Two distinct fears, answered separately: your members' fear of exposure, and your own.
The owner sees the aggregate — never the person. This isn't a policy promise; it's how the system is built.
The intelligence is produced by a walled-off role with RPC-only access — it can call aggregate functions, and it can never read an individual's row.
Any figure drawn from too few people is automatically suppressed (k-anonymity) — so nothing can be traced back to one person.
You see the shape of the community — the recurring needs, the clusters of common ground — and nothing that names anyone.
So the start is built to be low-risk: founder-led close support, a low-commitment start, a free trial month, and a boutique model — a few networks, supported well. You're not making a big-ticket procurement bet; you're starting a conversation and a trial, with your network and your data yours from day one.
Your network is always yours.
The two aren't in tension — they reinforce. Members are candid because they're protected; and that candour is exactly what makes the community intelligence trustworthy. Aggregate-and-auditable for you; private-and-unexposed for them. Each is the reason the other works.
Cross-check the privacy claims against our verification documents — ask in the conversation.
Start with a conversation →