Exporting a Payroll CSV for Part-Time Tutors

It is the Monday your payroll run is due, and you have eleven part-time tutors who worked scattered hours across two weeks. Some came in for a single 90-minute session, some covered three afternoons, one forgot to clock out on a Thursday. Your job is to hand the bookkeeper one file that adds up correctly the first time. Here is how to produce that file from start to finish.

What payroll actually needs in the file

Before you export anything, find out what your payroll person or service imports. Most want a flat CSV with one row per worker or one row per entry, and these columns:

• Worker name (and an ID if your payroll system keys on one)
• Date of the work
• Clock-in time and clock-out time
• Total hours, usually as a decimal like 2.50 rather than 2:30
• Sometimes a pay period label or a group/role column

The number that matters most is total hours per tutor for the pay period. Everything else is supporting detail so a disputed line can be checked. If payroll wants decimal hours and your timesheet shows 2:30, run those through a decimal and HH:MM converter so you are not doing the math in your head at 5pm.

Step 1: Pin down the pay period dates

Decide the exact start and end of the period before you pull data, because an off-by-one day is the most common reason a file gets bounced back. If you run a biweekly pay period, write down both dates, for example March 3 through March 16 inclusive. A tutor who worked the evening of the 16th should appear; one who worked the morning of the 17th should not.

Kangaroo Clock stores every entry in UTC and shows it in your local time, so a late session that crosses midnight is recorded at the real wall-clock time you saw, not shifted by a server somewhere. Confirm your dashboard is showing your timezone before you trust the boundaries.

Step 2: Close out forgotten clock-outs

You will almost always have at least one open entry where a tutor left without tapping out. Fix these before exporting, because an open entry has no total and will either drop the row or, worse, count from a stale start time to now.

Kangaroo Clock handles this with auto-close for stale entries. A forgotten clock-out is closed at its start time plus a cutoff you set, never at the current moment. So a tutor who clocked in at 4:00pm for a session that should have run two hours does not show up with sixteen hours because nobody tapped out until the next morning. The reported number stays honest, and you can adjust the single entry if you know the real end time.

Walk your open entries one at a time. For each, either set the correct clock-out from what you know, or let the cutoff apply. Do not leave any entry open when you export.

Step 3: Filter to the people and program you are paying

If your center pays SAT prep tutors on a different line than your elementary reading tutors, separate them now. Worker groups label people by role or program, and tags label the time entries themselves. Filter the report to the group you are paying so the CSV contains only those rows. Do the same per program if your grant or budget tracking needs hours split out. This is the same setup described in the guide to tracking tutor sessions for payroll, and getting groups right once saves you the filtering work every period after.

Step 4: Export the CSV

Run the export for your date range. Kangaroo Clock gives you a CSV with a stable column schema, meaning the columns appear in the same order with the same headers every single period. That matters more than it sounds. If your payroll import maps column 4 to total hours, it will still be column 4 next month, so an import template you build once keeps working. The export also includes a distinct-worker count, which is a fast sanity check: if you paid eleven tutors and the file says nine, two are missing and you know to look before you submit.

The full workflow for getting tutor hours into payroll via CSV is built around this export, with no per-seat fees no matter how many tutors clock in.

Step 5: Check the file before you send it

Open the CSV and do three checks:

1. Row count vs. people count. Match the distinct-worker number against your own list of who worked.
2. Outliers. Scan the total-hours column for anything implausible. A 14-hour entry for a part-timer usually means a clock-out that did not get fixed in Step 2.
3. Zero or blank totals. Any blank total is an entry that is still open. Go back and close it.

If you would rather verify a tricky multi-session day by hand, a time card calculator lets you add the segments and compare against the file. Catching one bad row here is far cheaper than a correction run after payroll has processed.

Step 6: Hand it off and keep a copy

Send the CSV to whoever runs payroll, and keep the same file. Kangaroo Clock does not move or hold money and is not a payroll processor, so the CSV is the handoff point between your hours and their system. Your copy is your record if a tutor questions a paycheck three weeks later. The export shows clock-in, clock-out, and the math between them per entry, so you can answer the question without reconstructing anything.

Making the next export take five minutes

The work above is front-loaded the first time. Once your worker groups and tags match how you pay, each future period is filter, close any open entries, export, spot-check, send. Workers never sign in and never install an app; they tap their name on a shared tablet, scan a posted QR code, or open a personal link. You can see that clock-in flow in the live kiosk demo before you set up a single tutor. Get the dates and groups right once and the CSV stops being a Monday-morning problem.