KI-Anrufzusammenfassung + HubSpot-Follow-up-Aufgabe
Fortgeschritten
Dies ist ein CRM, AI Summarization-Bereich Automatisierungsworkflow mit 12 Nodes. Hauptsächlich werden Hubspot, FormTrigger, HubspotTool, Agent, LmChatOpenAi und andere Nodes verwendet. KI-Anrufzusammenfassung + HubSpot-Follow-up-Aufgabe
Voraussetzungen
- •HubSpot API Key
- •OpenAI API Key
Verwendete Nodes (12)
Workflow-Vorschau
Visualisierung der Node-Verbindungen, mit Zoom und Pan
Workflow exportieren
Kopieren Sie die folgende JSON-Konfiguration und importieren Sie sie in n8n
{
"meta": {
"instanceId": "09423a3357ff64bdcc082268b9d577001317edbe377a3ccfb0b131ffb9554b30"
},
"nodes": [
{
"id": "62aa0111-3390-474f-9779-1d389398d3a0",
"name": "Kontakt per E-Mail finden",
"type": "n8n-nodes-base.hubspot",
"position": [
528,
224
],
"parameters": {
"operation": "search",
"authentication": "oAuth2",
"filterGroupsUi": {
"filterGroupsValues": [
{
"filtersUi": {
"filterValues": [
{
"value": "={{ $json['Contact email'] }}",
"propertyName": "email|string"
}
]
}
}
]
},
"additionalFields": {
"properties": "={{ [\"email\",\"firstname\",\"lastname\",\"jobtitle\",\"company\",\"country\",\"state\",\"city\",\"hs_language\",\"phone\",\"mobilephone\",\"lifecyclestage\",\"hs_lead_status\",\"hubspot_owner_id\",\"hs_email_last_open_date\",\"hs_email_last_reply_date\",\"hs_latest_meeting_activity\",\"hs_sequences_is_enrolled\",\"hs_sequences_enrolled_count\",\"createdate\",\"hs_lastmodifieddate\",\"hs_timezone\",\"notes_last_contacted\",\"hs_object_id\"] }}"
}
},
"typeVersion": 2.1
},
{
"id": "bc13ee93-f748-4058-9429-a25883e9c25e",
"name": "OpenAI-Chat-Modell",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"position": [
848,
448
],
"parameters": {
"model": {
"__rl": true,
"mode": "list",
"value": "gpt-4.1-mini"
},
"options": {}
},
"typeVersion": 1.2
},
{
"id": "1dbedf6b-1427-4127-a0bf-ac745c386e6a",
"name": "Formular: Transkript erfassen",
"type": "n8n-nodes-base.formTrigger",
"position": [
304,
224
],
"webhookId": "4f87bdf1-6594-4675-a20e-037a7f287da5",
"parameters": {
"options": {},
"formTitle": "Get transcript",
"formFields": {
"values": [
{
"fieldType": "email",
"fieldLabel": "Contact email",
"requiredField": true
},
{
"fieldType": "textarea",
"fieldLabel": "Paste transcript here",
"requiredField": true
}
]
}
},
"typeVersion": 2.3
},
{
"id": "e4ce0bc0-eaef-4d12-80af-8be92a32c059",
"name": "KI: Anruf zusammenfassen & Aufgabe entwerfen",
"type": "@n8n/n8n-nodes-langchain.agent",
"position": [
848,
224
],
"parameters": {
"text": "=### Inputs\n\n* **Contact info:** `{{ JSON.stringify($json.properties) }}`\n* **Call transcript (plain text):** `{{ $('Form: Capture Transcript').item.json['Paste transcript here'] }}`\n\n### Your goals\n\n1. Read the transcript and capture: participants, company, role, problem/opportunity, key requirements, objections/blockers, timeline, and any metrics (team size, volume, \\$\\$, dates).\n2. Write a **120–160 word** plain-English summary a teammate can skim.\n3. Create **one actionable follow-up task** that clearly moves the deal forward.\n4. Based on what you've learned from the call transcript, update any missing HubSpot contact information (blank fields) using the HubSpot \"Update Contact\" tool.\n\n### Rules\n\n* Be factual; **do not invent** details.\n* If data is missing, note it as “unknown”.\n* Normalize dates to `YYYY-MM-DD` (use next business day for “tomorrow”, +7 days for “next week”).\n* Keep original spelling for names/companies.\n* Use concise language; no fluff.\n\nRespond ONLY in the following JSON format:\n{\n \"Summary\": \"<120–160 word recap>\",\n \"Task name\": \"<imperative title, e.g., Schedule security review with IT lead>\",\n \"Task body\": \"<what to deliver, needed inputs, owner, due date and success criteria>\"\n}",
"options": {},
"promptType": "define",
"hasOutputParser": true
},
"typeVersion": 2.2
},
{
"id": "4fdbc6bf-5cf9-4eb1-9951-f865ec218126",
"name": "Anrufzusammenfassung protokollieren",
"type": "n8n-nodes-base.hubspot",
"position": [
1200,
224
],
"parameters": {
"type": "call",
"metadata": {
"body": "={{ $json.output.Summary }}",
"status": "COMPLETED"
},
"resource": "engagement",
"authentication": "oAuth2",
"additionalFields": {
"associations": {
"contactIds": "={{ $('Find Contact by Email').item.json.id }}"
}
}
},
"typeVersion": 2.1
},
{
"id": "23a2c237-2a38-48c8-9c6b-3ca2f07e1816",
"name": "Follow-up-Aufgabe erstellen",
"type": "n8n-nodes-base.hubspot",
"position": [
1424,
224
],
"parameters": {
"type": "task",
"metadata": {
"body": "={{ $('AI: Summarize Call & Draft Task').item.json.output['Task body'] }}",
"status": "NOT_STARTED",
"subject": "={{ $('AI: Summarize Call & Draft Task').item.json.output['Task name'] }}",
"forObjectType": "CONTACT"
},
"resource": "engagement",
"authentication": "oAuth2",
"additionalFields": {
"associations": {
"contactIds": "={{ $('Find Contact by Email').item.json.id }}"
}
}
},
"typeVersion": 2.1
},
{
"id": "8fcb293d-d6b2-4553-af27-cc340208fa8d",
"name": "Kontakt aus Transkript aktualisieren",
"type": "n8n-nodes-base.hubspotTool",
"position": [
1040,
448
],
"parameters": {
"email": "={{ $('Form: Capture Transcript').item.json['Contact email'] }}",
"options": {},
"authentication": "oAuth2",
"additionalFields": {
"city": "={{ /*n8n-auto-generated-fromAI-override*/ $fromAI('City', `Update this if the prospect mentions the city`, 'string') }}",
"country": "={{ /*n8n-auto-generated-fromAI-override*/ $fromAI('Country_Region', `Update this if the prospect mentions the country`, 'string') }}",
"jobTitle": "={{ /*n8n-auto-generated-fromAI-override*/ $fromAI('Job_Title', `Update this if the prospect mentions their job title`, 'string') }}",
"jobFunction": "={{ /*n8n-auto-generated-fromAI-override*/ $fromAI('Job_Function', `Update this if the prospect mentions their job function`, 'string') }}"
}
},
"typeVersion": 2.1
},
{
"id": "affbd7c1-1970-417b-b9e3-80ddea9a77bd",
"name": "Strukturierte Ausgabe parsen",
"type": "@n8n/n8n-nodes-langchain.outputParserStructured",
"position": [
1216,
448
],
"parameters": {
"jsonSchemaExample": "{\n\"Summary\": \"Hand from ClearPath spoke with Jordan (Plumber) regarding slow onboarding. Jordan’s company, which remains unknown, hires 25-40 new employees monthly...\",\n\"Task name\": \"Send demo invite and prep materials to Jordan\",\n\"Task body\": \"Deliver a calendar invite for the 2025-10-01 2:00 PM demo, including a short checklist for the IT lead to prepare and an outline for the two-week sandbox pilot.\"\n}"
},
"typeVersion": 1.3
},
{
"id": "f3513e1b-b8d2-45e6-a073-9ce960255765",
"name": "Haftnotiz",
"type": "n8n-nodes-base.stickyNote",
"position": [
240,
128
],
"parameters": {
"color": 7,
"width": 448,
"height": 304,
"content": "## Get call transcript and contact"
},
"typeVersion": 1
},
{
"id": "a287568d-b9ac-408a-b529-8705d5a7ea3a",
"name": "Haftnotiz1",
"type": "n8n-nodes-base.stickyNote",
"position": [
752,
128
],
"parameters": {
"color": 7,
"width": 880,
"height": 496,
"content": "## Add call summary, follow-up task and contact properties"
},
"typeVersion": 1
},
{
"id": "d15dd7dd-df8e-4e07-a23a-cacf7a8f0173",
"name": "Haftnotiz2",
"type": "n8n-nodes-base.stickyNote",
"position": [
-288,
-32
],
"parameters": {
"color": 5,
"width": 464,
"height": 656,
"content": "## AI Call Summary to HubSpot + Follow-Up Task\n\nPaste a recent call transcript and the contact’s email to auto-log a clean summary in HubSpot and create a follow-up task.\n\n## How it works\n* Capture the contact email and transcript via a form.\n* Look up the contact in HubSpot by email.\n* Send the transcript (and known contact fields) to AI to:\n * Summarize the call (120–160 words).\n * Propose a single, actionable follow-up task.\n * Suggest updates for missing contact properties (city, country, title, function).\n* Log the summary as a completed Call in HubSpot.\n* Create a Task in HubSpot with the AI-generated title/body.\n* Optionally update the HubSpot contact from the transcript hints.\n\n## How to use\n* Open the Form: Capture Transcript node and submit an email + transcript.\n* In AI: Summarize Call & Draft Task, you can tweak the prompt rules if needed.\n* Ensure your HubSpot OAuth is connected on all HubSpot nodes.\n* (Optional) Review the Update Contact from Transcript mappings before enabling in production."
},
"typeVersion": 1
},
{
"id": "c04d0fab-a1ba-4f6f-84b5-fcaf3ac7bc29",
"name": "Haftnotiz3",
"type": "n8n-nodes-base.stickyNote",
"position": [
1408,
432
],
"parameters": {
"color": 7,
"width": 288,
"height": 288,
"content": "### 💡 Customizing this workflow\n\n- Instead of a form, try using a google drive folder to fetch call transcripts\n- Take it a step further - extract the audio from the call recording using AI\n- Get creative with fields - add any addtional contact fields to the \"Update Contact\" tool and let the AI auto-populate those based on the transcript "
},
"typeVersion": 1
}
],
"pinData": {
"Form: Capture Transcript": [
{
"formMode": "test",
"submittedAt": "2025-09-26T11:15:57.804+02:00",
"Contact email": "miha.ambroz@pm.me",
"Paste transcript here": "Me: Hi prospect\r\nHans: Hi I am actually from hungary and I work as A product manager. Not sure you have that infor in the crm"
}
]
},
"connections": {
"4fdbc6bf-5cf9-4eb1-9951-f865ec218126": {
"main": [
[
{
"node": "23a2c237-2a38-48c8-9c6b-3ca2f07e1816",
"type": "main",
"index": 0
}
]
]
},
"OpenAI Chat Model": {
"ai_languageModel": [
[
{
"node": "e4ce0bc0-eaef-4d12-80af-8be92a32c059",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"62aa0111-3390-474f-9779-1d389398d3a0": {
"main": [
[
{
"node": "e4ce0bc0-eaef-4d12-80af-8be92a32c059",
"type": "main",
"index": 0
}
]
]
},
"affbd7c1-1970-417b-b9e3-80ddea9a77bd": {
"ai_outputParser": [
[
{
"node": "e4ce0bc0-eaef-4d12-80af-8be92a32c059",
"type": "ai_outputParser",
"index": 0
}
]
]
},
"1dbedf6b-1427-4127-a0bf-ac745c386e6a": {
"main": [
[
{
"node": "62aa0111-3390-474f-9779-1d389398d3a0",
"type": "main",
"index": 0
}
]
]
},
"8fcb293d-d6b2-4553-af27-cc340208fa8d": {
"ai_tool": [
[
{
"node": "e4ce0bc0-eaef-4d12-80af-8be92a32c059",
"type": "ai_tool",
"index": 0
}
]
]
},
"e4ce0bc0-eaef-4d12-80af-8be92a32c059": {
"main": [
[
{
"node": "4fdbc6bf-5cf9-4eb1-9951-f865ec218126",
"type": "main",
"index": 0
}
]
]
}
}
}Häufig gestellte Fragen
Wie verwende ich diesen Workflow?
Kopieren Sie den obigen JSON-Code, erstellen Sie einen neuen Workflow in Ihrer n8n-Instanz und wählen Sie "Aus JSON importieren". Fügen Sie die Konfiguration ein und passen Sie die Anmeldedaten nach Bedarf an.
Für welche Szenarien ist dieser Workflow geeignet?
Fortgeschritten - Kundenbeziehungsmanagement, KI-Zusammenfassung
Ist es kostenpflichtig?
Dieser Workflow ist völlig kostenlos. Beachten Sie jedoch, dass Drittanbieterdienste (wie OpenAI API), die im Workflow verwendet werden, möglicherweise kostenpflichtig sind.
Verwandte Workflows
HubSpot-Kontakte KI-Erweiterung
HubSpot-Kontakte KI-Erweiterung
Filter
Hubspot
Agent
+
Filter
Hubspot
Agent
12 NodesMiha
Kundenbeziehungsmanagement
AI-E-Mail-Antwort mit Slack-Genehmigung basierend auf HubSpot-Kontext
AI-E-Mail-Antwort mit Slack-Genehmigung basierend auf HubSpot-Kontext
If
Code
Gmail
+
If
Code
Gmail
19 NodesMiha
Kundenbeziehungsmanagement
Automatisierung der Klassifizierung und Weiterleitung von Support-Tickets
Automatisierung der Support-Ticket-Klassifizierung und -Weiterleitung von HubSpot zu Jira mit GPT
Set
Jira
Switch
+
Set
Jira
Switch
23 NodesPollupAI
KI-Zusammenfassung
Mit Ihren HubSpot-Daten (Kontakte und Transaktionen) chatten
Mit Ihren HubSpot-Daten sprechen (Kontakte und Geschäfte)
Hubspot Tool
Agent
Chat Trigger
+
Hubspot Tool
Agent
Chat Trigger
6 NodesMiha
Kundenbeziehungsmanagement
Analyse der Regionaleistung
Analyse der Verkaufs Region Leistung mit Bright Data MCP und GPT-4o
Set
Code
Gmail
+
Set
Code
Gmail
17 NodesYaron Been
Kundenbeziehungsmanagement
HireMind – KI-gesteuerter Lebenslauf-Verarbeitungsprozess
HR-KI-Lebenslauf-Screening und -Bewertung: GPT-4 und Google Workspace
If
Code
Slack
+
If
Code
Slack
26 NodesTrung Tran
Personalwesen
Workflow-Informationen
Schwierigkeitsgrad
Fortgeschritten
Anzahl der Nodes12
Kategorie2
Node-Typen7
Autor
Externe Links
Auf n8n.io ansehen →
Diesen Workflow teilen