Using Salary Confidential to identity a specific pay gap (Gender, Ethnic group etc)
Salary Confidential is built around two layers:
Polls — the overall container for your compensation search (e.g., “My next marketing role”).
Survey peer groups — the specific peer group(s) you choose to approach.
If you’re new to these concepts, see our FAQ item about how polls and survey peer groups work together
If you belong to an underrepresented group and suspect there’s a pay gap affecting you, you can use this Poll/Survey peer group structure to measure that gap directly in your own context.
A concrete example
Let’s say you’re a woman working as a Whatchamacallit in the Foo industry. You’ve identified a solid set of peers with similar backgrounds and roles.
You could:
- Create a new poll (“My next Whatchamacallit role”)
- Run one survey to all peers at once…
…but if you suspect men and women may be paid differently, you won’t be able to tell which data points belong to which group.
Instead, split your outreach strategically:
- Create two survey peer groups within the same poll — one targeting women, one targeting men.
Decide how many data points you want (e.g., eight total). Because surveys cost scales linearly with the number of responses, it costs the same to run:
- 1 survey peer group of 8 responses
- or 2 survey peer groups of 4 responses each
So you’re not penalized for splitting by subgroup.
When both survey peer groups are complete, you’ll receive:
- A survey report for each peer group ("Women watchmacallits" ; "Men watchmacallits")
- A poll-level view combining every result across both survey peer groups
The roll-up view still shows the full picture -- exactly what you'd have gotten from your big survey of 8 data points --, but the split approach gives you clean, unambiguous subgroup data: how much men with your role earn in your industry, and how much women earn.
Should you tell respondents you’re running differentiated survey peer group along sensitive characteristics?
That’s entirely up to you.
Each survey has a Survey Message — a short note that peers see when they land on the survey page. You can use that space to explain who you’re asking and why, including any demographic filters. Or you can choose not to mention it. Salary Confidential is agnostic, and we have no visibility into who you invite.
Sharing results across peer groups
If you’re running multiple surveys (e.g., a women peer survey and a men peer survey) and you want one group to see the other group’s outcomes, you can do that in the Results Message.
Respondents who complete a survey receive a secret key granting access to that survey’s results. That's the product promise of Salary Confidential: a respondent gets access to the data set they contribute to. But we consider that any other data access is going to be at the discretion of the Requester who owns the data.
In order to enable this, we have the Results Message box, which is a customizable note your verified respondents (with their keys) see inside their results view.
In this case, you can use this message box to paste your requester unlock link for another survey directly into that message. You can do this immediately when you first create the survey because, as a requester, we give you your future results view link (with your secret code) the moment the survey is created -- before the survey closes. You'll find your results report link under "Advanced" in the Survey tools in your dashboard
Example: If you’re running a women peer survey and want your women respondents to also see the men peer results, you can place a note in the Results Message of the women peer survey. It would include your own requester results link for the men peer survey group results report (with your requester unlock key in it). As the respondents of the women peer group surveys visit the results report for their survey (for which they got their own private keys when they participated), they will see your Result message there where you're sharing with them the results of this different survey peer group that they didn't participate in, but that you want to make available to them.
This lets you share cross-survey insights with any peer group you want — beyond the strict 'one-for-one' nature of the value exchange of the survey results access
Remember: Results keys (requesters' or respondents') only unlock a specific results report. There is no danger with sharing these keys if that's what you want to do: each key only unlocks one specific survey report and never anything else
Small samples and statistical significance
Salary Confidential surveys are intentionally small and highly targeted. That’s what makes them useful — but it also means the results are, by definition, anecdotal.
The data you collect is real. The patterns you observe may be highly systemic, or highly variable and it's not necessarily obvious which way your specific survey is leaning because sample bias is hard to detect by definition (In the example of our women peer group vs men peer group, there may be other characteristics within the participants that will affect the outcome in ways you can't detect but aren't strictly about their gender). So use these small surveys as pointed illustration in a phenomenon you are investigating, but remember that these could not hold out to any test of proof of discrimination, nor something you could reasonably expect to "take to court."