Benchmarks that need both people to agree.
Compare against a peer, an anonymised cohort or a role profile. Off by default, symmetric when on, and revocable without anybody being told.
What it shows
Sample figures against a cohort. The total is almost never the interesting part — the two gaps are.
A person, a cohort, or a role
They answer different questions, and the third one is the one we would point most people at.
Against a person you connected with
Both of you agree, both of you see the same view, and either of you can end it. A one-way comparison is surveillance; a mutual one is a study partnership.
Two people on the same path finding out that one is stronger on framing and the other on evidence — and then swapping memos before they submit.
Against a cohort
Your stage, your institution or your track, as an anonymised distribution. You see where you sit in the spread; you never see who anybody else is.
A cohort view that named people would be a leaderboard with extra steps, and nobody would opt into it twice.
Against a role
The comparison most people find most useful, because the benchmark is not a person at all — it is what a role demands, which cannot be discouraged by somebody else having a good month.
If comparison makes you anxious rather than useful, use this one and turn the other two off. It answers the question that actually matters.
Three rules, no exceptions
Comparison is the feature most easily turned into something unpleasant. These are the constraints that stop it.
Off until you turn it on
Comparison is disabled on a new account. Nobody can compare against you, and no view exists for them to open.
Symmetric or nothing
If you can see somebody, they can see you, in exactly the same detail. There is no one-way mode and there will not be.
Revocable, silently
End a comparison at any point. No notification is sent, and the historical view disappears for both sides.
Connection to comparison
Five steps, two of which are consent gates.
Connect first
Comparison is a thing between connections. You cannot compare against a stranger, and a connection request can be declined silently.
Both opt in
One of you proposes comparison, the other accepts. Until both have, no view is generated at all.
Pick the depth
Composite score only, skill components, or full breakdown with the underlying work. The stricter setting of the two applies.
Read the gaps, not the total
The total is almost useless. The useful output is the two skills where you are ahead and the two where you are behind.
End it whenever
One click, no notification, and the view is gone for both of you.
About comparison
Compare against
the job, not the person.
If peer comparison is not for you, the role benchmark answers the same question with nobody else in the frame.
We value your privacy
We use cookies to enhance your browsing experience, serve personalised content, and analyse our traffic. Non-essential cookies stay off until you choose. By clicking "Accept All", you consent to our use of cookies as described in our Cookie Policy.