Meeterboard

How Meeterboard calculates meeting cost

Transparent methodology: where the numbers come from, the exact formula, and what we deliberately do not claim.

Last updated:

Every dollar on Meeterboard is derived from a simple idea: annual total compensation can be spread evenly across a standard US business year, then divided down to seconds while a meeting runs. We do not pull live payroll data and we do not guess your personal salary. Instead, each attendee is priced from a Levels.fyi level averagefor a company and job family you select.

The core formula

We define one business year as 52 weeks × 40 hours × 3,600 seconds = 7,488,000 seconds.

For an attendee with annual total comp (TC) T:

per_second = T ÷ 7,488,000

Example: TC of $250,000.00 $0.0334 per second, or about $2.00 per minute.

Meeting total at any moment is the sum of each attendee's cost from their join time to now (or leave time), accounting for pause and drop rules below.

Where compensation data comes from

Headline averages come from Levels.fyi, which publishes aggregated compensation for tech roles by company, job family, and level. Meeterboard fetches public salary pages on our servers (not from your browser), parses level bands, and stores:

  • Company slug and display name
  • Job family (e.g. Software Engineer, Product Manager)
  • Level slug and labels (e.g. L5, Senior)
  • Average total comp, plus base, stock, and bonus when available

We refresh cached rows when they are older than seven days. Popular companies may be pre-seeded; the first visitor to a rare company may wait briefly while we ingest data. See our article How Meeterboard uses Levels.fyi data.

Why we snapshot TC when someone joins

When you add an attendee, we copy their level's annual TC into annual_tc_snapshot on that meeting row. Future Levels.fyi updates do not change an in-progress meeting. That keeps the live clock stable and matches how you made decisions when the meeting started.

Meeting states and time rules

Ready → Running → Paused → Ended

  • Ready: meeting created but timer not started; elapsed time and cost are zero.
  • Running: clock advances; per-second costs accumulate.
  • Paused: elapsed and costs freeze at the pause timestamp.
  • Ended: final totals use end time; no further ticking.

Attendee join, drop, and rejoin

  • Join time starts billing for that attendee.
  • Drop sets a leave time; cost stops growing unless they rejoin.
  • Rejoin starts a new interval (drop time in the middle is excluded from cost).
  • Ending a meeting stamps leave times for anyone still present.

Single-company meetings

All attendees in a meeting use pay bands from the host company'scached Levels.fyi data. You cannot mix Google levels with Amazon levels in one clock today. That constraint keeps search and UI manageable and avoids ambiguous cross-company level mapping.

Spectate and leaderboard

Spectate shows the highest-cost live meeting with anonymized animal aliases instead of custom names. Leaderboard ranks ended meetings by total cost for live, daily, weekly, monthly, and all-time windows. Live leaderboard rows are server snapshots; the homepage teaser and spectate page tick client-side once per second.

Identity without accounts

We assign a random UUID cookie (mc_uid) so meetings persist on your browser. No password. You can copy your ID from Profile and restore it elsewhere. Details in our Privacy Policy.

Limitations (read before trusting the number)

  • Averages, not individuals: Levels.fyi bands smooth across many submissions; your coworker may earn far above or below the average.
  • US business year: 52 × 40 hours ignores PTO, overtime, and non-US norms (e.g. 37.5-hour weeks).
  • Employer cost vs. employee comp: we show loaded labor at TC, not fully loaded employer cost (benefits, facilities, taxes).
  • Equity volatility: stock components in TC can swing with market prices; Levels.fyi snapshots reflect their methodology at fetch time.
  • Coverage gaps: some companies and families have sparse data; we surface what Levels.fyi returns.

Further reading