Ask ten superintendents how far out they plan and you'll get ten answers. On a ground-up high-rise, a six-week window makes sense — you've got long-lead steel, a tower crane, and enough duration that six weeks is barely a slice of the job. On a tenant improvement, six weeks is often most of the project. Plan that far out on a ten-week build-out and half of what you drew will be wrong by the time you get there, because the tenant hasn't picked their finishes yet and the landlord hasn't cleared your after-hours access.
After running a lot of TI jobs — office suites, retail, medical, the occasional restaurant that made me question my career choices — I've landed on four weeks as the horizon that actually fits the work. Not because four is a magic number, but because it lines up with the three things that govern a TI schedule: how fast material shows up, how far a sub will commit a crew, and how long a tenant takes to answer a question. This is a look-ahead article, so let me explain why that horizon works and, more usefully, how to run it.
Why TI Work Breaks the Long-Horizon Habit
Tenant improvements are short, tight, and unforgiving. A typical build-out runs eight to sixteen weeks. There's almost no total float, so a two-day slip in one trade isn't absorbed by the schedule — it lands directly on your certificate of occupancy date. And you're stacking a dozen trades into a few thousand square feet where two crews can't physically stand in the same corner.
That combination changes how you plan. On a long job, a look-ahead is a rolling detail window in front of a big master schedule. On a TI, the four-week window often is most of the remaining job. You're not zooming into a small piece — you're looking at the whole thing at working resolution. That's exactly why the horizon has to be short enough to stay honest and long enough to give people time to react.
The Four-Week Sweet Spot
Four weeks works because it matches the real cycle times on a TI:
- Material lead times. Most of what goes into a standard build-out lands inside four weeks — hollow metal frames and doors, standard commercial carpet and LVT, acoustic ceiling grid and tile, light fixtures, standard plumbing fixtures. The stuff that blows past four weeks is predictable: custom millwork, specialty glass and storefront, switchgear, long-lead HVAC units, anything with a "quick-ship" upcharge you didn't pay. Those you track as constraints from day one, separate from the rolling window. Everything else, four weeks captures cleanly.
- Sub commitment. A drywall or electrical foreman running three or four jobs at once can tell you honestly what he'll have available for the next four weeks. Push him to commit crews eight weeks out on a small job and he'll give you a number he has no intention of honoring — because he doesn't know yet either. Four weeks is inside his real planning horizon, which means the commitments you get are commitments you can hold him to.
- Tenant response time. Four weeks is roughly how long it takes a tenant's organization to actually make and approve a decision — not the decision itself, but the meeting, the sign-off, the person who's traveling. Ask for an answer inside four weeks and you're being reasonable. Ask inside one and you're the reason it's late.
Planning by Area, Not Just by Date
The single biggest scheduling mistake on TI work is planning by trade and date while ignoring space. You draw framing for week one, drywall for week two, and it looks fine on paper — until you realize the electrician needs to rough-in the walls the framers just built, in the same square footage, before the drywall goes up. Now three trades want the same corner in the same week and nobody told them.
Run your look-ahead by area. Break the floor plate into zones — Suite A, the corridor, the core restrooms, the open office — and sequence each zone through the standard interior flow: layout, top-track and framing, in-wall rough-in (electrical, plumbing, low-voltage, fire sprinkler drops), inspection, insulation, one side of drywall, then the other, tape and finish, prime, ceiling grid, above-ceiling rough, ceiling inspection, tile, paint, flooring, trim and doors, fixtures and devices, final clean.
When you lay that out zone by zone across four weeks, the trade stacking jumps off the page. You can literally see where two crews collide and slide one zone a few days so the work flows in a wave through the space instead of a pileup. This is where a visual, location-based look-ahead tool earns its keep — LookAheadWall lets you build the plan by area and trade flow rather than as a flat bar chart, so the collision is obvious before it happens instead of at 6:45 a.m. when both foremen show up with their crews.
The Buffers That Save You
The schedule that fails is the one drawn with zero space between trades. On paper, drywall finishes Friday and paint starts Monday. In reality, the mud isn't dry, the sander hasn't been through, and the painter walks a wall that isn't ready. Build buffers where the trade handoffs are genuinely lumpy:
- Rough-in to cover: hold a day for the in-wall inspection. Never schedule insulation or one-side drywall the same day the inspector is supposed to sign the rough. He might not show, or he might red-tag something, and now you've buried a defect behind the board.
- Tape-and-finish to paint: give last-coat mud time to dry and get sanded — usually a day, more in a cold or humid space. Paint over green mud and you'll see every seam under the wall washers.
- Ceiling grid to tile: keep tile out until the above-ceiling inspection passes. Drop tile early and you'll be pulling it back out when the inspector wants to see the fire caulking and the flex whips.
- Flooring near the end: resilient flooring and carpet go in late, after the messy trades are done, and hard-surface flooring often needs the space acclimatized and the slab moisture-tested first. Rushing this is how you end up replacing $12-a-yard carpet because a painter tracked across it.
A day of buffer at each of these handoffs feels like slack you can't afford on a tight job. It's the opposite. Those buffers are what keep one slipped trade from cascading into your CO date.
Inspections and the Path to CO
On TI work the inspections come fast and they gate everything. In-wall rough, above-ceiling, fire alarm and sprinkler, sometimes a separate low-voltage or health-department inspection on a medical or food job — then final building, fire, and the certificate of occupancy. Your look-ahead should carry every one of them as a hard milestone, because a missed or failed inspection on a tight job is a direct hit to the finish date.
Two rules of thumb. First, know your jurisdiction's real lead time for scheduling an inspection — some want 24 hours, some want three days, and finding that out the morning you need the inspector is too late. Second, the whole schedule should always show the path to CO-ready. On a TI, the CO inspection is the finish line, and if you can't see the chain of predecessors that has to close before you can call for it, you don't actually have a schedule — you have a wish.
The Landlord Constraints Nobody Draws
Here's what sinks first-time TI schedulers: they plan the construction and forget the building. TI work happens inside somebody else's occupied property, and the base building imposes constraints that have nothing to do with your trades:
- Freight elevator reservations and loading dock windows — often a two-hour slot you have to book days ahead.
- No hot work, no core drilling, no loud demo during business hours in an occupied building.
- After-hours access that requires a security escort or advance notice.
- Tie-ins to base-building systems — power, water, HVAC, fire alarm — that need landlord approval and frequently the landlord's own contractor, on the landlord's timeline.
Every one of these belongs in the look-ahead as a constraint with a date. That system tie-in that needs the base-building engineer? If you don't flag it three weeks out, you'll find out you need it the week you need it, and the landlord's guy is booked. The four-week window gives you exactly enough runway to line these up — if you actually put them on the plan.
Managing Tenant Decisions Without Getting Blamed for Them
Tenants change their minds. It's the nature of the work — they're moving into this space and they want it right. The trick isn't preventing changes; it's making the schedule consequences visible so the decision gets made in time or the tenant owns the delay.
Track finish selections, equipment choices, and layout decisions as dated constraints tied to the work they gate. If the tenant's paint and carpet selections have to be locked before you can order, and the order plus lead time is three weeks, then the selection deadline is three weeks before you need to install — put that date in front of them in writing. When a tenant sees "we need your flooring pick by the 14th or the move-in slips," decisions happen. When they just hear "soon," they don't. A shared, visible look-ahead does a lot of this arguing for you, because the tenant can see their own decision sitting on the critical path.
Running the Weekly Cycle Fast
The four-week horizon is the frame; the weekly work plan is the engine. Every week you pull the next week's work from the look-ahead, confirm every task is actually ready — material on site, predecessor done, area clear, inspection passed, manpower committed — and only then let a foreman commit to it. Anything with an open constraint doesn't go on the weekly plan; it goes on the constraint log to get cleared first.
On TI jobs this cycle runs hot. Constraints that take weeks to clear on a big job clear in a day or two here — a submittal, a delivery, an inspection. Because of that, your Percent Plan Complete should be high. On a well-run TI with a tight crew and a short chain of command, hitting 85% or better of your weekly commitments is realistic, and if you're consistently down in the 60s, the problem usually isn't the crews — it's that you're committing to work that was never actually ready. The measure of a good TI look-ahead is boring: what you said would happen this week is what happened.
Common Ways TI Schedules Go Wrong
- Over-planning. Drawing a six-week detailed plan on an eight-week job. You'll spend hours maintaining fiction. Keep the detailed horizon at four weeks and a lighter milestone view beyond it.
- Assuming small means casual. "It's just a build-out, we don't need a real look-ahead." Small jobs need the coordination more, not less, because there's no float to recover a mistake.
- Zero buffers. Butting trades edge to edge to hit an aggressive date, then watching the whole thing cascade off the first slip.
- Ignoring tenant and landlord as scheduled parties. They aren't spectators — their decisions and access windows are predecessors, and they belong on the plan with dates like any subcontractor.
- Weekly plans full of not-ready work. Committing to tasks with open constraints, then wondering why you keep missing.
The Bottom Line
Four weeks fits tenant improvement work because it matches the actual clock speed of the job: material lands inside it, subs will honestly commit within it, tenants can decide across it, and it's short enough that the plan stays true. Run it by area so you catch trade collisions before the crews do, build a day of buffer at the ugly handoffs, carry every inspection and landlord constraint as a dated milestone, and drive it with a fast, honest weekly cycle.
Good look-ahead software — LookAheadWall included — makes this easier by letting you plan visually by location and trade flow, share the live plan with your subs and the tenant, and update it from your phone as things move on site. But the tool only amplifies the discipline. Get the horizon right, respect the handoffs, and put the landlord and the tenant on the schedule where they belong, and you'll finish TI jobs on time often enough that the repeat work starts finding you.