AI email drafting

AI email drafting where your team reads, edits and sends

Much of a small team’s email is the same request written again, then chased when nobody replies. We set up workflows that write those emails as drafts from your own records, and your people read, edit and send every one.

A woman working on a laptop at a wooden desk with a notebook and a mug, in a bright room with plants

Where this time goes

  1. The same request, again

    Availability requests to partners get written one at a time.

  2. Chasing the quiet ones

    Follow-ups to partners who have gone quiet depend on someone remembering to send them.

  3. Gaps filled with a guess

    A quick draft can cover a missing detail with a plausible one.

  4. Your voice, kept consistent

    Tone and format drift when different people write the same kind of email.

How the workflow runs

  1. Write down your voice and sources. Your formats, tone and trusted records go into your workspace, so drafts start in your house style.
  2. Read your records. The workflow reads from your own records, such as email, shared folders and spreadsheets.
  3. Create the draft. The email is created as a draft. Where it lacks a detail, it is set up to ask your team rather than invent one.
  4. Your team reads and edits. A person checks the facts and changes the wording as needed.
  5. Your team sends. A person sends the email. Follow-ups for quiet partners are drafted the same way.

Example

Client example · Travel agency

Journey workflows, with the team sending everything

A travel agency that plans multi-stop journeys for its guests wanted help with the work between enquiry and departure. Within the workspace we set up around its journey process, the workflows draft partner availability requests and follow-ups, and prepare proposals and pre-departure messages. Emails are created as drafts. The team reads, edits and sends them. Some steps in the journey are still done by hand.

This describes what was built, not measured results.

A woman working on a laptop at a wooden desk with a notebook and a mug, in a bright room with plants

What stays with your team

  • Reading and editing

    A person reads every draft and edits it before it goes anywhere.

  • Sending

    The workflows create drafts. Your people send them.

  • The facts

    Your team checks rates, details and names before anything reaches a guest, partner or client.

  • Who approves what

    We agree this with you before the build.

  • Professional sign-off

    Work requiring professional sign-off remains with the authorised person.

Questions

Will the workflow email people by itself?
No. In our travel agency example the workflows create email drafts, and a person reviews and sends them.
Can AI get details wrong?
Yes. AI can make errors or miss context. The workflows read from your own records and are set up to ask your team for details they do not have rather than invent them. Your team checks the result before it reaches anyone.
Does it work with our email, and how is the content handled?
That is checked in discovery. The travel agency example ran on email, shared folders and spreadsheets. Data handling and platform suitability must be checked for the specific use. These controls do not guarantee compliance or replace professional judgement.

Start with this task.

Build with AI sets up AI workflows, with practical lessons for the team, for owner-led businesses in Cape Town and across South Africa.

See the three setup options