AI CRM Integration: Automate Contact Management & Follow-ups

AI CRM Integration

Use Cases

1. Lead enrichment: Auto-fill contact data
2. Email generation: Personalized outreach
3. Meeting summaries: Auto-update CRM
4. Next best action: AI recommendations
5. Sentiment analysis: Prioritize hot leads

HubSpot + AI

Zapier Workflow

Trigger: New contact in HubSpot
→ ChatGPT: Generate personalized email
→ HubSpot: Create email task
→ Slack: Notify sales rep

API Integration

import requests
from openai import OpenAI

HUBSPOT_API_KEY = "your-key"
client = OpenAI()

def enrich_contact(contact_id):
    # Get contact from HubSpot
    contact = get_hubspot_contact(contact_id)

    # AI enrichment
    prompt = f"""Based on this info, suggest:
    - Industry
    - Company size
    - Pain points
    - Best approach

    Contact: {contact['company']}, {contact['title']}
    """

    suggestions = client.chat.completions.create(
        model="gpt-4",
        messages=[{"role": "user", "content": prompt}]
    ).choices[0].message.content

    # Update HubSpot
    update_hubspot_contact(contact_id, {"notes": suggestions})

Salesforce + AI

Einstein Integration

• Built-in AI features
• Lead scoring
• Opportunity insights

Custom AI

from simple_salesforce import Salesforce

sf = Salesforce(username='...', password='...', security_token='...')

def ai_lead_scoring(lead_id):
    lead = sf.Lead.get(lead_id)

    score_prompt = f"""Score this lead (0-100):
    Company: {lead['Company']}
    Industry: {lead['Industry']}
    Title: {lead['Title']}
    Engagement: {lead['Email_Opens__c']} opens
    """

    score = get_ai_score(score_prompt)

    # Update Salesforce
    sf.Lead.update(lead_id, {'AI_Score__c': score})

Pipedrive + AI

n8n Workflow

{
  "nodes": [
    {
      "name": "Pipedrive Trigger",
      "type": "n8n-nodes-base.pipedriveTrigger",
      "webhookId": "new_deal"
    },
    {
      "name": "AI Analysis",
      "type": "n8n-nodes-base.openAi",
      "parameters": {
        "operation": "message",
        "text": "Analyze this deal and suggest next steps: {{$json.title}}"
      }
    },
    {
      "name": "Update Pipedrive",
      "type": "n8n-nodes-base.pipedrive",
      "parameters": {
        "operation": "update",
        "resource": "deal",
        "notes": "={{$json.choices[0].message.content}}"
      }
    }
  ]
}

Meeting Transcription → CRM

def meeting_to_crm(audio_file, deal_id):
    # Transcribe
    transcript = client.audio.transcriptions.create(
        model="whisper-1",
        file=audio_file
    )

    # Extract action items
    analysis = client.chat.completions.create(
        model="gpt-4",
        messages=[{
            "role": "user",
            "content": f"""Extract from meeting:
            - Key discussion points
            - Action items (who, what, when)
            - Next meeting date
            - Deal stage recommendation

            Transcript: {transcript.text}"""
        }]
    )

    # Update CRM
    update_deal(deal_id, analysis.choices[0].message.content)

Smart Follow-ups

def generate_followup(contact_id, last_interaction):
    context = get_contact_history(contact_id)

    email = client.chat.completions.create(
        model="gpt-4",
        messages=[{
            "role": "user",
            "content": f"""Write follow-up email:
            Contact: {context['name']} at {context['company']}
            Last interaction: {last_interaction}
            Their interest: {context['notes']}
            Our product: {product_description}

            Goal: Move to next stage
            Tone: Professional, helpful
            Length: <150 words"""
        }]
    )

    return email.choices[0].message.content

---

Found an issue? Open an issue!