Free, no obligation
Send me last month's timecards.
Every vendor in this category will tell you their overtime is correct. Rather than ask you to believe ours, run it against hours you already paid. Send last month's timecards and I will put them through the same engine our payroll exports use, then send back a written comparison of where the numbers differ and which rule caused each gap.
Three business days
A written comparison, or a straight answer that we could not parse your file. One or the other, inside three days.
Nothing is stored
The analysis runs in memory and the rows are gone when the answer comes back. There is no table for them to land in. Anonymise the names if you like - the arithmetic is identical.
We will tell you if you are right
The most likely outcome outside California is that your numbers already agree. You will hear that plainly, and I will not follow up.
Where the numbers usually differ
Four places, in roughly this order. None of them are exotic - they are the ordinary consequences of tracking a weekly total.
Daily overtime that a weekly total cannot see
Four twelve-hour days is 48 hours. A weekly calculation finds 8 hours of overtime. California finds 16, because each day past 8 counts on its own - and the weekly rule then adds nothing, because only 32 hours are still straight time. Three eleven-hour days is 33 hours: weekly finds nothing at all, the rules find 9 hours of overtime.
Rounding
Five days of 8h05m rounded down to 8h00m is 25 minutes a week that somebody worked and nobody paid for. 29 CFR 785.48(b) permits rounding punch times, not a day's total.
Overtime counted twice
The opposite error, and it costs you rather than the worker: hours already paid as daily overtime being counted again toward the weekly 40. Labor Code 510(a) forbids it in as many words.
The seventh consecutive day
First 8 hours overtime, everything past 8 at double - and keyed to your workweek rather than to Sunday, which is where most systems get it wrong even when they implement it.
Three times I'll tell you not to bother.
No vendor publishes this section. It is here for the same reason the rest of this site publishes the gaps in our own compliance engine: the only way to be worth believing on the hard claims is to be honest about the easy ones.
Everyone works a straight five-day week outside California
If nobody exceeds eight hours in a day and your state has no daily overtime rule, a weekly total is the correct answer and we will find nothing. That is most of the country.
What I'll do instead: tell you that in one line, and not follow up.
You already run a compliance-grade payroll service
If ADP, Gusto or a payroll bureau is already applying state daily-overtime rules to your hours, they are doing the part we do. The question for you is whether the hours going in are measured or remembered - which is a different product conversation.
What I'll do instead: point you at the part that might actually be broken, which is usually the timesheet, not the arithmetic.
You need this settled today
Three business days is a real constraint of a one-person operation, not a negotiating position. If you are mid-dispute or mid-audit, that is too slow to be useful.
What I'll do instead: give you the free calculator and the statute pages so you can do it yourself this afternoon.
Common questions
What do I need to send?+
Last month's timecards as a CSV or spreadsheet export, with an employee name, a date and hours per day. A pay rate column turns the answer into dollars rather than hours, but it is optional. If your system exports something odd, send it anyway - if we cannot parse it we will say so rather than guess.
What do I get back?+
A written comparison within three business days: your hours run through our wage-and-hour engine, next to what a weekly-only calculation produces from the same hours, per employee. Every difference is traced to the rule that caused it and the statute it comes from. If we cannot parse your file we will tell you that instead, inside the same three days.
What happens to my data?+
It is never stored. The analysis runs in memory and the rows are gone when the answer comes back - there is no table in our database for them to land in, which is a design decision rather than a policy promise. We do not need real names either: replace them with Tech 1, Tech 2 and the arithmetic is identical.
What if my numbers are already right?+
Then we tell you that and you keep your money. It is the most likely outcome if you are outside California and everyone works a normal five-day week, and it is the reason the offer is worth making at all. An analysis that always finds a problem is a sales pitch.
Is this a trick to get my employee data?+
It would be a poor one, given we tell you to anonymise the names and store nothing. What we get out of it is a conversation with someone who has seen our arithmetic against their own payroll, which is a better sales pitch than anything we could write.
Who actually does this?+
Me, personally - Punchless is a small operation and this is a founder-run process, not a queue. That is also why the turnaround is three business days rather than three minutes, and why there is a limit to how many of these I can do in a week.
Every rule used in the comparison is published with the statute behind it and the test that checks it, including the gaps we have not closed. Punchless is not a law firm and this is not legal advice - it is arithmetic against published rules, which is a narrower and more useful thing.