Binee can remember lasting preferences for you, like how you want summaries formatted or which client matters most. Memory is personal: your memories belong to you in this workspace, and no teammate can see them.

The Binee Assistant button at the top of the app opens the same Binee chat beside the Memory page too, so you can save a preference conversationally right there, for example "remember that I want summaries as bullet points", with the same confirmation card as in chat.

While your list is empty, the Memory page shows a clearly labeled example memory so you can see the shape of a good one: a lasting preference, not a one-off request. The example is just an illustration; it is never stored.

Binee also learns light context from conversations (like facts about your business) to answer better; those learned notes appear in the same list, labeled by where they came from: Added by you, Saved from chat, or Learned from conversations.

Delete any memory from the Memory page in the sidebar, or ask Binee in chat to forget it (deletion also shows a confirmation card). Forgetting is a hard delete: the memory is gone, not archived.

Saved preferences guide Binee's default behavior, but an explicit instruction in your current message always wins over a saved preference.

Saved preferences also shape your insights. When one of your automations fires, Binee writes the finding with your preferences as reader priorities: they decide what the finding leads with and emphasizes, never whether something matches an automation. Example: after you save "negative client intent matters most to me", an automation that recaps your sales meetings leads every recap with how the client sounded, while a teammate with no such preference gets a plain recap of the same kind of meeting. This is personal by design: your preferences shape only your own insights.