline it?
Critical path = the chain of phases with zero schedule float. If a phase on the critical path slips a week, the whole project slips a week. If a phase off the critical path slips a week, the project may not slip at all. On a residential gut rehab the critical path almost always runs through design → permits → demo → framing → MEP rough-in → rough inspection → drywall → finishes → final inspection → punch list. Phases like windows-doors and tile run in parallel with siblings (roofing dry-in / flooring) and have float. The Gantt strip outlines the critical-path bars in rust so operators can see at a glance which phases need protection.
Why are inspection gates rendered as crisp vertical markers?
Inspection days aren't fuzzy. The inspector either shows up or doesn't, and the gate either passes or doesn't. The rest of the Gantt uses banded ends to signal that real construction phase boundaries are uncertain — drywall doesn't actually finish on day 119, it finishes the week of day 119 — but inspections are crisp dates by design. Visually they're 3-day rectangles rendered as a sharp 3px vertical bar in primary rust.
How accurate is this calculator's output?
Planning range, not contractor schedule. Phase durations are SiteworkMath calibrations anchored on Census 2024 build-duration data, AIA contract cadences, HomeAdvisor / Fixr cross-checks, and flipper experience in the Chicago metro. Sublinear sqft scaling (^0.7) reflects that crews flex with project size. The 15-20% buffer on the ceiling accounts for inspection re-dos, weather windows, and supplier ghost-jobs that aren't directly on any phase. The cost calc next door surfaces the dollar-side of the same project decision.
What I’d do next
- Pair the schedule with the cost — gut rehab cost calc
The same project decision has two halves: dollars and calendar. Run the cost calc with matching inputs to surface the feasibility budget alongside this schedule horizon.
- Read the planning-range methodology
What's anchored on Census 2024, what's a SiteworkMath calibration, and how the buffer cushion is sized.
- Refine the foundation-work line — concrete yardage calc
If foundation work is in scope (and adds the foundation phase to the timeline), drop into the concrete calc to size the slab / footing / pier order.
By James Wu. Planning-range methodology and per-cluster sourcing tiers in methodology. Owner-builder slowdown anchored on US Census 2024 Survey of Construction (single-family build durations: contractor 12.1mo / owner-builder 15.2mo → 1.26x). Phase durations are SiteworkMath calibrations anchored on flipper experience + AIA contract cadences + HomeAdvisor / Fixr cross-checks. Critical path computed via standard forward + backward CPM pass. Engine logic in lib/sitework/timeline/timeline.ts. Numbers are SiteworkMath planning ranges, not contractor schedules — the schedule caveat is part of the output for that reason. Not structural, financial, or legal advice. Full methodology.