Our salary data methodology

Sources

Three inputs, weighted in this order: (1) signed offer letters and placement records from our own pre-sales searches; (2) employer briefs and approved bands supplied during live searches; (3) candidate-reported current compensation, verified against offer documentation where available. Self-reported figures with no corroboration are excluded.

Sample and window

1,749 verified US data points, collected between January 1, 2026 and August 8, 2026. Roles are US-based only; equity, sign-on bonuses and benefits are excluded from both base and OTE.

Publication rules

Every figure on this site is one record in a single compensation dataset, and each record carries its own sample size, collection window and source class. Templates read those records directly, so a page cannot type a number that contradicts them. Three rules decide whether a record is allowed to appear:

• Published. Our own median, from a cut with at least 30 verified data points. The cut’s own n is shown next to it, alongside the dataset it was drawn from.
• Small sample. A cut below 30 is withheld and reads “not yet published”. One exception is shown with an explicit small-sample label rather than hidden, marked with a dagger wherever it appears.
• Modelled. Where we hold no separate sample for a cut, we may scale the national figure by that cut’s own base index. Modelled figures are labelled as modelled in the copy and in the structured data, carry no sample size, and are never presented as sourced medians.

Coverage: the sample behind each cut

A national n tells you little about a metro or a sector page, so each cut publishes its own. These are the cuts where we currently hold a sample of our own.

What this data does not cover

Non-US roles, contract and day-rate engagements, equity value, sign-on and retention bonuses, benefits, and any cut with fewer than 30 data points. Individual metro and sector cuts are drawn from the same sample, so a narrow cut carries a smaller n than the headline figure. Third-party figures shown alongside ours on each page are attributed and linked; we neither clean nor restate them.

Definitions

Cleaning

Duplicate offers to the same candidate are collapsed. Roles mis-titled against their actual scope (a delivery consultant advertised as a pre-sales consultant, for example) are re-mapped to the function performed, not the title advertised. Segments with fewer than 30 data points are not published.

Update cadence

Quarterly. Each refresh restates the full rolling 12-month window rather than appending, so figures can move down as well as up. The last update date appears at the top of every page.

Corrections

If a published figure doesn't match what you're seeing in your own hiring, tell us — we review challenges against the underlying sample and publish corrections on this page.