We had a client who held eleven Microsoft Teams calls a week. And at the end of every single one, someone said “right, I’ll send a follow-up with the action points.”
That email arrived about 40% of the time. When it did, it was vague. When it was vague, nobody knew who owned what. And when nobody knew who owned what, the same conversation happened again the following Tuesday.
Sound familiar? Good. Because this is the part where it stops.
The real problem isn’t the meetings
The note-taking problem is not that people are lazy. It’s that we’re asking one person to do two cognitively opposite things at the same time.
You cannot genuinely listen to someone and simultaneously write a structured summary of what they’re saying. So what you get is fragments. Half-sentences. “Discuss risk thing with Sarah?” written at 10:43am and meaningless by 4pm.
The gap between “we discussed it” and “it got done” is a process gap. And process gaps are exactly what AI workflow automation is built for.
What the automated version looks like
When a call ends, Microsoft Teams passes the meeting transcript into a Make.com automation workflow. That workflow sends the transcript to an AI model, which identifies action items, assigns them to whoever was mentioned, attaches due dates based on context, and pushes the whole lot straight into your task management tool.
Every action item lands as a task. Assigned. Due-dated. Categorised. Before anyone has left the call.
No copying. No pasting. No “I’ll send the follow-up.” Done.
One of our SME clients was spending three hours a week across their team chasing actions from meetings. That’s 150 hours a year of people doing admin that a workflow handles in seconds.
You don’t need a developer for this
Three things required:
Microsoft Teams with transcription enabled. It’s a setting in your admin panel. If you’re on a Business or Enterprise licence, it’s almost certainly already there.
A Make.com account. This is the workflow automation platform we use for most of our UK SME clients. It connects Teams to your task manager — Asana, Trello, Monday.com, or a spreadsheet — using a visual builder. No code.
A prompt that actually works. This is where most DIY attempts fall down. A vague prompt produces vague tasks. A prompt built around your team’s language and how your meetings actually run produces something you’d trust.
The bit people don’t expect
The biggest benefit isn’t the time saved. It’s the accountability shift.
When action items are written manually, there’s always ambiguity. “Was that definitely assigned to me?” With an automated system, there’s a record. Timestamped. Attributed. In the task manager before anyone’s closed their laptop.
Nobody can say “I didn’t realise that was mine.” It’s on the board. It has their name on it.
That shift in accountability, across a small business team, over twelve months, is worth considerably more than three hours a week.
If you hold more than a few Teams calls a week and your follow-up process is “someone tries to remember,” this is worth a conversation.
We offer a free 45-minute AI Readiness Audit. No pitch. No obligation. A PDF report lands in your inbox afterwards.
Book a free AI Readiness Audit → tier3-solutions.com/contact
