Make.com Case Studies: Real Businesses, Real Results

Case Study 1: B2B SaaS Lead Gen Agency

Context: A 5-person lead generation agency serving B2B tech companies. Previously spending 40 hours/week on manual prospecting and outreach.

Problem: Lead research was taking 2-3 hours per prospect. Writing personalised outreach took another 30 minutes. Sequences fell behind because the team couldn't keep up.

Make.com solution:

The agency built a three-Scenario automation stack:

1. Lead sourcing Scenario: Apollo.io API → pull leads matching client ICP → Clearbit enrich → Airtable log with full company data

1. Personalisation Scenario: Triggered when new Airtable row created → Claude: generate 3-paragraph personalised outreach email referencing specific company data → Gmail: create draft in client's inbox for final review → HubSpot: create contact

1. Follow-up Scenario: Daily schedule → check which leads received email 1 but no reply after 3 days → Claude: generate follow-up → Gmail: send → update Airtable

Results:

• Lead research time: 2-3 hours → 5 minutes (95% reduction)
• Outreach capacity: 20 leads/week → 200 leads/week
• Response rate: maintained (personalisation quality preserved by Claude)
• Team capacity freed for strategy and client work

Case Study 2: E-commerce Brand Support

Context: A DTC brand with 50,000 customers, handling 200+ support emails per day with a 3-person support team.

Problem: 80% of support tickets were repeat questions (order status, returns, shipping). Team was drowning in tickets and response times were 48+ hours.

Make.com solution:

1. Classification Scenario: Gmail trigger → Claude: classify into 8 categories (order status, return request, product question, complaint, etc.) + extract order ID if present

1. Auto-response Scenario: For routine categories (order status, shipping estimates) → Shopify API: fetch order data → Claude: write personalised response with specific order details → Gmail: send immediately → Zendesk: close ticket

1. Escalation Scenario: For complaints and complex issues → Claude: assess sentiment and urgency → Zendesk: create P1/P2 ticket → Slack alert with context summary

Results:

• Auto-resolved: 65% of tickets (no human involved)
• Response time on auto-resolved: 3 minutes vs 48 hours
• Human team focused on: complaints, complex issues, high-value customers
• CSAT: improved from 3.8 to 4.4/5 (faster responses + consistent quality)

Case Study 3: Content Agency Scaling

Context: A content marketing agency producing SEO content for 15 B2B clients. Each client needed 8-12 articles/month. Team of 6 writers couldn't scale.

Problem: Writers were spending 2 hours on research + outline before writing a single word. The "writing" itself was the smaller part of the work.

Make.com solution:

1. Brief-to-outline Scenario: Airtable trigger (new content brief) → Semrush API: keyword data + SERP analysis → Claude: generate detailed research brief with outline, sections, key points, data to include → Airtable: update row with brief

1. Outline-to-draft Scenario: Triggered when brief status = "Ready" → Claude: generate 2000-word first draft → Claude (second call): generate meta + social captions → WordPress: create draft → Slack: notify writer

1. Post-publish Scenario: WordPress webhook → Claude: generate social media content (LinkedIn, Twitter, email newsletter) → Buffer: schedule for the week

Results:

• Research/outline time per article: 2 hours → 10 minutes
• Writers now start with a complete draft, not a blank page
• Output capacity: 8-12 articles/client → 20-25 articles/client
• Writer satisfaction increased (more time on craft, less on admin)

Case Study 4: Recruiting Firm

Context: A technical recruiting firm with 3 recruiters, placing 10-15 candidates/month.

Problem: Candidate research (GitHub, LinkedIn, technical skills assessment) took 45 minutes per candidate. Initial outreach had low response rates because messages were generic.

Make.com solution:

1. Candidate research Scenario: Webhook (new candidate referral) → GitHub API: fetch public repos and activity → LinkedIn (Phantombuster) → Claude: generate candidate assessment (technical strengths, relevant projects, likely interests, personalised outreach angle) → Airtable: log comprehensive profile

1. Personalised outreach Scenario: Triggered on profile ready → Claude: write personalised InMail/email referencing specific GitHub projects and skills → Recruiter Slack DM: "Review and send this outreach" with one-click approve

1. Interview prep Scenario: When candidate moves to interview stage → Claude: generate client-specific prep guide based on job requirements and candidate profile → Email to candidate

Results:

• Research per candidate: 45 min → 8 min
• Outreach response rate: 12% → 31% (more personalised)
• Placements/month: 10-15 → 20-25
• Client satisfaction improved (better-prepared candidates)

Key Lessons from These Case Studies

Start with your highest-friction task. All four cases targeted the process taking the most time. The ROI was immediately visible.

Human review is not a weakness. In Cases 1 and 4, humans review AI drafts before sending. This maintained quality while dramatically increasing speed. Don't skip human review for high-stakes communications.

Volume multiplies value. The more a process runs, the more valuable the automation. Cases 2 and 3 had the highest ROI because they processed hundreds of items per day.

Claude's personalisation beats templates. All four cases used Claude to write context-specific content. Generic templates would have produced lower results.

Build in quality monitoring. All four teams tracked quality metrics (response rate, CSAT, placement rate) to verify the automation maintained the quality they needed.

Recommended Tools

• Make.com — Automation platform across all cases
• Claude API — AI personalisation and analysis
• HubSpot — CRM in sales and support cases
• Airtable — Data management across all cases
• Apollo.io — Lead sourcing in Case 1
• Clearbit — Enrichment in Cases 1 and 4