Time windows
A retail delivery window is a courtesy. A home health time window is often a clinical fact wearing a scheduling problem's clothes. What makes them hard, what actually helps, and what to ask.
Most scheduling advice talks about the day as if any visit could go anywhere in it. In home health, a lot of visits cannot. A patient on insulin needs to be seen before they eat. A dialysis patient is simply not home three days a week. A wound needs redressing on a clock the wound sets, not the route. Time windows are not a scheduling nicety here; they are clinical facts wearing a scheduling problem's clothes.
A retail delivery window is a courtesy — early is fine, a little late is fine. A home health time window is often a physiological or logistical hard stop:
None of these are preferences a schedule can round off. A route that is efficient on paper but misses three of these in one day did not save time — it created three problems that each now need a phone call.
| What the window requires | What it does to the route |
|---|---|
| Insulin before breakfast | Forces an early, often inconvenient first stop, whatever the rest of the day's shape wants. |
| Dialysis days | Removes the patient from the map for that date, and narrows the visit to a short pocket either side. |
| A redressing interval | Drifts forward a little each week if the visit keeps landing late, until it stops matching the order. |
| Two disciplines, one home | Needs two routes to agree on a window neither of them owns alone. |
A schedule built by hand tends to solve the loudest window in the room — usually the earliest one — and let the rest land wherever the day happens to put them. Holding five or six patients' windows in working memory while also sequencing drive time is a genuinely hard combinatorial problem, which is what a solver is for and a spreadsheet is not.
"Before 9, insulin" or "not Tuesdays or Thursdays, dialysis" has to be a field the schedule reads, not a fact one clinician happens to remember — and the day gets planned without it the one time that person is out.
A visit that must land in a 45-minute pocket at 7am is more expensive to schedule than one that can go anywhere between 10 and 4, and the plan should show that rather than treat every visit as equally flexible.
When a day cannot hold every window, the honest answer names which visit did not fit and why, so a person can decide — rather than software quietly picking one to run late.
A single hard window rarely breaks a day by itself. Three or four of them, spread across a caseload also being driven to in some order, is what turns a route into a puzzle — and the driving between them is not free. The Windshield Tax calculator is a quick way to see what a week of that driving actually costs before you decide how much slack to build around each window.
Obillions is in private testing on iPhone, with a public TestFlight next. If time windows are what makes your week hard to plan, see what a schedule that treats drive time and windows as one problem looks like, or read about what happens when one of those windows falls through anyway.
A constraint on when a visit can happen beyond just the date — before breakfast for insulin, not on dialysis days, within a wound's redressing interval, or when a caregiver is actually present. Unlike a delivery window, missing it is often a clinical or logistical failure, not just an inconvenience.
The patient is not home on treatment days, and on the days around treatment may only be home for a narrow stretch. That removes the patient from the map entirely on some dates and compresses the visit into a small pocket on others, which a schedule has to track per patient rather than assume.
By showing, before the day is driven, when a wound-care visit is on track to land outside its interval — not after the fact on chart review. A visit that lands a little late each week and is never flagged will drift past the ordered interval without anyone deciding that.
No. A visit with a 45-minute window is more constrained than one that can happen anytime in a six-hour span, and a schedule that treats them as equally flexible is guessing on the tight ones rather than planning for them.
The honest answer names which visit did not fit and why — too many tight windows, not enough hours between them — so a person can decide what to do, rather than software quietly picking one to run late.
No, and treating them as the same is how a patient gets told a time nobody actually confirmed. A window being satisfiable by the route is a solver's answer; a patient agreeing to it is a separate step that should stay separate.
A voice agent for a day of visits. You say what changed and the day re-plans itself: the order, the drive times, and who nearby could take the hour a cancellation just freed. Nothing changes until you confirm, and your voice and photos never leave your phone.
In private testing now. A public TestFlight comes next, then the App Store.