Your sales emails, read and qualified for you

No, and that is the starting point. Your teams keep their addresses, their email client and their habits. Mail Box AI connects to your mailboxes — Microsoft 365, Google Workspace or IMAP — reads what comes in and produces a view alongside. A project that requires changing tools fails after three weeks; this one requires nothing.

Yes, and the setup allows for it. Every conclusion carries a confidence score. Above the threshold you set, the enquiry is classified automatically; below it, it goes to human review. A salesperson can correct a qualification at any time, and those corrections are used to adjust the rules. The observed rate is around 91%, which leaves roughly one enquiry in eleven to check.

By the actual subject matter, not just the message's subject line. A reply sent from another email client, an internal forward, a rewritten subject: these cases break the usual linking. The analysis draws on the people involved, the references quoted, the attachments and the content to recognise that it is the same case.

You set the scopes: a salesperson sees their own cases, a manager their team, management the whole. Personal mailboxes are not opened to everyone by virtue of the connection: only conversations attached to a sales case appear in the shared view.

No, it feeds it. A CRM waits to be filled in; that is precisely what nobody does. Mail Box AI produces the record, the need, the stage and the history from the emails, then pushes them into your CRM. If you don't have one, the built-in management view may be enough to begin with.

Across the deployments observed: about 70% less time spent handling emails, 50% more sales responsiveness, and a third more conversion on the opportunities detected. These figures depend on your volume and your current sales discipline — we recalculate them on your own data during the testing phase, before any commitment.

They remain yours. They are analysed to produce your qualifications, are not used to train a shared model, and are deleted at your request. Hosting and retention are set out in the contract; for sensitive activities, we look at dedicated hosting.

Mailboxes are connected within forty-eight hours. Importing the history, merging the threads and tuning the analysis to your products takes two to four weeks depending on volume. The pipeline becomes usable as soon as testing ends.

In Morocco, personal data processing falls under Law 09-08 and the CNDP: the emails analysed contain personal data — your prospects' as well as your employees' — and must appear in your declaration. Informing the data subjects, and informing your teams of the scope analysed, should be planned for. Processing is logged, and deletion requests are kept and exportable. We are not lawyers: have your setup and your CNDP declaration reviewed by your counsel.

The GDPR applies in full: legal basis, informing data subjects, retention period and the right to erasure. Two points deserve attention here — the analysis covers exchanges containing third-party data, and it also concerns your employees, which brings in staff information duties and sometimes consultation of employee representative bodies. The European AI Act applies on top. Processing is logged and exportable. We are not lawyers: have your setup reviewed by your counsel.

In the United States, state privacy laws — including California's CCPA — grant rights of access and deletion to the people whose data is processed, including in a business context. Since the analysis also covers your employees' messages, staff notification falls under each state's employment law. Processing is logged and exportable. We are not lawyers: have your setup reviewed by your counsel.

Yes. We connect one mailbox to a sample of your history and show you what the analysis draws from it: qualifications, merged threads, opportunities detected. You judge on your own emails rather than on a prepared demo.