In short

Most broken promises to clients are not deliberate. Someone said they would send something, nobody wrote it down, and a week passed. The improved process reads meeting notes and client email for promises and follow-ups, turns each one into a task with an owner and a date, and drafts the reply or the cover note where one is due.

The system finds and prepares. Every message to a client is approved by a person before it is sent.

  • Sector Professional services firms
  • Role Account manager or team lead
  • Job Following up what was promised

The job today

At the end of a meeting, people agree to send a proposal, a document or an answer. Notes are taken, but actions are buried in them. Follow-ups depend on each person's memory and inbox. Clients chase, which is the first time the team notices something was missed.

Where it breaks

  • Actions are buried in meeting notes and long email threads.
  • It is not clear who owns each follow-up.
  • Clients have to chase, which damages the relationship.
  • Managers have no view of what has been promised and is still open.

The improved process, step by step

  1. Read notes and emailAfter each client meeting, and as client email arrives, the system reads the notes and threads chosen for this.
  2. Find the promises AI preparesIt picks out commitments such as "we'll send", "I'll check" or "by Friday", with the exact words and where they came from.
  3. Create the taskEach promise becomes a task with a suggested owner and date, for the team lead to confirm.
  4. Draft the follow-up AI preparesWhere a reply or cover note is due, an assistant drafts it in your tone, ready to attach what was promised.
  5. A person approves each send A person approvesThe owner checks the draft, attaches anything needed and approves it. Nothing goes to a client without that approval.
  6. Show what is still openA short list each morning shows promises due today and anything overdue.
Catching promises and follow-ups after client meetings: process diagramSteps: 1. Read notes and email; 2. Find the promises (AI prepares); 3. Create the task; 4. Draft the follow-up (AI prepares); 5. A person approves each send (a person approves); 6. Show what is still open 01 Read notes and email 02 AI PREPARES Find the promises 03 RULES CHECK Create the task 04 AI PREPARES Draft the follow-up 05 A PERSON APPROVES A person approves each send 06 Show what is still open
The process at a glance. Teal steps are prepared with AI help; the amber step is where a person approves.

Where a person must approve

A person approves

Every message to a client is approved by a person before it is sent. Anything that sends, changes or cannot be undone waits for that approval.

The system never makes new promises or agrees dates on the team's behalf.

The data boundary

  • The system reads meeting notes and client email chosen for this purpose. It does not read personal or unrelated mailboxes.
  • It can create draft tasks and replies. It cannot send email or change client records on its own.
  • Each task links to the words it came from, so it can be checked.
  • Every task, draft and send is logged.

Checklist before you start

  • Meeting notes are saved somewhere the system can read.
  • We know which mailboxes hold client email.
  • We have one place where tasks should live.
  • Each client has a named owner.
  • We have agreed how quickly follow-ups should go out.
  • We know who approves messages for each client.

This is an illustrative walkthrough of a common pattern. It is not a client case study, and the details will differ in your organisation.