Build enough senior living and you eventually draw a memory care job, and the first time you do, you find out the hard way that it does not schedule like a normal assisted living wing. On the surface it looks like the same corridor of resident rooms, a dining room, a couple of activity spaces. Underneath it's a secured, alarmed, camera-covered building full of details that have to be commissioned and inspected before a single resident with dementia can walk the halls. Miss the coordination on those details and you don't get a punch list item. You get a facility that can't open, because the state won't license it and the operator won't accept the keys.
This is a walk through what actually drives a memory care schedule, where these jobs get away from superintendents, and how to sequence the specialized scopes so your look-ahead reflects reality instead of the optimistic bar chart the master schedule started as.
Why memory care is its own animal
Memory care serves residents with Alzheimer's and other forms of dementia, and the entire building is designed around one governing fact: these residents will try to leave, and some of them cannot be allowed to. That single requirement — elopement prevention — ripples into almost every trade. Doors get delayed egress hardware and mag-locks tied to the fire alarm. The whole envelope becomes a controlled perimeter. Courtyards get fenced and gated so a resident can wander outside without wandering off. Nurse call becomes a wander-management system with door alarms and, on many projects, resident-worn tags read by antennas in the door frames.
None of that is exotic on its own. What makes it hard is that these are integrated life-safety systems, not stand-alone scopes. The access control talks to the fire alarm. The wander-management heads sit in the same door frames as the hardware and the electric strikes. Low-voltage, door hardware, electrical, and fire alarm all converge on the same openings, and they converge late — right when everyone is trying to close out. Your look-ahead has to see that convergence coming weeks out, because the trades won't warn you. They'll each show up assuming the opening is ready for them and find the other two haven't been.
The secured perimeter is the critical path — treat it that way
On a normal building, doors and hardware are a finish item you can chase during punch. On memory care, the secured perimeter is the thing the operator's license hangs on, so it belongs on your critical path from the day you set door frames. Sequence it deliberately:
- Frames and blocking first, and get them right. Delayed-egress hardware, mag-locks, and wander-antenna heads all need backing and conduit stubbed into the frame or the wall above it. If the electrician doesn't know a given opening is a secured opening before the drywallers close the wall, you're cutting it open later. Mark every secured and alarmed opening on the field set and walk it with the EC before rock goes up.
- Rough-in the low-voltage and power to the openings during the wall. Card readers, door position switches, electric strikes, mag-locks, and antennas each want their own home run. This is the step teams skip, then pay for.
- Hang doors and install hardware, then integrate. The mechanical install is quick. The integration — tying the mag-lock release to the fire alarm, mapping every door into the access control head end, testing the delayed-egress timing — is not, and it can't start until the fire alarm and access control panels are alive.
Build that as a trade-flow sequence in your look-ahead, one opening type flowing frame → rough-in → close → hang → integrate → test, and you'll spot the openings that are three weeks behind the pack long before they become the reason the building can't be commissioned. A location-based weekly work plan is exactly the right tool here, because "the doors" isn't one activity — it's the same short chain repeated across forty openings, and you care about which ones are lagging, not the aggregate percent complete.
Therapeutic design details that quietly eat your float
Memory care buildings are designed to reduce agitation and support residents who are disoriented, and those design intentions land on the trades as a pile of small, fussy, easy-to-miss requirements. Individually none is a big deal. Collectively they're where the finish schedule goes sideways because a super treated them like ordinary finishes.
- Flooring transitions. Dark thresholds and abrupt color changes read as holes or steps to a dementia resident, so specs push for continuous, low-contrast transitions and no shiny floors that look wet. That constrains flooring products, transition strips, and sometimes the whole floor-flatness spec. Catch it in submittals, not when the flooring installer improvises a transition you have to tear out.
- Handrails everywhere, and continuous. Corridors want continuous rail, which means blocking behind the drywall in more places than a normal corridor. Same rule as the doors: it has to be in before rock.
- Lighting and glare. Even, glare-free light with good color rendering, often on circadian controls that ramp through the day. That's a lighting-controls commissioning item, and controls commissioning is a schedule sinkhole on every job — start the programming and functional testing early.
- Memory boxes, wayfinding, and cueing. Recessed shadow boxes outside resident doors, color-coded neighborhoods, tactile cues. More blocking, more coordination, more little rough-in items that live in the walls.
The through-line: most of the therapeutic scope is rough-in you can't see and can't skip. Once the wall is closed, adding it back is demolition. Put a "secured / memory-care rough-in complete" hold point in your weekly work plan for each unit before you release drywall, and make the framer, EC, and low-voltage tech sign off on it together. That five-minute walk is the cheapest insurance on the job.
Wander paths and secured courtyards
A good memory care plan gives residents a continuous loop to walk — indoors and out — with no dead ends, because a dead end confuses and frustrates a wandering resident. Outdoors, that loop runs through a secured courtyard, and the courtyard is a coordination trap. The fence or wall has to be tall enough and detailed so nothing becomes a foothold, the gates tie into access control, and the landscaping can't include anything toxic or anything a resident could climb. Site and landscape trades rarely think about elopement, so this scope needs the same perimeter-integration attention as the interior doors. Schedule the courtyard gate hardware and its access-control tie-in alongside the interior secured openings, not as a loose site item that shows up at substantial completion.
Commissioning and licensing: the real finish line
Here's the part that surprises superintendents coming from market-rate work: substantial completion is not the finish line. The finish line is the operator getting a license, and that requires the security and life-safety systems to be commissioned and, on most jobs, witnessed by the authority having jurisdiction and the state licensing surveyor. Those inspections book weeks out and they don't move for you.
Work backward from the survey date and pull it into your look-ahead as a hard milestone with the dependencies hung off it:
- Fire alarm accepted first. Access control and delayed egress can't be fully tested until the fire alarm is inspected and accepted, because the whole system releases the secured doors on alarm. Fire alarm final is a predecessor to nearly everything security-related — sequence it early, and treat any slip in it as a slip in the license date.
- Integrated functional testing. Every secured door tested for alarm release, delayed-egress timing, and access-control mapping. Nurse call and wander management tested end to end. Budget real days for this and expect a punch cycle — it never passes clean the first pass.
- Owner and surveyor walks. The operator's own team will test elopement scenarios harder than any inspector. Build a buffer between "we think it works" and the licensing survey so you have room to fix what the first walk finds.
A one- to two-week buffer between integrated testing complete and the licensing survey is not padding — it's the difference between opening on time and slipping a month waiting for the surveyor's next available date. Protect it, and don't let earlier trades erode it.
Making the look-ahead do the work
Memory care is where short-interval scheduling stops being a nice-to-have. The three- to six-week look-ahead is the window where you catch the opening that's missing its rough-in, the courtyard gate that never got its access-control home run, the fire alarm final that's drifting toward the license date. Pull those specialized scopes into the same rolling weekly plan you run everything else on, sequence them as explicit trade flows across locations, and share it with the low-voltage, fire alarm, and door-hardware subs who otherwise never see each other's schedules. This is exactly the coordination a tool like LookAheadWall is built to make visible — the same secured-opening chain repeated across the building, laid out by location so the laggards jump off the wall instead of hiding inside a rolled-up percentage.
The superintendents who deliver these jobs on time aren't smarter about drywall or paint. They just refuse to treat the secured, therapeutic, and life-safety scopes as finishes to be chased at the end. They put them on the critical path early, they hold the wall until the rough-in is verified, and they work backward from the licensing survey instead of forward from substantial completion. Do that, and a memory care job runs like any other well-planned building. Skip it, and you'll spend the last month cutting open walls you already closed.