The five questions we get before every signing.

No. Quillhall runs on isolated per-tenant infrastructure and uses your messages only to draft replies in the same session — never as training data. Drafts stay inside your tenant, encrypted at rest and in transit, and only the operatives assigned to your account can read them. You can request export or deletion at any time, and we publish the same data-handling policy we sign with enterprise customers.

Your voice profile is built from your own prior correspondence and lives alongside the rest of your tenant data — it is never pooled across customers. Each draft is grounded in the thread it replies to: the model pulls cited context, writes in the language the thread started in, and stops short of inventing facts. Nothing ships without your approval, and you can flag any draft for a regeneration or a manual pass.

Quillhall is priced on outcomes, not seats. A single principal on Pro covers one mailbox, one calendar, and one Slack workspace at a flat monthly rate — adding a second channel or a co-principal steps you up a tier rather than adding line items. The pricing page lists the breakpoints by outcome (drafts shipped, briefings delivered, channels kept).

Google Workspace and Microsoft 365 are first-class, both for mailbox and calendar, including shared mailboxes and delegated calendars. Slack covers triage notifications and channel replies. Investor portals and CRMs connect via read-only OAuth bridges that pull the thread context the agent needs without exposing writes back upstream. New integrations land with the same SLA as the core surfaces.

Always-on means a live triage queue staffed in three geographic shifts, with measured monthly uptime on record. Overnight and weekend coverage is the same queue, not an after-hours voicemail — urgent threads are triaged within minutes, drafts are reviewed on the next business morning, and the morning brief lands before the inbox floods. The SLA spells out response targets by tier and credits you when we miss them.