Tools - Team Benchmark
Benchmarking one role is easy. Knowing whether your whole team is priced to keep is the harder question, and it is usually the one that gets answered after somebody resigns. Enter the team, see who is below band, what it would cost to fix, and where the pressure is.
Your team's pay never leaves this page. Every figure below is worked out by JavaScript running on your own machine, against bands already published on this site. Nothing you type is sent to us, nothing is saved, and closing the tab discards all of it. There is no email gate on the result, because we would rather you used this honestly than gave us an address.
| Name or initials | Discipline | Level | City | Base (thousands) |
|---|
Copied. It is plain text, so it will paste into anything.
The team against market
-
Mean position in band
Person by person
Add an engineer to see the read.
If this says what you suspected it would, the useful next step is a market read on the roles you are worried about, not a spreadsheet.
Talk to us about your team →Methodology
Each engineer is placed against the band for their discipline, level and city: the 25th to 75th percentile of accepted offers from active Re:Sourced searches in 2026, the same dataset behind the Salary Checker, the Salary Index and the Salary Guide. Position in band is where they fall between the two ends, so 0 per cent is the floor and 100 per cent is the ceiling. Someone can sit below 0 or above 100, and the tool says so rather than clamping it.
Cost to fix is given twice, because they answer different questions. To the floor is what it takes to stop anyone being under-market, which is the number that stops someone leaving. To the midpoint is what it takes to put the team on the market's middle, which is the number that stops them looking. Both are annual base, before on-costs.
Fully loaded payroll adds the statutory employer costs for each person's tax region, the same figures behind the Cost-to-Hire calculator. Because it is keyed by region rather than currency, a team split across Sydney and Melbourne is loaded at each state's payroll tax rather than a blend.
Nothing here converts between currencies. A team is benchmarked within a single currency, and a city in another one is left out of the totals rather than added at a rate we do not hold. A multi-currency team is two reads, not one.
The risk tiers are blunt on purpose, and every flag states the rule that produced it, because somebody is going to ask why it says that about a named person. This sees base salary and nothing else. People leave over manager, mission, progression and equity too. A flag means pay is one of the reasons available to them, which is a reason to have the conversation, not a prediction that they will go.
Common questions
No. Every figure is calculated on this page by JavaScript running on your own machine. Nothing you type is sent to Re:Sourced or to any server, there is no save, and closing the tab discards it. The page itself loads fonts and analytics like any other page, but no pay figure is included in that.
Three tiers, all based on base salary against the published band. More than 10 per cent below the band floor is high: that is a gap a competing offer closes in one move. Below the floor at all is elevated. Inside the band but in the bottom quarter is worth watching. Everything else is not flagged.
You can mix cities that share a currency, such as Sydney, Melbourne and Brisbane, or New York and San Francisco. You cannot mix currencies, because adding them up would need an exchange rate and we deliberately hold none. Benchmark a multi-currency team as one read per currency.
No, and this tool does not claim it is. People leave over manager, mission, progression and equity as much as base. What a band position tells you is whether pay is one of the reasons available to them. Treat a flag as a prompt to have the conversation, not as a prediction.