Fireflies + Slack: coaching notes after every call
Your reps finish a call, and the next 20 minutes disappear into replays, half-written notes, and “what did the prospect actually push back on?” By the time you’ve found the objection, the coaching moment is already cold.
Sales managers feel this weekly. Enablement leads end up chasing consistency across reps. And if you run an agency or a small sales team, you’re the one doing the coaching too. This Fireflies Slack coaching automation turns every finished call into structured coaching notes and objection handlers, delivered where you already work.
You’ll see how the workflow pulls the transcript, gets an AI evaluation, posts a clean Slack summary, and saves a Google Drive report you can reuse in training.
How This Automation Works
Here’s the complete workflow you’ll be setting up:
n8n Workflow Template: Fireflies + Slack: coaching notes after every call
flowchart LR
subgraph sg0["When clicking ‘Execute workflow’ Flow"]
direction LR
n0["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/webhook.dark.svg' width='40' height='40' /></div><br/>Fireflies Webhook Trigger"]
n1["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/httprequest.dark.svg' width='40' height='40' /></div><br/>Get AI App ID"]
n2["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/code.svg' width='40' height='40' /></div><br/>Gets Latest Meeting"]
n3@{ icon: "mdi:cog", form: "rounded", label: "Get outputs of an AI app 1", pos: "b", h: 48 }
n4@{ icon: "mdi:cog", form: "rounded", label: "Get outputs of an AI app 2", pos: "b", h: 48 }
n5@{ icon: "mdi:play-circle", form: "rounded", label: "When clicking ‘Execute workf..", pos: "b", h: 48 }
n6@{ icon: "mdi:brain", form: "rounded", label: "OpenAI Model", pos: "b", h: 48 }
n7@{ icon: "mdi:robot", form: "rounded", label: "Call Analysis Output Structure", pos: "b", h: 48 }
n8@{ icon: "mdi:cog", form: "rounded", label: "Create file from text", pos: "b", h: 48 }
n9@{ icon: "mdi:cog", form: "rounded", label: "Wait", pos: "b", h: 48 }
n10@{ icon: "mdi:robot", form: "rounded", label: "AI Call Analyzer", pos: "b", h: 48 }
n11@{ icon: "mdi:cog", form: "rounded", label: "Get a list of transcripts", pos: "b", h: 48 }
n12@{ icon: "mdi:cog", form: "rounded", label: "Get a summary of a transcript", pos: "b", h: 48 }
n13@{ icon: "mdi:swap-vertical", form: "rounded", label: "Prepare Call Data", pos: "b", h: 48 }
n14@{ icon: "mdi:swap-vertical", form: "rounded", label: "Format Analysis Results", pos: "b", h: 48 }
n15["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/slack.svg' width='40' height='40' /></div><br/>Send Rep Feedback to Slack"]
n9 --> n11
n6 -.-> n10
n1 --> n12
n10 --> n14
n13 --> n10
n2 --> n1
n14 --> n15
n0 --> n9
n11 --> n2
n3 --> n4
n4 --> n13
n15 --> n8
n12 --> n3
n7 -.-> n10
n5 --> n11
end
%% Styling
classDef trigger fill:#e8f5e9,stroke:#388e3c,stroke-width:2px
classDef ai fill:#e3f2fd,stroke:#1976d2,stroke-width:2px
classDef aiModel fill:#e8eaf6,stroke:#3f51b5,stroke-width:2px
classDef decision fill:#fff8e1,stroke:#f9a825,stroke-width:2px
classDef database fill:#fce4ec,stroke:#c2185b,stroke-width:2px
classDef api fill:#fff3e0,stroke:#e65100,stroke-width:2px
classDef code fill:#f3e5f5,stroke:#7b1fa2,stroke-width:2px
classDef disabled stroke-dasharray: 5 5,opacity: 0.5
class n5 trigger
class n7,n10 ai
class n6 aiModel
class n0,n1 api
class n2 code
classDef customIcon fill:none,stroke:none
class n0,n1,n2,n15 customIcon
Why This Matters: Coaching Dies in the Gap After Calls
Call coaching usually fails for a boring reason: the admin work is heavier than the insight. You have recordings in one place, transcripts in another, feedback in a random Slack thread, and “best practice” living in someone’s head. Meanwhile, objections repeat. Pricing pushback. Competitor comparisons. “Send me something.” If you’re not capturing those patterns consistently, you’re coaching based on memory, not evidence. And honestly, memory is biased toward the loudest moment, not the real turning point.
The friction compounds. Here’s where it breaks down.
- Managers spend about an hour per rep each week hunting for moments worth coaching.
- Reps get vague feedback because nobody wants to rewatch a 40-minute call to find one missed question.
- Objections are logged inconsistently, so you can’t tell what’s trending across the team.
- Good coaching notes disappear into DMs and never make it into onboarding or playbooks.
What You’ll Build: Fireflies Transcript Analysis Posted to Slack + Saved to Drive
This workflow starts when Fireflies notifies n8n that a meeting has ended. After a short delay (so Fireflies has time to finish syncing), n8n pulls the most recent transcript and summary, then collects the AI app outputs Fireflies provides. Those pieces are assembled into a single “call payload,” which is sent through an AI Agent powered by an OpenAI Chat Model. The agent evaluates objection handling, call effectiveness, and extracts the key objections that came up. Then it generates practical objection handlers your reps can reuse on the next call. Finally, n8n formats everything into a structured report, posts it to Slack for fast coaching, and saves the same report as a Google Drive document for your records and training library.
The workflow begins at a Fireflies webhook trigger. From there it retrieves the right transcript, runs a structured AI analysis, and outputs the results in two places: Slack for visibility and Google Drive for long-term reuse.
What You’re Building
| What Gets Automated | What You’ll Achieve |
|---|---|
|
|
Expected Results
Say you review 10 calls per week across the team. Manually, even a “quick” review is about 20 minutes to find the right transcript, skim it, and write coherent coaching notes, which is roughly 3 hours of admin. With this workflow, the trigger is automatic and the AI report shows up in Slack a few minutes after each call, plus a Google Drive doc is saved for later. You still do the coaching, but the scavenger hunt is gone. Most teams get a couple hours back every week right away.
Before You Start
- n8n instance (try n8n Cloud free)
- Self-hosting option if you prefer (Hostinger works well)
- Fireflies for meeting transcripts and summaries.
- Slack to deliver coaching notes instantly.
- OpenAI API key (get it from your OpenAI API settings page)
- Google Drive to store the formatted report documents.
Skill level: Intermediate. You’ll connect accounts, paste an API key, and adjust a prompt, but you won’t be writing an app.
Want someone to build this for you? Talk to an automation expert (free 15-minute consultation).
Step by Step
A Fireflies “meeting ended” event kicks it off. Fireflies calls your n8n webhook when the meeting is complete. The workflow waits briefly so the transcript and AI outputs are fully available.
The workflow pulls the right transcript and supporting data. n8n retrieves the transcript list, selects the most recent meeting, then fetches the transcript summary and the AI app outputs associated with that call.
AI turns raw conversation into coaching assets. A LangChain-style AI Agent using an OpenAI Chat Model evaluates objection handling and call effectiveness, then produces structured fields (so you don’t get a messy wall of text). This is also where your sales methodology can be baked in via the prompt notes.
Results land in Slack and Google Drive. The formatted insight report is posted to your chosen Slack channel for quick review. The same content is turned into a Google Drive document so it’s easy to search, share, and reuse later.
You can easily modify the objection categories to match MEDDICC, SPICED, Challenger, or your internal playbook. See the full implementation guide below for customization options.
Step-by-Step Implementation Guide
Step 1: Configure the Webhook Trigger
Set up the incoming webhook that starts the workflow when Fireflies completes a call, and keep the manual trigger for testing.
- Open Fireflies Incoming Hook and set HTTP Method to
POST. - Set Path to
fireflies-call-completed. - Keep Manual Start Trigger connected to Retrieve Transcript List for manual test runs.
Step 2: Connect Fireflies and Pull the Latest Transcript
Retrieve the newest transcript after a short delay to ensure Fireflies has finished processing.
- In Delay for Sync, set Unit to
minutesand Amount to8. - Open Retrieve Transcript List and set Operation to
getTranscriptsList. - Credential Required: Connect your firefliesApi credentials in Retrieve Transcript List.
- In Select Recent Meeting, keep the JavaScript code as-is to sort by
dateStringand output the latest transcript ID.
Step 3: Fetch Fireflies AI App Outputs and Build the Call Payload
Use the transcript ID to pull AI app outputs and assemble the data needed for analysis.
- In Fetch AI App Identifier, set URL to
https://api.fireflies.ai/graphqland Method toPOST. - Set JSON Body to
={ "query": "query GetAIAppsOutputs($transcriptId: String) { apps(transcript_id: $transcriptId) { outputs { transcript_id user_id app_id created_at title prompt response } } }", "variables": { "transcriptId": "{{ $json.transcript_id }}" } }. - In Fetch AI App Identifier, update the Authorization header value from
Bearer [CONFIGURE_YOUR_TOKEN]to your Fireflies API token. - Open Fetch Transcript Summary and set Transcript ID to
={{ $('Select Recent Meeting').item.json.transcript_id }}. - Credential Required: Connect your firefliesApi credentials in Fetch Transcript Summary, Retrieve AI App Output A, and Retrieve AI App Output B.
- In Retrieve AI App Output A, set App ID to
={{ $('Fetch AI App Identifier').item.json.data.apps.outputs[0].app_id }}and Transcript ID to={{ $('Fetch AI App Identifier').item.json.data.apps.outputs[0].transcript_id }}. - In Retrieve AI App Output B, set App ID to
={{ $('Fetch AI App Identifier').item.json.data.apps.outputs[1].app_id }}and Transcript ID to={{ $json.data.transcript_id }}. - In Assemble Call Payload, map fields like transcript to
={{ $('Fetch Transcript Summary').item.json.data.summary.short_summary }}, keyObjections to={{ $('Retrieve AI App Output A').item.json.data.response }}, and objectionHandler to={{ $('Retrieve AI App Output B').item.json.data.response }}.
Step 4: Set Up the AI Insight Generation
Configure the AI agent with the OpenAI model and structured output parser to produce clean JSON and a Slack summary.
- Open Sales Call Insight Agent and keep the Prompt text as defined for objection analysis and Slack summary rules.
- Ensure Sales Call Insight Agent uses input variables from Assemble Call Payload, such as
{{ $('Assemble Call Payload').item.json.repName }}and{{ $('Assemble Call Payload').item.json.transcript }}. - Confirm OpenAI Chat Model is connected as the language model for Sales Call Insight Agent.
- Credential Required: Connect your openAiApi credentials in OpenAI Chat Model.
- Verify Analysis Output Schema is connected to Sales Call Insight Agent as the output parser with the provided JSON schema example.
Step 5: Configure Output and Formatting
Format the AI output, send a Slack message, and generate a Google Doc summary.
- In Format Insight Results, map fields like repName to
={{ $('Assemble Call Payload').first().json.repName }}and callDate to={{ $json.output.metadata.call_date }}. - In Post Slack Feedback, set Text to the provided template with expressions such as
{{ $('Sales Call Insight Agent').item.json.output.slack_summary }}. - Credential Required: Connect your slackOAuth2Api credentials in Post Slack Feedback.
- In Generate Doc from Text, set Operation to
createFromText, and keep the Name and Content expressions that reference Format Insight Results. - Credential Required: Connect your googleDriveOAuth2Api credentials in Generate Doc from Text.
Step 6: Test and Activate Your Workflow
Run a manual test to confirm data flow, then enable the webhook for production use.
- Click Execute Workflow using Manual Start Trigger to run a test with the latest Fireflies transcript.
- Verify that Post Slack Feedback posts a message containing the Slack summary and call details.
- Confirm Generate Doc from Text creates a document in the selected Google Drive folder.
- Activate the workflow and point Fireflies to the Fireflies Incoming Hook URL for live processing.
Troubleshooting Tips
- Fireflies credentials can expire or need specific permissions. If things break, check your Fireflies API access and connected app permissions first.
- If you’re using Wait nodes or relying on Fireflies to finish processing, timing varies. Bump up the delay if the transcript fetch sometimes returns empty.
- OpenAI prompts that are too generic create generic coaching. Add your objection taxonomy and “what good looks like” early, or you will be editing outputs forever.
Quick Answers
About 30 minutes if Fireflies, Slack, and Drive are already connected.
No. You’ll connect accounts, paste your OpenAI key, and tweak the prompts.
Yes. n8n has a free self-hosted option and a free trial on n8n Cloud. Cloud plans start at $20/month for higher volume. You’ll also need to factor in OpenAI API costs, which are usually a few cents per call depending on transcript length.
Two options: n8n Cloud (managed, easiest setup) or self-hosting on a VPS. For self-hosting, Hostinger VPS is affordable and handles n8n well. Self-hosting gives you unlimited executions but requires basic server management.
Yes, and you should. The easiest win is changing the prompt inside the Sales Call Insight Agent to match your sales methodology and objection categories. You can also swap the Slack destination (post to a rep-specific channel or a manager channel), and change the Google Drive folder used by the “Generate Doc from Text” step so each team gets its own library.
Usually it’s API access or expired credentials on the Fireflies side. Reconnect Fireflies in n8n, then confirm the webhook is still active and pointing to the correct n8n URL. If failures happen only sometimes, it’s often timing: the transcript isn’t ready yet, so increasing the “Delay for Sync” wait fixes it.
A normal small-team setup can handle dozens of calls per day without drama, as long as your OpenAI and Fireflies limits aren’t being hit. On n8n Cloud, volume depends on your plan’s monthly executions. If you self-host, there’s no execution cap, but your server resources and transcript length will decide how many run smoothly at once.
Often, yes, because this workflow benefits from multi-step logic (waiting for sync, selecting the latest meeting, enforcing structured AI output, formatting, and then sending to two destinations). Zapier and Make can do pieces of it, but you may end up with multiple zaps/scenarios and weaker control over the “shape” of the AI output. n8n also gives you the self-hosting option, which matters when call volume grows. If you want simple “transcript to Slack” only, those tools can be fine. If you want consistent coaching notes with objection handlers and an archive, n8n is the more practical build. Talk to an automation expert if you want help choosing.
Once this is running, coaching notes show up while the call is still fresh. You keep the human judgment, and the workflow handles the busywork.
Need Help Setting This Up?
Our automation experts can build and customize this workflow for your specific needs. Free 15-minute consultation—no commitment required.
