User behaviour
Intent, before the bid
Watchlist adds, lot follows, push opens, email clicks, login cadence — the engagement signals that explain why a bidder is leaning in, or quietly stepping back.
Data domains
The engagement layer
User Behaviour sits between Traffic (anonymous attention) and Bids (committed action). It records the considered engagements your registered bidders make: adding a lot to a watchlist, following a category, opening a push notification, clicking a link in an emailed catalogue, logging in to check on a watched lot.
Because these signals are emitted by identified users, they're the strongest leading indicator the platform has for both intent ('this bidder is leaning into our sale next week') and disengagement ('this VIP hasn't opened the last four emails — flag for relationship'). The domain rolls them up per customer, per sale, and per category.
Ask your data
From a question to a re-engagement shortlist
Ask which VIPs are quietly disengaging. The MCP server computes the engagement score, contrasts it with the prior period, and returns a re-engagement list.
Access & governance
Behaviour is the most actionable PII
Behavioural data is sensitive because it directly enables targeted outreach. Default analyst access is aggregated; per-customer scores are gated to engagement and account-management roles.
Aggregate distributions and cohort-level engagement metrics are open to all analyst roles.
Per-customer engagement score and channel-level activity require the `customers.engage` capability.
Disengagement-risk lists are surfaced only to account-management roles, with a 14-day re-export cooldown to avoid over-contact.
Related domains
Behaviour closes the loop with engagement
Let's talk
Ready to modernize your auction house?
Book a personalised demo and see Auction Rabbit tailored to your sale calendar