Every draft your team writes, checked in Outlook and Gmail against the rules that govern lettings and sales. It names the rule, explains the exposure and says what to weigh up — and never blocks the send.
The Property Ombudsman, 2024–2025 annual reviews
SafeReplies has flagged emails from our team that could easily have turned into deposit disputes or regulatory fines. We review them straight from the notification and rectify them quickly — none of which we'd have spotted until it was too late.
Aaron ShortSafeReplies is a preventative tool that can flag risks before you send.
Responsibility for what leaves your outbox stays with the sender and their organisation.
A promise made three messages up is caught too.
Senders learn as they go.
Nothing is blocked. Nothing slips through quietly.
Who sent it and how serious — by email, within minutes. The sentence and the rule are one click away.
Put it right before they reply. Record what you did in one click.
Repeat flags show as a pattern, so coaching goes where it's needed.
Spots frustration, escalation and formal complaints in what the other person has written.
Watch, Escalating or Formal, with the line that triggered it.
Mark it handled, dismiss it, or tell us the flag was wrong.
Admins see everything — every branch and team org-wide, all flags, plus members, branches, policies and billing controls.
Illustrative dataEvery team's compliance, before they hit send.
No. SafeReplies never stops an email leaving. If your draft looks fine you'll see nothing. If we spot a potential issue we flag it, explain why it's a risk, and set out what to consider before you send — then it's your call. You can edit the wording or send it as it stands. If a draft is high risk, SafeReplies asks you to confirm when you press Send — one click and it goes. We never send, alter or approve email on your behalf.
Across more than 1,000 test emails per industry pack, SafeReplies caught around 9 in 10 high-risk issues. It will not catch everything — no pre-send check can, and wording risk depends on context we may not see. Treat it as a preventative second pair of eyes that removes most of the obvious exposure, not a guarantee that an email is safe or compliant.
We don't archive your inbox or use your emails to train models. Your data is stored in the UK. The check itself is run by AI providers (OpenAI and Anthropic), which are contractually barred from retaining or training on your content. Checked drafts are kept for 14 days with contact details removed, so we can check the scanner's accuracy, and flagged emails are kept encrypted for 14 days so a manager can review them. Both are then deleted. After that we keep only the flag record — the type of risk, the short phrase that triggered it and which user — to power your reports.
Outlook on the web and the new Outlook app, and Gmail in Chrome. Our Chrome extension also covers Outlook on the web. It isn't available yet on mobile apps, the classic Outlook desktop app or browsers other than Chrome.
Yes. Start with a 7-day free trial — we take your card up front but won't charge you until the trial ends. Cancel any time before then and you won't be charged.
7 days free for your whole team. Cancel anytime.