New construction is a blank page. You pour the slab, you frame the walls, and the building comes up more or less the way the drawings said it would. Renovation is the opposite. You're working inside somebody else's decisions from thirty or fifty or a hundred years ago, and half of those decisions are hidden behind plaster you haven't opened yet. The schedule you build on Monday can be wrong by Wednesday afternoon when the drywall crew finds a cast-iron stack the plumber swore was long gone.
That's the real challenge of managing renovation and retrofit work, and it's why the tools and habits that keep a new-construction job on track don't automatically transfer. You're not managing a plan so much as managing the rate at which reality contradicts the plan. Good field management and look-ahead scheduling on a reno job isn't about locking down a bar chart six months out. It's about building a short window you actually trust, documenting what you find fast enough to keep decisions moving, and re-sequencing without losing the crew. Here's how experienced supers actually run these jobs.
Assume the drawings are a hypothesis, not a fact
On a gut renovation, the existing-conditions set is your best guess, drawn from as-builts that may never have matched what got built and definitely don't reflect the three tenant fit-outs since. Treat it that way. Before you commit a single crew to a sequence, walk the space with the foremen who'll do the work and mark up what's actually there.
The most valuable hour on any reno starts is the demo-eve walk. You pull ceiling tiles in a few representative bays, open one chase, and look above the corridor. You're checking for the things that blow up schedules: unexpected structure, live conduit that isn't on any panel schedule, abandoned-in-place mechanical, and the classic — a floor slab that steps two inches where nobody drew a step. What you find there decides whether your first three weeks are real or fantasy.
Build a short window you trust, then roll it
The instinct on an uncertain job is to plan less. That's backwards. Uncertainty is exactly why you want a disciplined look-ahead — you just keep the committed window short and honest. A rolling three-week look-ahead works well on reno: week one is firm and made-ready, week two is being cleared of constraints, week three is a sketch you fully expect to redraw.
The rule that separates a working plan from wishful thinking is constraint screening. Nothing goes into your firm week until it's made-ready — meaning the space is available, the preceding work is done and inspected, the material is on site, the shop drawings are approved, and any required shutdown or abatement is cleared. On new work, most tasks pass that screen. On renovation, a huge share fail it because of a discovered condition, and that's the whole point: the screen catches the fantasy before it becomes a crew standing around at 7 a.m.
This is where location-based, visual weekly work planning earns its keep over a traditional Gantt. When your plan is organized by area — this wing, this floor, this stack of units — you can see at a glance where a discovered condition has stalled a location and shift the crew to the next available area instead of sending them home. A tool like LookAheadWall is built around that location-and-trade-flow view for exactly this reason: on a reno you re-sequence constantly, and you need to see the whole board when you do it.
Document conditions fast, because the RFI clock is your real critical path
Here's the thing about a discovered condition: the demo itself takes an hour, but the decision about what to do can take two weeks if you let it. On renovation, your critical path is very often the RFI-and-change-order loop, not the field work. The unknown structural beam doesn't stop you — waiting for the engineer's response about the beam stops you.
So the discipline is to compress the documentation-to-decision cycle:
- Photograph everything, immediately, with a tape or scale in the shot. A dated photo of the opened wall with a measurement in frame is worth more than three paragraphs of description, and it's the evidence you'll need when the change order gets questioned.
- Get the RFI out the same day, not at the end of the week. A four-day-old discovery that hasn't been submitted is four days you'll never get back.
- Log the discovery against the schedule, not just in a binder. The value of tracking a found condition in your field or look-ahead software is that it's tied to the location and the affected tasks — so when you re-plan, you can see what that beam actually held up.
- Keep a running discovery log the owner can see. On occupied and institutional work especially, showing the owner a clean record of what you found and when protects the relationship when the number climbs.
Sequence the destruction, not just the construction
Everybody plans the build. Fewer people plan the demo, and on renovation the demo sequence is where jobs get hurt. Selective demolition means removing what goes while protecting what stays, and the order matters as much as it does on the way up.
A few hard-won rules. Shore before you cut anything that might be doing structural work — and on old buildings, plenty of things are doing structural work that weren't designed to. Cap and verify utilities before demo touches them; the difference between a dead line and a live one is a flooded floor or a very bad day for the guy with the sawzall. Sequence abatement ahead of any disturbance, and treat the abatement clearance as a hard predecessor in your look-ahead — nobody frames behind a wall that hasn't been cleared. And protect the stuff that stays before the adjacent demo starts, not after somebody's already chipped the terrazzo you're keeping.
Protection is a scheduled activity, not an afterthought
If the finish, the floor, the storefront, or the historic staircase is staying, protecting it is a line item with a start date, a responsible party, and a verification step — the same as any other task. The failure mode is treating protection as something that "happens." It doesn't. Somebody installs it, somebody checks it, and it stays checked through the dusty trades.
Build the protection installation into the week before the demo or rough trades hit that area, put a hold on the following work until protection is verified, and re-verify after the messy trades pass through. On historic and high-finish jobs, a walk with the owner's rep to sign off on protection before demo begins saves an argument later that you will otherwise lose.
Coordinate around the people still using the building
Occupied renovation — a hospital wing that's still admitting, an office floor above a running trading desk, apartments with tenants in place — adds a layer that new work never has. The occupants set constraints your schedule has to obey, and they're not flexible.
Noise and vibration windows, after-hours-only cutovers, negative-pressure containment for infection control, dust barriers, protected egress paths, elevator and loading-dock time slots — every one of these is a real constraint that belongs in your look-ahead the same way a material delivery does. The tie-in that can only happen Sunday at 2 a.m. isn't a footnote; it's a fixed point the rest of the week plans around. Notify occupants of impacts well ahead — a week's notice for a shutdown is a minimum, not a courtesy — and coordinate the system shutdowns for MEP tie-ins with the building's operations people, not just your own subs. The shutdown you didn't clear is the one that trips a fire alarm on a full floor.
MEP tie-ins to existing systems: map before you cut
Existing mechanical, electrical, and plumbing is the single richest source of renovation surprises. Panel schedules lie. That "spare" circuit feeds something. The chilled-water main you're tapping serves a wing you didn't know was on it.
Before any tie-in, trace and verify the existing system to the point of connection, and don't trust the label — ring it out, lock it out, and confirm it's dead. Coordinate the shutdown window as a hard-scheduled event with the building. Where you're joining old pipe to new, plan for adapters and transition fittings you probably can't buy off the shelf, and get them on order early because a two-week lead on a weird fitting will quietly become your critical path. Pressure-test and, on the electrical side, megger the new runs before you close anything back up — chasing a fault after the wall's skinned is a day you'll never enjoy.
Match, phase, and expect the change orders
Two more realities of existing-building work. First, matching. When you're patching into existing brick, plaster, terrazzo, or millwork, get a physical sample approved against the real thing before anyone runs production quantities — the match that looks fine on a submittal reads as a scar under the actual corridor lighting. Build that sample-approval step into the look-ahead as a predecessor to the finish work, because a rejected match after installation is a full redo.
Second, phasing. Occupied and operational renovations usually move in phases with temporary conditions between them — temporary walls, temporary power, temporary exits. Each transition is its own mini-project with a handoff, and the temporary condition itself has to be scheduled, inspected, and eventually removed. Track the phase transitions explicitly; the seams between phases are where things fall through.
And accept up front that renovation generates change orders at a rate new construction never will, because discovery drives change. That's not a failure of planning — it's the nature of the work. What separates a profitable reno from a bloody one is the speed and cleanliness of the loop: found condition, photo, RFI, direction, priced change, re-sequenced plan. Tighten that loop and keep the record clean, and the surprises stay surprises instead of becoming disputes.
The short version
Renovation rewards humility and speed, not a bulletproof long-range schedule. Keep your committed window short and screened for real constraints. Walk the space before you commit crews. Document discoveries the same day you find them and tie them to the location and the tasks they affect. Treat demo sequence, protection, abatement, and occupant constraints as scheduled activities with real predecessors — because they are. Do that, and the plan bends with each new find instead of breaking. The building will still surprise you. It always does. The goal is to be surprised on Tuesday and re-planned by Wednesday, with the crew working the whole time.