The first design-build job I ran, I made the mistake of treating the design side like it was somebody else's problem. I had my construction lookahead dialed in — slab pours, framing, MEP rough-in, all sequenced and staffed. What I didn't have on that board were the design deliverables feeding those activities. So when the structural engineer was three weeks late releasing the stamped foundation drawings, my board still cheerfully said we poured on schedule. We didn't. We sat. And the owner, who was supposed to have picked a curtain wall system by then, hadn't even been asked to.
That's the whole difference with design-build. On a traditional design-bid-build job, the design is basically done before you mobilize, and your lookahead only has to worry about who's swinging a hammer. On design-build, design and construction run at the same time, and the schedule that only tracks the hammer is lying to you. The decisions being made in the design office this week are the constraints your crews hit in six weeks. If those decisions aren't on the board, you're flying blind and you won't know it until concrete is waiting.
Put design on the same board as the trades
The single most useful move is also the one people resist: put design activities directly into your look-ahead, right alongside the field work, in the same six-week window. Not in a separate design schedule that lives in the architect's inbox. The same board your foremen look at.
What belongs there:
- Design package release dates — the day a set of drawings is issued for construction, not "issued for review." Those are two different milestones and confusing them is how you end up building off a 60% set.
- Review and approval cycles — owner review, permitting, and any AHJ turnaround. These eat calendar time you don't control, so they need real durations, not zero.
- Shop drawing and submittal windows — when a sub can start drawings, when they're due, and how long review takes.
- Owner and design decisions — finish selections, system choices, anything that's still open. Each one is a constraint with a name and a due date.
When design lives on the same weekly work plan as construction, the dependency becomes visible to everyone. Your framing foreman can see that his start is riding on a slab detail that's still in review, and he'll be the one asking about it in the Monday meeting — which is exactly who you want asking.
Work the decisions backward from the field
Here's the mental habit that separates people who run design-build well from people who get run over by it: you schedule design decisions backward from the field work they unlock, not forward from when the design team feels like getting to them.
Take a simple example. You need to pour footings on the 15th. The rebar has a two-week lead. The fabricator needs approved shop drawings before they'll fab, and shop drawing review is a week. The shops can't be drawn until the structural design is released. Walk that chain backward and the structural release date isn't "sometime soon" — it's a hard date roughly five weeks before your pour, and if it slips, everything downstream slips with it.
Every construction activity in a design-build job has a chain like that hanging off the back of it, and half those links live in the design office. The value of a good look-ahead here isn't that it draws you a pretty bar chart — it's that it forces you to make those chains explicit and assign real dates and owners to the design links, the same way you would to a concrete pour. A rolling lookahead schedule that only shows the field half of the chain is telling you a comforting story that isn't true.
Long-lead items are where design-build jobs actually die
If a design-build project blows its schedule, nine times out of ten you can trace it to a long-lead item that was ordered late because the design that defined it finished late. Switchgear, elevators, chillers, structural steel, curtain wall, custom AHUs — these are the items with 12 to 30 week leads, and they all depend on design being firm enough to release a purchase order.
Pull those items out and reverse-schedule them first, before you worry about anything with a short lead. For each one, you need three dates on the board: when the design has to be locked to define it, when the PO has to be cut, and when it lands on site. Then check that the design-lock date is actually achievable given where the design team is today. If it isn't, you've just found your real critical path, and it's weeks earlier than you thought.
The trap is optimism. It is very easy to write down that the mechanical design "will be done by March" because that's what everyone hopes. Design durations are softer than construction durations — a wall takes as long as it takes to build, but a design can iterate three extra rounds because the owner keeps changing their mind. Pad design activities more than you think you need to, especially anything that requires an owner decision to close out. A one-to-two week buffer on a straightforward construction handoff is normal; on a design deliverable that's waiting on an owner selection, two to three weeks of float is not paranoid, it's realistic.
Run design commitments through the same planning discipline
Design-build is where Last Planner-style planning really earns its keep, because it makes designers commit to dates in front of the same room that's committing to field work. In a good weekly planning session, the design lead should be making promises the same way the trade foremen do: "the structural release will be issued Thursday," "the owner sign-off on finishes is committed for the 12th." Those are commitments you track and measure, and when they're missed, you find out why — same as a missed field task.
The mechanism that matters most is constraint tracking. Before any construction activity is allowed to be "ready," you screen it for constraints, and on design-build a huge share of those constraints are design and decision constraints: missing information, an unmade owner call, an unresolved clash, a submittal still in review. If you're only screening for materials, labor, and equipment — the classic three — you're missing the category that actually holds design-build jobs up. Add "design/decision" as a constraint type and force every activity through it before it's cleared to start.
Where the two sides hand off — and where they tear
The interfaces between design and construction are where design-build jobs bleed schedule, and they're worth watching harder than anything else on the board.
- Preliminary vs. final information. Crews will build off preliminary drawings if you let them, and then rework when the final differs. Your board should clearly flag which information is "for construction" and which is still "for coordination only." A foreman scheduling app or shared field view that shows this distinction saves a lot of demo.
- Field-back-to-design feedback loops. Design-build's advantage is that the field can inform the design — an existing condition, a smarter detail, a constructability fix. But that only helps if there's a scheduled path for it. Build in RFI and design-response windows; a two-day-turnaround expectation that everyone knows about beats an open-ended "we'll get to it."
- BIM and clash resolution. If you're modeling, clash detection and resolution is a real activity with a real duration, not a background hum. Schedule the coordination sessions and the resolution time before rough-in, not during it. Closing a wall on an unresolved clash is a mistake you pay for twice.
Subs who design and build their own scope
On most design-build jobs, some trades — often MEP and fire protection — carry their own design responsibility. Their schedule has an extra front end you have to account for: engineering, shop drawings, coordination, then fabrication, then install. If you only schedule their install date, you've skipped the two months of design and fab that have to happen first.
Get those subs into the planning conversation early and pull their design and coordination milestones onto the same look-ahead. When a mechanical sub is doing their own layout, their coordination with the structural and the ceiling grid is a scheduled dependency, not a favor. The earlier that shows up on the board, the fewer nasty surprises during rough-in.
What the tool actually needs to do
You don't need a $100k enterprise platform to run design-build scheduling well, but whatever you use has to handle a few things honestly:
- Show both design and construction activities in the same weekly view, so the whole team sees the dependencies without hunting for a second schedule.
- Express design-to-construction links clearly — this pour waits on that release — so a slip in design visibly ripples into the field.
- Let the whole integrated team — design, construction, owner, key subs — look at and update the same plan, because the second design and construction are working off different schedules, you're back to design-bid-build with worse coordination.
This is the space a visual, location-based look-ahead tool is genuinely built for. Something like LookAheadWall lets you lay design deliverables and field activities on the same weekly board, connect trade-flow sequences so a downstream crew's start is visibly tied to the release it depends on, and share the live plan with the subs on their phones. The point isn't the software — it's that everyone is arguing over the same set of dates instead of three conflicting ones.
The mistakes that keep showing up
After enough of these jobs, the failure modes rhyme:
- The construction-only board. A beautiful field schedule with no design on it. It will show you green right up until the day concrete is waiting on drawings.
- Optimistic design durations. Assuming design finishes on the hoped-for date. It rarely does, and the buffer you didn't build is the delay you eat.
- Siloed planning. Design plans in one meeting, construction in another. That's just design-bid-build with the walls knocked down and none of the coordination benefit — the whole reason to do design-build is that the two teams plan together.
- Owner decisions with no due dates. "The owner will pick finishes eventually" is not a schedule input. Every open owner decision needs a date on the board and someone chasing it.
Bottom line
Design-build only pays off if the schedule reflects how the work actually flows — design and construction braided together, decisions upstream driving crews downstream. The discipline is simple to state and hard to hold: get the design deliverables and owner decisions onto the same look-ahead as the field work, schedule them backward from the crews they unlock, screen every activity for design constraints before you call it ready, and watch the handoffs between the two sides like they're the critical path, because they usually are.
Do that, and the overlap between design and construction becomes the advantage it's supposed to be instead of the thing that quietly sinks you. Skip it, and you'll find out the hard way — standing next to a ready crew and a truck full of concrete, waiting on a drawing that was due three weeks ago.