Clock-In Errors for Remote Cleaners: Run These 7 Checks First

The expensive version of this problem is not the missed punch. It is the supervisor who fixes thirty of them a month by typing numbers into a spreadsheet at 11 p.m. on payroll Sunday, with no note about who authorized what and no record of why.

Eleven months later a former cleaner claims she worked 6 unpaid hours a week at a bank branch, and your entire defense is a spreadsheet that a manager could have edited any time. That is the real cost of treating clock-in errors as an annoyance instead of a process.

When a remote cleaner cannot clock in, work seven checks in order before you touch the timesheet: phone browser and login, location permission, GPS accuracy against the geofence radius, cell signal, the correct location record, the shift window, and the cleaner's own account. Then log a documented manual correction, not a silent edit.

What follows is the triage sequence in cost order. The first three checks take under a minute each and resolve most tickets. The last four cost more time, so you only run them when the cheap ones pass.

Why does GPS clock-in fail for cleaners who are actually on site?

Almost every "the app is broken" ticket is one of four physical realities, and none of them are the cleaner lying.

First, satellite geometry. GPS.gov states that GPS-enabled smartphones are typically accurate to within about a 4.9 m radius under open sky. Under open sky. A cleaner standing in a basement loading dock, a below-grade parking garage, or the interior corridor of a 20-story tower has no open sky at all.

Second, fallback positioning. When a phone browser cannot get a satellite fix, the browser geolocation service falls back to Wi-Fi network positioning or IP address positioning. IP-based coordinates can land in the wrong city entirely, which is why a punch sometimes shows up 40 miles from the building.

Third, permissions. A browser location prompt that got dismissed once will keep failing silently until someone resets it in the phone's settings.

Fourth, the pin. If the coordinates on your location record were geocoded from a mailing address, they may sit at a street centroid or at the leasing office of a three-building campus, not at the door your crew actually uses.

Error signatures and what they usually mean

Work these 7 checks in order, cheapest first

Check 1: Was it one cleaner or the whole building?

Pull the punch log for that site and that night before you call anybody. This one question splits your entire diagnostic tree in half and takes about 60 seconds.

Pass: other cleaners at the same site clocked in normally. The problem lives on one person's phone or one person's account, so skip to Check 2.

Fail: nobody at that site got in. Stop chasing the cleaner. Go straight to Check 6 and audit the location record, because a bad pin or a radius that is too tight will fail every person who walks through that door, every night, until you fix it.

Check 2: Is the cleaner on the right link, the right login, and a real browser session?

Have them screenshot what they see. Half of "it won't let me in" tickets are a stale bookmark, a shared login someone else changed, or private browsing mode wiping the saved session and the location permission every single time.

Pass: they reach the clock-in screen, see their own name, and see their assigned shift. The session is fine.

Fail: they see a login wall, a blank page, or someone else's name. Fix the account, not the geofence. Two specific culprits worth naming: incognito or private tabs, and a cleaner using a family member's phone that is still logged in as a different employee.

Check 3: Is location permission actually granted, with precise location on?

This is the single highest-yield check on the list and it costs one phone call. On iPhone, walk them to Settings, then Privacy and Security, then Location Services, then their browser, and confirm it is set to "While Using the App" with Precise Location switched on. On Android, check both the system location setting and the browser's site permission for your clock-in page.

Pass: permission is granted, precise location is on, and Low Power Mode or battery saver is off. Move to Check 4.

Fail: permission is denied, or precise location is off, or the phone is in battery saver. Precise location off is the sneaky one, because the browser still returns a position, just a deliberately fuzzy one that can miss a 150 m radius by a wide margin. If they use a VPN, have them turn it off for clock-in, since VPN traffic pushes browser positioning toward the IP address of a server in another state.

Check 4: Did they have a data connection where they were standing?

Ask one question: where exactly were you when you tapped the button? Freight elevators, sub-basements, mechanical rooms, interior stairwells, and concrete parking decks kill both the satellite fix and the cellular upload.

Pass: they had two or more bars at the punch spot and other apps loaded normally. The connection is not your problem.

Fail: they were underground or deep in the building core. The fix is a documented punch spot, not a technology change. Pick a specific place at each site where a fix is reliable, usually the main entrance, the dock door, or the vestibule, and write it into the site instructions so it is the same for the regular cleaner and for any sub covering the route.

Check 5: Was the recorded position inside the radius, and by how much?

Now look at the actual numbers on the failed attempt. You want the recorded coordinates and the distance from the pin, because 40 m outside a 150 m radius and 4,000 m outside it are two completely different problems.

Pass: the position sits inside the radius, or misses it by a small margin consistent with an indoor fix. Widen the radius for that location and move on.

Fail: the position is off by kilometers. That is fallback positioning, not travel. Go back to Check 3. Worked example: Meridian Plaza is a three-building campus. Your location pin sits at the property entrance, but the crew works Building C, roughly 210 m away across the parking field. At a 150 m radius every punch fails by about 60 m, forever. The fix is a separate location record per building, not a 400 m radius that would also cover the neighboring restaurant.

Check 6: Are the coordinates on the location record correct?

Open the site in a satellite map view and confirm the pin sits on the building your crew cleans, ideally on the entrance they use. Addresses imported in bulk from a CSV are the usual offender, because a geocoder will happily place a suite in a business park at the park's main road.

Pass: the pin lands on the right roof and the radius circle covers the doors and the crew's parking area without swallowing a neighboring tenant.

Fail: the pin is on a road, on the wrong building, or on a leasing office two blocks away. Correct it from the satellite view and re-verify with the crew's next real punch, not with your own phone from the parking lot at noon.

Check 7: Was there a shift to clock into?

Plenty of "rejected" punches are correct rejections. A sub covering a route may have no assignment on that location, a one-off work order may never have been created, or the cleaner may be 40 minutes early because they finished the previous stop fast.

Pass: a shift exists, on the right location, on the right date, and the attempt fell inside the allowed early window.

Fail: no assignment exists, or the attempt was outside the window. Fix the schedule, then decide whether your early window is realistic. A 15 minute pre-shift window is fine for a fixed-site crew. Route cleaners hitting five small accounts in one night need more room, or you will manufacture exceptions every single night.

How do I fix a missed clock-in without creating a wage-and-hour problem?

An unrecorded shift still has to be paid. The Fair Labor Standards Act obligates you to keep accurate records of hours worked, and the Department of Labor's Wage and Hour Division specifies retention periods: payroll records for three years, and the time cards and work schedules on which wage computations are based for two years.

So the correction itself is the record. Treat every manual edit as an audit entry with five fields: who reported it, the claimed in and out times, the reason code, the corroborating evidence, and who approved it.

1. The cleaner reports the failure the same night, in writing, in a channel you keep. Not a text to a supervisor's personal phone.
2. The supervisor names a reason code from a short fixed list: permission denied, no signal, outside radius, no shift assigned, phone dead, forgot.
3. Attach corroboration. A dated inspection photo, a supply request timestamp, a message sent from the site, or a client sign-in sheet all work.
4. A second person approves the edit. Never let the person who reports the hours be the person who approves them.
5. The cleaner confirms the corrected total before payroll runs.

On rounding: 29 CFR 785.48(b) permits rounding recorded time to the nearest 5 minutes, tenth of an hour, or quarter hour, provided the practice does not, over time, fail to compensate employees for all time actually worked. Neutral means it cuts both ways. Rounding that only ever rounds down is the version that gets litigated.

Check your state before you lean on "it was only a few minutes." In Troester v. Starbucks (2018), the California Supreme Court declined to apply the federal de minimis doctrine to small increments of regularly occurring off-the-clock work under state law. Small and routine is exactly the pattern a missed-punch habit creates.

How wide should the geofence radius be?

A radius is a tradeoff between false rejections and useless precision. Too tight and you generate exceptions every night at an indoor site. Too loose and the punch proves nothing more than "somewhere in this shopping center."

These are operational starting points, not a standard. Set them per site, then adjust based on your actual exception log after the first two weeks.

Is it a clock-in error or is someone gaming it?

You cannot tell from one incident, and you should not try. You tell from the shape of the pattern over four to six weeks.

Genuine technical failures cluster by place and by device: the same building, the same phone model, the same corridor. They happen at clock-out as often as clock-in. Convenient failures cluster by person and by time: the same cleaner, always at the front end of the shift, always with a claimed start time that is exactly the scheduled start time.

Three questions separate them. Does this person's failure rate exceed everyone else's at the same site? Do their failures happen only on the in punch and never the out punch? Do their claimed times land on the schedule to the minute, when successful punches never do?

If all three point the same way, that is a supervision conversation with documented evidence, not a technology purchase. Be honest about what geofenced browser clock-in can and cannot prove: it records a position at the moment of the punch. It does not follow anyone between punches, and it does not detect a spoofed location. Presence for the whole shift is proven by inspection scores and photo evidence, not by a single coordinate.

Frequently asked questions

Do I have to pay a cleaner who forgot to clock in?

Yes, if they worked. The FLSA requires you to pay for all hours worked and to keep accurate records of them, and a failed punch does not undo the work. You can discipline a repeated failure to follow the timekeeping procedure, but withholding pay as the penalty is how you turn a $40 problem into a Department of Labor complaint.

How long should I keep GPS clock-in records?

Follow the FLSA retention floors as your minimum: three years for payroll records, two years for the time records and schedules that wage computations rest on. Many operators keep timekeeping data longer, because state wage claim windows and client audit requests can reach back further. Make sure your export path works before you need it, not during a dispute.

Can a cleaner clock in from home if the geofence is wide?

If your radius reaches their apartment, yes. That is the main argument against solving every exception by widening the circle. Size the radius to the property, split multi-building campuses into separate records, and handle indoor sites with a documented punch spot at grade level instead of a 500 m geofence.

What if the cleaner's phone is too old to run browser clock-in?

Confirm it is really the phone before you buy anything. Most "old phone" tickets turn out to be a denied location permission or a browser that has not been updated in two years. If the device genuinely cannot get a fix, options are a shared site device kept in the janitor closet, a supervisor punching the crew in from their own phone at the door with a written policy, or a documented paper exception log for that route.

How many clock-in exceptions per week is normal?

There is no published benchmark, so set your own baseline. Count exceptions for two weeks, then track the ratio of exceptions to total shifts. What matters is the trend and the concentration: a steady rate spread across many sites is friction, while a rising rate at one site or from one person is a specific problem with a specific fix.

Where CleanTrack360 fits

CleanTrack360 runs geofenced GPS clock-in and clock-out in the phone browser, so there is no app install to chase on a mixed fleet of personal devices. The native app is still in development. The default geofence radius is 150 m and it is configurable per location, which is what lets you tighten a strip mall unit and widen a warehouse yard instead of applying one number everywhere. Locations can be imported in bulk by CSV, so if you are auditing pins after reading Check 6, you can reload corrected coordinates in one pass. Reports export to CSV for payroll and for the retention requirements above.

Be clear about the limits, because they matter to your policy: position is captured at clock-in and clock-out only, there is no breadcrumb tracking between punches, and there is no GPS spoofing detection. Coverage for the rest of the shift comes from inspections with custom checklists, photo evidence and automatic scoring, plus location-based team messaging channels where cleaners can report an exception the same night in a place you keep. Plans are Starter at $99 per month for up to 5 team members, Pro at $199 for up to 20, and Business at $249 for up to 50, priced per plan rather than per user, with a 14-day free trial and no credit card required.