Migrating from Zapier/Make to n8n: The Complete Migration Playbook

You've decided to migrate from Zapier or Make to n8n. Smart move — you'll save 70-90% on automation costs. But migration is the hard part. Here's the playbook, refined from dozens of real migrations.

Phase 0: The Audit (1-2 Days)

Before touching n8n, document everything you're migrating.

Step 1: Export All Zaps/Scenarios

Zapier: Settings → Export Data → Export all Zaps (CSV)

Make: Scenario list → Export (JSON)

Step 2: Categorize by Complexity

| Category | Description | Migration Effort | |---|---|---| | Simple | Trigger → 1 action. E.g., Gmail → Slack | 30 min each | | Medium | Trigger → filter/transform → 2-3 actions | 1-2 hours each | | Complex | Multi-path, code steps, 5+ actions | 3-5 hours each | | Critical | Revenue-affecting, can't fail | Add 50% buffer for testing |

Step 3: Prioritize

Migrate in this order:

1. Non-critical simple Zaps — Build momentum, learn n8n
2. Medium Zaps — Your core automations
3. Complex Zaps — Give yourself time to re-architect
4. Critical Zaps — Parallel run (both Zapier + n8n) for 2 weeks before cutting over

Phase 1: Set Up n8n (1 Day)

Follow our self-hosting guide for a $5/month VPS setup. You'll need:

• Ubuntu 24.04 VPS
• Docker + Docker Compose
• Caddy for SSL
• PostgreSQL (optional, but recommended for migration — easier backup/restore)

For migration, set up a staging n8n instance first. Don't migrate straight to production.

Phase 2: Rebuild Workflows (1-3 Weeks)

Zapier → n8n Translation Table

| Zapier Concept | n8n Equivalent | |---|---| | Trigger | Trigger node (same concept) | | Action | Regular node | | Filter | IF node | | Formatter | Set node or Function node | | Paths | Switch node | | Code step | Function node (full JavaScript) | | Delay | Wait node | | Webhook | Webhook node |

Pattern: Simple Zap → n8n

Zapier Zap: New Gmail → Filter: from boss → Create Google Sheets row

n8n equivalent:

1. Gmail Trigger node (poll for new emails)
2. IF node: {{ $json.from }} contains "boss@company.com"
3. Google Sheets node: Append row

Build time: 10 minutes.

Pattern: Multi-Path Zap → n8n

Zapier: New Typeform entry → Path A (positive) → Slack #wins | Path B (negative) → Slack #urgent

n8n equivalent:

1. Webhook node (Typeform sends POST)
2. Set node: extract feedback text
3. Switch node: route based on sentiment (positive / neutral / negative)
4. Slack node A (#testimonials)
5. Slack node B (#urgent-support)

Build time: 20 minutes.

Pattern: Code Step → Function Node

Zapier Code Step (JavaScript): ```javascript const price = parseFloat(inputData.price); const discount = price > 100 ? price * 0.1 : 0; return { finalPrice: price - discount }; ```

n8n Function node (identical code): ```javascript const price = parseFloat($json.price); const discount = price > 100 ? price * 0.1 : 0; return { json: { finalPrice: price - discount } }; ```

Nearly identical. But n8n supports full NPM imports, async/await, and multiple output items — capabilities Zapier's sandboxed code steps can't touch.

Phase 3: Data Migration

If your Zapier Zaps reference stored data (Storage by Zapier), export it:

```bash

Zapier storage export

curl https://store.zapier.com/api/records?secret=YOUR_SECRET ```

In n8n, use Workflow Static Data or a database:

```javascript // n8n equivalent — static data per workflow const staticData = $getWorkflowStaticData('global'); staticData.lastProcessedId = '12345'; ```

Phase 4: Parallel Run (2 Weeks)

For critical workflows, run both systems simultaneously:

1. Keep the Zapier Zap active
2. Activate the n8n workflow
3. Compare outputs daily (spot-check or automated comparison)
4. After 2 weeks with zero discrepancies, deactivate the Zapier Zap

This is the safest cutover strategy. Yes, you're paying for both for 2 weeks. Worth it to avoid a silent failure on a revenue-critical workflow.

Phase 5: Team Training (Ongoing)

Your team needs to know how to use n8n. Key topics:

• Editor basics: 1-hour walkthrough
• Credential management: How to add API keys without exposing them
• Workflow activation/deactivation: Don't edit live workflows
• Error monitoring: Where errors show up, how to debug
• Importing templates: How to use FlowForge templates as starting points

Common Migration Pitfalls

Pitfall #1: Underestimating Filter Logic

Zapier's Filters are simple AND/OR. n8n's IF nodes are more powerful but require learning the expression syntax. If you have complex filters, budget extra time.

Pitfall #2: Polling Interval Differences

Zapier polls every 1-15 minutes depending on your plan. n8n self-hosted can poll as frequently as you want, but every 1 minute is the practical minimum. Adjust your expectations — some triggers will fire faster on n8n.

Pitfall #3: Webhook URL Changes

If you have Zaps triggered by external webhooks (Stripe, GitHub), you'll need to update those webhook URLs to point to your n8n instance. Plan a coordinated cutover.

Pitfall #4: Overlooking Idempotency

n8n doesn't automatically deduplicate trigger events. If Gmail triggers on the same email twice, n8n runs twice. Add idempotency checks for critical workflows (see our error handling guide).

The Numbers: Before and After

A typical migration of 25 Zaps:

| | Zapier (Professional) | n8n Self-Hosted | |---|---|---| | Monthly cost | $49 + premium app fees (~$80) | $5 (VPS) | | Task limit | 2,000 tasks | Unlimited | | Annual cost | ~$960 | ~$60 | | 3-year cost | ~$2,880 | ~$180 | | Data ownership | On Zapier servers | On your servers |

Savings: $900/year. $2,700 over 3 years.

Ready to Migrate?

The hardest step is starting. Audit your Zaps this week. Pick the simplest one and rebuild it in n8n tonight. You'll finish in 30 minutes and wonder why you didn't migrate sooner.

Browse FlowForge templates → — all 25 n8n workflows are direct replacements for common Zapier/Make automations. Import, connect your accounts, and you're live in under 20 minutes.