Exception management
Four things need you this morning. The other thirty-seven do not.
Everything that went to plan is already verified and out of your way. What is left is a short list of decisions only a person can make.
The first fifteen minutes of your day
This is the screen you open with your first coffee, and it is where most owners spend their time in CareVerify. It is worth describing exactly.
It is 8:04 on Monday. Forty-one visits happened since Friday afternoon. Thirty-seven of them started roughly on time, at the right address, and ended when they were supposed to. You will never see those. They are verified and they are done.
The four that are left are on one screen. One caregiver has not checked in for an 8am visit and it is now 8:04, so that one is at the top with a phone number next to it. One arrived twenty-two minutes late on Friday. One checked in 140 metres from the client's address. One forgot to check out of a Friday afternoon visit and it is still showing as in progress.
You make one phone call and three decisions. By 8:20 the queue is empty and you have not looked at a single visit that was fine.
How something ends up in the queue
- 1
The system compares what happened to what was scheduled
Every check-in and check-out is measured against the visit it belongs to, as it arrives.
- 2
Anything outside tolerance becomes an exception
Late, missing, out of area, unfinished, overlapping. Each with its own trigger, not one vague 'needs review' flag.
- 3
Urgent ones do not wait for you to look
A missed visit texts the office 30 minutes after the scheduled start. You do not find out about it on Monday.
- 4
You resolve it with a reason
Confirm what happened, correct a time, or mark the visit as not delivered. The reason is stored with the record.
- 5
It leaves the queue and enters the audit trail
Resolved exceptions do not vanish. The original, the correction and the reason all stay attached to the visit.
What raises an exception
Six types, each with a specific trigger. Named exceptions tell you what to do. A generic flag just tells you to go and look.
Late check-in
Triggered when: More than 15 minutes after the scheduled start
Why it matters: A client may be sitting waiting, and that is the call you would rather make than receive
Missed visit
Triggered when: No check-in 30 minutes after the scheduled start
Why it matters: The office is texted immediately. This is the one that cannot wait until morning
Outside the geofence
Triggered when: Location outside your radius, or accuracy too poor to trust
Why it matters: Audit risk. Usually a phone problem, occasionally not, and you want to know which
No check-out
Triggered when: Still in progress an hour past the scheduled end
Why it matters: Someone forgot. Left alone, the hours are wrong and payroll is wrong
Manual edit
Triggered when: Any office correction to a recorded time
Why it matters: Every correction needs a reason and stays in the record, including yours
Overlap
Triggered when: One caregiver checked in at two clients at once
Why it matters: Impossible, so it is either a mistake or something worse. Either way you look at it today
Details that keep the queue worth opening
A queue you can actually empty
If everything raised a flag, you would stop reading it within a week. Thresholds are set so that a normal week produces a handful of items, not forty. An exception queue nobody empties is the same as having no queue.
Exceptions reach payroll before payroll surprises you
Anything unresolved shows up when you run the export for that period. The connection is deliberate: an unresolved visit is somebody's hours, and it should block a clean export rather than quietly reduce a total.
Your resolutions are audited too
Office corrections are recorded with who made them, when, and why. That is not distrust of you. It is what makes the visit record hold up when somebody asks how a time came to be changed.
Missed visits do not wait to be discovered
Thirty minutes after a scheduled start with no check-in, the office gets a text. A client waiting is a problem that gets worse with every hour it stays unnoticed, and that is not something to learn about on Monday.
What this does not do
- It does not decide for you. Nothing is auto-approved and nothing is auto-rejected. Whether a caregiver who checked in 140 metres away was at the right house is a judgment about your client and your staff, and software guessing at it would put a guess into your audit trail.
- It does not call the caregiver. It puts the number in front of you. Automated chasing of people who are usually mid-visit creates more problems than it solves.
- It does not judge whether care was good. It flags mismatches between plan and record. Quality of care is a supervision question, not a data one.
The rest of the product
Visit Verification
GPS and time captured at the door, checked against the schedule, and recorded whether or not the phone cooperates.
Scheduling
A week grid that catches double bookings before they happen and copies repeat visits forward.
Caregiver App
Works with no signal, never blocks a check-in, and lets caregivers see their own verified hours.
Payroll Export
Unresolved visits surface before you export, so nothing is silently dropped from someone's pay.
See what your Monday queue would look like
14 day trial. No card required. The clock starts at your first caregiver check-in.