Methodology

How Tippd calculates pay

Every number on Tippd comes from hospitality workers reporting their own shifts. Nothing is scraped from job ads, modelled, estimated, or filled in when it is missing. If we do not have the data, the page says so instead of showing a number.

The unit: one venue, one role, one shift

A report covers a single shift at a single venue: the role worked, hours worked, tips taken home after tip-out, and the hourly wage when the worker knows it. Roles are never mixed — a bartender and a busser never appear inside the same figure, because they are different jobs with different pay.

What we ask for

Venue, role, hours, tips after tip-out, hourly wage if known, the day of the week, and the season and year. We deliberately do not ask for an exact date, a name, an email, or an employee ID. A report should not be traceable to the person who filed it.

Shift total and effective hourly

A shift total is hours × hourly wage + tips taken home. Effective hourly is that total divided by the hours worked. When a report did not include a wage, we show tips only. We do not fill in a tipped minimum and present the result as reported earnings, because half of that number would be ours rather than the worker’s.

Medians, not averages

One New Year’s Eve can drag an average somewhere useless. Typical figures are medians and are written with a leading ~, rounded to the nearest dollar. Ranges show the middle half of reported shifts, so a good night and a slow night stay visible instead of being flattened into one number.

Annual estimates

Where an annual figure appears, it is the typical shift total × 4 shifts a week × 48 weeks a year, rounded, and it only appears once enough wage-inclusive reports sit behind it. It is a scenario, not a salary, and the assumption is printed beside it every time. Hospitality schedules are not steady, so treat it as arithmetic rather than a forecast.

Sample size and publishing thresholds

Sample size appears next to every benchmark, and small samples are labelled as small rather than dressed up. A cut needs at least 5 reports before we publish figures for it at all. A pay page is only listed for search once it carries at least 8 reports across 3 venues; a venue page needs at least 3 reports. Thin pages stay unlisted rather than being padded out.

Freshness

Each page shows when it was last calculated and the reporting periods behind it. Because we record season and year rather than exact dates, the data ages in seasons. Dates move when data moves, never to look fresher than the data is.

What we do not claim

Reports are not currently verified against payroll, and we do not describe them as verified. We do not know how long anyone has worked in the industry, only how long they say they have worked at the venue they are reporting on. We do not publish job listings or salary ranges from employers. Everything on Tippd is worker-reported, and the limits of that are part of the number.

Privacy

No names, no emails, no exact shift dates. Reports carry a season and year. Nothing attached to a report can identify who filed it. More about Tippd.

Frequently asked

Is the data verified?: Not currently. Reports are anonymous and self-reported. We protect the numbers with medians, sample-size thresholds, and by publishing nothing for a cut that is too thin, rather than by claiming a verification we do not perform.
Why medians instead of averages?: Because hospitality income is spiky. One exceptional night moves an average and does not move a median, and the median describes the shift you are more likely to actually work.
Why do some pages not show a number?: Because the cut has too few reports behind it. We would rather show nothing and say we are still collecting than publish a figure built on two shifts.
How is an annual figure calculated?: Typical shift total multiplied by 4 shifts a week and 48 weeks a year. It is a scenario with the assumption printed next to it, not a salary.
Are tips reported before or after tip-out?: After. We ask for what the worker actually took home, because that is the number that matters.