Internal Comms & Meetings

How to write meeting notes as Small IT and ITOps Teams

Internal Comms & MeetingsFor Small IT and ITOps Teams3 apps10 steps~20 min to set up

You're a two-person IT team running standup calls, vendor syncs, and incident reviews every week — and nobody is writing anything down consistently. Your Jira tickets reference decisions that were made on a call three weeks ago, but the context lives in someone's head. When an L1 ticket escalates or a new hire joins, you're re-explaining the same reasoning you documented nowhere. You use Google Meet or Zoom for most calls, but notes either don't get written, land in a Slack message nobody pins, or go into a Notion runbook nobody reads. Action items get lost. Follow-ups fall through. You're too busy fixing the next fire to circle back on the last one.

Build this on Starch → See the apps used

Internal Comms & MeetingsFor Small IT and ITOps Teams3 apps10 steps~20 min to set up

Outcome

What you'll set up

A real-time transcription and summary layer for every IT standup, incident debrief, and vendor sync — with key decisions and action items automatically extracted after each call ends

A searchable meeting archive you can query by topic, date, or person — so when someone asks 'what did we decide about the Okta SSO rollout?' you find the exact moment in thirty seconds

Automated task capture that turns action items from meeting notes into tracked to-dos with owners and due dates, connected to your existing Jira or Notion workflow

The Starch recipe

Apps, data, and prompts

The combination of Starch apps, the data sources they pull from, and the prompts you use to drive them.

Apps used

Meeting Notes Task Manager Knowledge Management

Data sources & config

Meeting Notes connects to Google Calendar (Starch syncs your Google Calendar data on a schedule) to know which meetings to capture. Knowledge Management connects to Notion (connect Notion from Starch's integration catalog; the agent queries it live) to push runbook entries and summaries into your existing workspace. Task Manager works standalone — you capture tasks from chat or auto-extracted action items, and you can connect Jira or Asana from Starch's integration catalog to push tasks live when your workflow runs.

Prompts to copy

Transcribe this week's IT standup and extract all action items. Assign each one to the person who said they'd do it. Flag anything that was blocked or escalated.

Summarize this incident debrief call. What was the root cause we agreed on? What are the follow-up tasks and who owns them? What should go into the runbook?

Search my meeting history for every time we discussed the Salesforce license audit. Pull the decisions and any open action items we haven't closed.

Add a task: review Zoom license reclaim list before Thursday's vendor sync — P2 priority.

Create a runbook entry from last Tuesday's AWS cost anomaly debrief. Use the summary and decisions from that meeting as the source.

Run these in Starch → or paste them into your favorite agent

Walkthrough

Step-by-step

1 Start with the Meeting Notes app from the Starch App Store — it comes pre-built for real-time transcription, post-call summaries, and action-item extraction. No configuration required to get the first meeting captured.

2 Connect your Google Calendar: Starch syncs your Google Calendar data on a schedule so Meeting Notes knows your upcoming IT standups, incident reviews, and vendor syncs without manual entry.

3 Run your next standup or incident debrief as normal. After the call ends, Starch generates a structured summary — key decisions, open questions, and named action items with the person who owns each one.

4 Review the extracted action items. For anything that needs to be tracked formally, tell the Task Manager: 'Add a P1 task — confirm Okta SSO group mapping before Friday — owned by me.' It captures it with deadline and priority.

5 For tasks that belong in Jira, connect Jira from Starch's integration catalog; the agent queries it live and can push new tickets directly when your automation runs.

6 After an incident debrief, prompt the Knowledge Management app: 'Create a runbook entry from today's AWS cost anomaly call. Use the meeting summary as the source and tag it under Infrastructure.' This keeps your Notion runbook current without someone manually writing it up at 6pm.

7 Set up a weekly automation: 'Every Friday, pull all meeting summaries from this week, list every open action item that hasn't been marked done in Task Manager, and post a digest to the #it-team Slack channel.' Connect Slack from Starch's integration catalog for the delivery step.

8 When a new IT hire joins or an L1 ticket escalates, use the meeting history search: 'Find every meeting where we discussed the Jamf enrollment workflow and pull the decisions.' You get context in seconds instead of asking a teammate to reconstruct it.

9 For vendor syncs — Torii reviews, Zoom license audits, Salesforce seat reclaims — Meeting Notes captures the commitments the vendor made. You can search later: 'What did the Torii rep say about auto-reclaim thresholds in our February call?'

10 Over time, your Knowledge Management app builds a searchable institutional memory — meeting summaries, runbook entries, and decision logs all in one place — so the answer to most IT questions stops living in your head and starts living somewhere your teammates can find it.

See this running on Starch

Connect your tools, describe what you want, and the agent builds it. Closed beta is free.

Try it on Starch →

Worked example

March 2026 AWS Cost Anomaly Incident — Post-Incident Review

Sample numbers from a real run

Unexpected EC2 spend flagged by AWS Cost Checker	4,200
Action items extracted from 45-min debrief call	7
Tasks pushed to Jira from meeting summary	4
Runbook entry created in Notion from call transcript	1
Minutes spent writing the summary manually	0

On March 11, the AWS Cost Checker flagged a $4,200 spike in EC2 spend — a dev environment that wasn't torn down after a load test. You ran a 45-minute post-incident debrief on Google Meet with your IT lead and one engineer from the backend team. Starch transcribed the call in real time. After the call ended, Meeting Notes generated a structured summary: root cause (missing teardown automation), three immediate remediation steps, and seven action items — four owned by you, two by the backend engineer, one pending finance sign-off on the refund request. You told the Task Manager: 'Add P1 task — implement Lambda teardown trigger for dev environments — due March 18.' It added it with the right priority and deadline. You then told the Knowledge Management app: 'Create a runbook entry from today's AWS cost anomaly debrief — tag it under Cloud Cost Controls and link it to the March 11 incident.' The runbook entry was in Notion within two minutes, written from the actual call, not from memory. The following Friday, the weekly digest surfaced two open action items that still hadn't been marked done — including the finance refund request — so nothing quietly disappeared.

Measurement

How you'll know it's working

Action items closed per week vs. action items opened in meetings (closure rate)

Mean time from incident debrief to runbook entry published

Number of IT decisions documented and searchable vs. decisions that exist only in someone's head

Time spent on meeting admin per week (writing summaries, chasing follow-ups) before and after

Repeat tickets caused by undocumented decisions or missed follow-ups

Comparison

What this replaces

The other ways teams handle this today, and how the Starch version compares.

Otter.ai or Fireflies

Good transcription, but action items don't connect to your Jira, Notion, or Slack — you still manually copy things over after every call.

Notion meeting notes template

Works if someone writes it up, but in a two-person IT team under incident load, nobody does — it stays blank or gets one sentence.

Google Docs + manual Jira tickets

Free and already in your stack, but you're the one writing the doc, extracting the action items, and creating the tickets — that's 20 minutes per meeting you don't have.

Loom async updates

Useful for async communication, but Loom recordings aren't searchable by decision or action item — you can't query 'what did we decide about the Okta rollout' across six months of videos.

On Starch RECOMMENDED

One platform — meeting notes, task manager, knowledge management all running on connected data. Setup in plain English; numbers stay current via scheduled syncs and live agent queries.

Try it on Starch →

FAQ

Frequently asked questions

Does Starch actually join and record Google Meet or Zoom calls?

Meeting Notes works with your call recordings and transcripts — it doesn't need to drop a bot into the call itself. If your Google Meet is connected via Google Calendar (Starch syncs your calendar data on a schedule), it knows which meetings happened. You can also feed it a transcript directly. Check the Meeting Notes app for the current capture methods.

Can it push action items directly into Jira tickets, or do I still have to do that manually?

You can connect Jira from Starch's integration catalog; the agent queries it live when an automation runs. Tell Starch: 'After each IT standup, create a Jira ticket for any action item tagged as P1 or P2.' It handles the ticket creation. Lower-priority tasks can stay in Task Manager if you prefer.

We use Notion for our runbook. Will Starch write directly into it?

Yes. Connect Notion from Starch's integration catalog; the agent queries it live and can write pages when your automation runs. You describe the structure you want — 'create a page under IT Runbooks, tag it by incident type, and populate it from the meeting summary' — and Starch builds that automation.

What if the call has sensitive vendor contract discussions or personnel topics? Where does the transcript go?

Starch is not SOC 2 Type II certified today — that's worth knowing if your org has strict data residency requirements. Transcript data lives in Starch's database. If your IT policy restricts where call recordings can be stored, check with your security lead before connecting sensitive call data.

We don't have a consistent meeting schedule — a lot of IT work is reactive. Does this still help?

Yes. Meeting Notes doesn't require a recurring calendar invite. You can trigger it manually for any call — an unplanned incident bridge, a vendor escalation, a quick architecture sync. The meeting archive builds up over time regardless of whether the meetings were scheduled or ad hoc.

How is this different from just making everyone take turns writing the Confluence or Notion doc?

The person whose turn it is to take notes is also the person trying to follow the conversation and make decisions. Something always gets missed, and the doc stops getting written after three weeks when the rotation breaks down. Starch does it automatically after every call — the output quality doesn't depend on whose week it is or whether anyone remembered.