Method
How to plan a home health route by hand, and when to stop.
Done well, a hand-planned day is close to what software would produce. Done the usual way, it costs an hour and a tank a week. The method, step by step, and the threshold where it stops being worth it.
Most home health routes are still planned by hand, on paper or in the head, the night before or in the driveway. Done well, a hand-planned day is close to what software would produce. Done the usual way, in the order the visits were entered, it costs an hour and a tank of fuel a week. This is the method, step by step, and the point at which it stops being worth doing by hand.
Before you start: three facts per patient
A route is only as good as what you know about each stop. For every visit on the day, write down:
- The window. Not "morning" but the real constraint: insulin before 8:30, wound care after the aide has been, home from dialysis by 3. Most patients have no hard window; the two or three who do decide the whole shape of the day.
- The length. A routine visit and an admission are not the same block. Be honest about the admission; it is the one you will be late from.
- Whether the time is agreed. A time you have already told the patient is fixed. A time you are proposing is movable. Mark them differently on the page or you will move one you promised.
Step 1: pin what cannot move
Put the agreed times and the hard windows on the timeline first. These are the skeleton. If two of them collide, that is a phone call, not a routing problem, and it is better made the night before than at the door.
Step 2: cluster by neighbourhood, not by list order
Look at the remaining addresses on a map, not a list. Group the ones within a few minutes of each other. The single largest saving in hand planning comes from refusing to cross town twice: if two patients live near each other, they are seen back to back, whichever order the referrals arrived in.
Step 3: thread the clusters through the pins
Now place the clusters between the fixed points so that the day moves in one direction where it can. Start far and work home, or start near and end far, but avoid the zigzag. A day that goes east, west, east again has a drive in the middle that produces nothing.
Step 4: put lunch where the map says, not where the clock says
A break at noon on the far side of town is a break you will skip. Put it where there is a natural gap between clusters, inside the window you are actually willing to eat in.
Step 5: add the drive times and check the arithmetic
Estimate each leg. In town, twenty-five to thirty miles an hour door to door is realistic once parking and finding the house are counted; on rural roads, more. Add the visit lengths. If the day ends after the hour you are willing to end, something has to come out, and it is far better to know that at 9pm than at 4pm.
Step 6: decide, in advance, what falls out
This is the step almost nobody does by hand, and it is the one that makes the difference on a bad day. Which visit moves to tomorrow if the admission runs long? Which one can be given to a colleague? Write the answer on the plan. When the day breaks, you execute a decision instead of making one at a red light.
What this gets you, and what it does not
| Hand planning handles | Hand planning cannot handle |
|---|---|
| Five or six stops with one or two windows | Eight stops with four windows: the number of valid orders is in the thousands |
| A plan made the night before | Re-planning at 9:05 when the first patient cancels and you are already driving |
| One clinician's day | A week across a team, where who gets which patient decides the driving before anyone leaves the house |
| Estimating drive times you know | Drive times on roads you do not know, which is every travel nurse's first month |
When to stop doing it by hand
The honest threshold is not the number of patients; it is the number of times the plan changes. A stable caseload of five with no windows can be planned on a napkin for years. A caseload that cancels twice a week, with admissions that run long and windows that move, breaks the napkin every other day, and each break costs the miles this page started with.
That is the case we built Obillions for: the day is ordered by a real solver within the windows, and when it breaks you say what changed and it re-plans, including who nearby could take the hour. If you want to know what your current order is costing, the Windshield Tax calculator is free and takes a minute.
Questions people actually ask
How do I plan a home health route by hand?
Write each patient's time window, visit length and whether the time is agreed. Pin what cannot move, cluster the rest by neighbourhood on a map, thread the clusters between the pins so the day moves in one direction, place lunch in a natural gap, add drive times, and decide in advance which visit falls out if the day runs long.
What is the biggest mistake in hand-planned routes?
Driving in the order the visits were entered. Two patients who live near each other should be seen back to back whatever order the referrals arrived in; crossing town twice is where most of the wasted miles come from.
How long should I allow between visits?
In town, twenty-five to thirty miles an hour door to door is realistic once parking and finding the house are counted. On rural roads allow more. Underestimating this is why hand-planned days end late.
What do I do when a patient cancels mid-morning?
Have the answer ready before the day starts: who nearby is due a visit this week and could take the hour. A cancellation handled from the car with no list is usually an hour lost.
When is a route too complex to plan by hand?
It is less about patient count than about change. Eight stops with several time windows have thousands of valid orders, and a day that is re-planned two or three times a week is past what paper can keep up with.
Does Obillions replace hand planning?
It does the ordering within time windows and the re-planning when the day breaks, by voice. The judgement about windows, agreed times and which visit can give way is still yours; the app asks before it commits to anything.
Obillions
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.