Building AI Automations for Pipedrive with Make.com & Zapier

From Concepts to Implementations

The previous chapter explained AI integration architecture. This chapter gets practical—you'll build real automation workflows that deliver immediate value.

We'll focus on Make.com for detailed examples because it offers more flexibility for complex scenarios. Zapier alternatives are noted where the platform handles things differently.

Workflow 1: Automated Lead Enrichment

When a new lead enters Pipedrive, automatically research the company and populate custom fields with AI-generated insights.

What This Workflow Does

New deal created → Fetch company website → AI analyzes the business → Custom fields populated with company description, industry vertical, potential pain points, and decision-maker suggestions.

Setup Steps

Step 1: Create the trigger

Module: Pipedrive → Watch Deals
Select: New deals only
Filter (optional): Only deals with organization attached

Step 2: Get organization details

Module: Pipedrive → Get Organization
Organization ID: Map from trigger (organization_id)
This retrieves the company name and website.

Step 3: Fetch website content

Module: HTTP → Make a request
URL: Map organization website URL
Method: GET
Parse response: Yes
Note: Add error handling for sites that block scraping or don't exist.

Step 4: Extract text from HTML

Module: Text Parser → HTML to Text
HTML: Map from previous HTTP response body
This converts the webpage to readable text for AI processing.

Step 5: AI analysis

Module: OpenAI → Create a Chat Completion
Model: gpt-4
System message:

You are a B2B sales research analyst. Analyze company websites and provide:
1. Company description (2-3 sentences)
2. Industry vertical
3. Company size estimate (startup/SMB/mid-market/enterprise)
4. Top 3 likely pain points relevant to [your product category]
5. Recommended decision-maker titles to target

Respond in JSON format with keys: description, industry, company_size, pain_points (array), target_titles (array)

User message: Map the extracted website text (truncate if needed to fit token limits)

Step 6: Parse AI response

Module: JSON → Parse JSON
JSON: Map from OpenAI response content
This converts the AI output to usable data.

Step 7: Update deal with enrichment

Module: Pipedrive → Update Deal
Deal ID: Map from trigger
Custom fields: Map parsed JSON values to your custom fields

Custom Fields to Create in Pipedrive

• AI Company Description (Large text)
• AI Industry (Single option or text)
• AI Company Size (Single option)
• AI Pain Points (Large text)
• AI Target Titles (Large text)
• AI Enrichment Date (Date)

Error Handling

Add a router after the HTTP module to handle failures:

• Route 1: Website fetched successfully → Continue to AI
• Route 2: Fetch failed → Update deal with "Enrichment failed - manual research needed"

Workflow 2: AI-Powered Email Sequences

Generate personalized email sequences based on deal context, then schedule them for sending.

What This Workflow Does

Deal reaches "Qualified" stage → AI generates 3-email sequence → Emails stored in Pipedrive activities scheduled for future dates.

Setup Steps

Step 1: Create the trigger

Module: Pipedrive → Watch Deals
Filter: Stage changed to "Qualified"

Step 2: Gather context

Module: Pipedrive → Get Deal (full details)
Module: Pipedrive → Get Person (primary contact)
Module: Pipedrive → Get Organization
Module: Pipedrive → List Activities (recent activities on this deal)

Step 3: Generate email sequence

Module: OpenAI → Create a Chat Completion
System message:

You are an expert B2B sales copywriter. Create a 3-email sequence for a newly qualified lead.

Email 1: Value-focused introduction (send day 1)
Email 2: Case study / social proof (send day 4)  
Email 3: Soft close with meeting request (send day 7)

Each email should be:
- Under 150 words
- Personalized to the contact's role and company
- Reference their likely pain points
- Include a clear, low-friction CTA

Respond in JSON format:
{
  "emails": [
    {"subject": "", "body": "", "send_day": 1},
    {"subject": "", "body": "", "send_day": 4},
    {"subject": "", "body": "", "send_day": 7}
  ]
}

User message: Include deal value, contact name, title, company name, industry, and any notes from previous activities.

Step 4: Parse and create activities

Module: JSON → Parse JSON
Module: Iterator → Iterate over emails array
For each email:

Module: Pipedrive → Create Activity
Type: Email (or your custom type for draft emails)
Subject: Map email subject
Note: Map email body
Due date: Calculate from today + send_day
Deal ID: Map from trigger

Variation: Send Directly via Email Platform

Instead of creating Pipedrive activities, integrate with your email sending tool:

• Lemlist: Create campaign, add lead, schedule emails
• Smartlead: Add to sequence with custom dates
• Reply.io: Create personalized sequence

Workflow 3: Dynamic Deal Scoring

AI evaluates deals periodically and assigns probability scores with explanations.

What This Workflow Does

Daily schedule → Fetch all open deals → AI scores each deal → Update probability and add AI notes.

Setup Steps

Step 1: Schedule trigger

Module: Schedule → Run scenario every day at 6 AM

Step 2: Fetch open deals

Module: Pipedrive → List Deals
Status: Open
Limit: Adjust based on pipeline size

Step 3: Iterate and enrich each deal

Module: Iterator → For each deal
Module: Pipedrive → List Activities for deal
Module: Pipedrive → Get Person (primary contact)
Module: Pipedrive → Get Organization

Step 4: AI scoring

Module: OpenAI → Create a Chat Completion
System message:

You are a sales forecasting expert. Analyze deals and predict close probability.

Consider these factors:
- Activity recency and frequency
- Stakeholder engagement level
- Deal age vs typical sales cycle
- Value and complexity
- Stage progression velocity

Provide:
1. Probability score (0-100)
2. Confidence level (low/medium/high)
3. Top 3 factors influencing the score
4. Recommended action

Respond in JSON:
{
  "probability": 75,
  "confidence": "medium",
  "factors": ["Recent demo went well", "No executive sponsor yet", "Budget confirmed"],
  "recommendation": "Identify and engage C-level sponsor before proposal"
}

User message: Deal details, activity history, contact info, stage, value, age.

Step 5: Update deal

Module: Pipedrive → Update Deal
Probability: Map from AI response
Custom field "AI Score Factors": Map factors
Custom field "AI Recommendation": Map recommendation
Custom field "AI Score Date": Today

Performance Optimization

For large pipelines, this workflow can be expensive and slow. Optimize by:

• Scoring only deals that changed recently
• Using cheaper models for initial scoring, reserving GPT-4 for high-value deals
• Caching organization/contact data to reduce API calls
• Processing in batches with delays to stay within rate limits

Workflow 4: Meeting Follow-Up Automation

When a meeting activity is completed, AI generates and sends a follow-up email with next steps.

What This Workflow Does

Activity marked done (type: Meeting) → AI generates follow-up based on notes → Email sent and logged to Pipedrive.

Setup Steps

Step 1: Create trigger

Module: Pipedrive → Watch Activities
Filter: Type = Meeting, Status changed to Done

Step 2: Get context

Module: Pipedrive → Get Deal
Module: Pipedrive → Get Person
Activity note: Map from trigger

Step 3: Generate follow-up email

Module: OpenAI → Create a Chat Completion
System message:

You are a sales professional writing a follow-up email after a meeting.

Based on the meeting notes, create an email that:
1. Thanks them for their time
2. Summarizes 2-3 key discussion points
3. Confirms any agreed next steps
4. Proposes the next meeting or action
5. Keeps the tone professional but warm

Keep the email under 200 words.

Respond in JSON:
{
  "subject": "Following up on our conversation",
  "body": "Email body here"
}

User message: Meeting notes, contact name, deal details, previous context.

Step 4: Send email

Module: Gmail → Send Email (or your email provider)
To: Contact email
Subject: Map from AI
Body: Map from AI
Note: Consider adding manager CC for important deals

Step 5: Log to Pipedrive

Module: Pipedrive → Create Activity
Type: Email
Done: Yes
Subject: Map from AI
Note: "Auto-generated follow-up: " + email body

Human Review Option

For important deals, add a review step:

1. Instead of sending, create a draft email
2. Notify the rep via Slack/email with the draft
3. Wait for approval via button or trigger
4. Send only after human review

Workflow 5: Win/Loss Analysis

When deals close (won or lost), AI analyzes patterns and provides insights.

What This Workflow Does

Deal marked Won/Lost → AI analyzes deal history → Insights logged and summary sent to manager.

Setup Steps

Step 1: Create trigger

Module: Pipedrive → Watch Deals
Filter: Status changed to Won OR Lost

Step 2: Gather full deal history

Module: Pipedrive → Get Deal
Module: Pipedrive → List all Activities (no limit)
Module: Pipedrive → Get Organization
Module: Pipedrive → Get Person

Step 3: AI analysis

Module: OpenAI → Create a Chat Completion
System message:

You are a sales analytics expert. Analyze completed deals to extract learnings.

For this {{outcome}} deal, analyze:
1. Timeline: How long was the sales cycle? Was it faster/slower than expected?
2. Engagement: Activity patterns, response times, stakeholder involvement
3. Turning points: What key moments moved the deal forward or backward?
4. What worked: Tactics, messages, or approaches that helped
5. What could improve: Opportunities missed or approaches that didn't work
6. Competitor insights: Any competitor mentions and their impact
7. Replication: For wins, what's repeatable? For losses, what to avoid?

Respond in JSON:
{
  "summary": "2-sentence summary",
  "timeline_analysis": "",
  "engagement_quality": "low/medium/high",
  "turning_points": [],
  "what_worked": [],
  "improvements": [],
  "competitor_notes": "",
  "key_lesson": ""
}

User message: Complete activity history, deal details, all notes.

Step 4: Update deal with analysis

Module: Pipedrive → Update Deal
Custom fields: Map analysis components

Step 5: Notify team

Module: Slack → Post Message (or Email)
Channel: Sales-wins or Sales-learnings
Message: Formatted summary of the analysis with key lessons

Platform Comparison: Make.com vs Zapier

When to Use Make.com

• Complex workflows with multiple branches
• Heavy data transformation needs
• Cost-sensitive high-volume scenarios
• Advanced error handling requirements
• Custom HTTP requests and webhooks

When to Use Zapier

• Simple, linear automations
• Non-technical users building workflows
• Quick setup priority over customization
• Integration with apps Make.com doesn't support
• Teams already invested in Zapier ecosystem

Key Takeaways

• Lead enrichment saves 15-20 minutes per new deal with automated research
• AI-generated sequences maintain personalization at scale
• Dynamic scoring provides objective, consistent deal evaluation
• Automated follow-ups ensure no meeting lacks proper recap
• Win/loss analysis extracts learnings that would otherwise be lost