n8n vs Make vs Custom Code: 2026 Comparison

You have decided to automate. Now you need to pick a platform. The choice between n8n, Make, and writing custom code is not about which is “best.” It is about which fits your specific constraints: budget, complexity, team skills, and scaling plans.

Quick Overview: What Each Platform Does

n8n is an open-source workflow automation platform with a visual editor. It can be self-hosted on your own infrastructure or used through n8n Cloud. It is developer-friendly, highly extensible, and increasingly AI-native with built-in support for language model integrations. The community is large and growing fast. The learning curve is steeper than Make, but the ceiling is much higher.

Make (formerly Integromat) is a visual, cloud-hosted automation platform that sits between Zapier’s simplicity and n8n’s power. The execution model is based on scenarios and operations rather than tasks, which changes how pricing scales. It handles complex logic including iterators, aggregators, and multi-branch routing. Non-technical teams can build reasonably sophisticated workflows without developer involvement.

Custom code means building your automations as software, typically using Python or Node.js, hosted on your own infrastructure or a cloud provider. Maximum flexibility, maximum responsibility. Nothing is off-the-shelf. Everything is exactly as you build it.

Pricing Comparison at Real Scale

Headline pricing is misleading. What matters is cost at the volume your business actually runs.

	n8n Cloud	n8n Self-hosted	Make	Custom code
Entry tier	£20/month (2,500 executions)	Hosting cost only: ~£10-30/month	£9/month (10,000 ops)	Dev cost + hosting
At 50k ops/month	~£100/month	~£15-40/month hosting	~£45/month	Variable
At 500k ops/month	~£500/month	~£30-80/month hosting	~£200/month	Variable
Maintenance burden	Low	Medium (self-managed)	Low	High

The self-hosted n8n model becomes very attractive at scale. Your hosting cost is essentially fixed regardless of execution volume. For high-volume automations, the difference between £500/month on n8n Cloud and £40/month self-hosted adds up to significant savings over a year.

Make’s operations model counts every action as an operation, not just each workflow run. A workflow that sends an email, updates a CRM record, and queries a database counts as 3 operations. Plan accordingly when estimating costs for complex workflows.

Custom code has no platform fees but carries development and maintenance costs. At low complexity, this is usually the most expensive option. At very high volume or unique requirements, it becomes the most cost-effective.

Flexibility and Limitations

Make’s limitations: Execution time caps at 40 minutes per scenario on most plans. Complex, long-running processes need to be broken into multiple connected scenarios. The visual interface is excellent for moderate complexity but becomes unwieldy for very complex logic. Error handling options are functional but less configurable than n8n or custom code.

n8n’s limitations: The learning curve is real. Getting productive with n8n takes longer than Make. Self-hosted deployments require technical capability to set up, maintain, and update. The visual interface can become cluttered with very large workflows. Community support is strong but commercial support is more expensive than you might expect.

Custom code’s limitations: Every feature must be built and maintained. No visual interface for non-technical stakeholders. Higher build cost for standard integrations. When a connected API changes, you fix it yourself. Debugging requires developer time rather than a platform support ticket.

When to Use Each

Use Make when:

Your team includes non-technical people who will build or maintain workflows
You need something working quickly with minimal setup
Your automation needs are in the “marketing and operations” category: CRM updates, email sequences, form handling, basic data routing
Workflows are moderately complex but do not require loops over large datasets or heavy computation
You want cloud-hosted with no infrastructure responsibility

Use n8n when:

Your team has technical capability or you have a developer building and maintaining workflows
You need self-hosted due to data residency requirements (relevant for GDPR-conscious UK businesses)
Workflows are complex: branching logic, iteration over large datasets, AI model integration
Volume is high enough that per-operation cloud pricing becomes expensive
You want an active open-source community and the ability to build custom nodes

Use custom code when:

Your automation requires logic that no platform handles well (unusual business rules, proprietary algorithms, integration with internal systems that have no standard connectors)
Volume is very high and platform costs become significant at scale
You need maximum reliability with custom monitoring, retry logic, and error handling
The automation represents a competitive advantage you do not want to expose to a third-party platform
You have technical resources who can own the maintenance

The Hybrid Approach

Most sophisticated operations end up using more than one approach. This is not a compromise. It is the sensible architecture.

Make for the marketing stack: lead capture, CRM updates, email sequences, Slack notifications. These are Make’s sweet spot and non-technical marketers can maintain them.

n8n for operations: invoice processing, data pipelines, complex approval workflows, AI-assisted document handling. Technical team builds and maintains these. Self-hosted for data control.

Custom code for critical business logic: the proprietary scoring algorithm, the bespoke client portal integration, the high-volume data transformation that needs to run a million times per day reliably.

The platforms coexist without conflict. They each handle the layer they are best suited for.

Before Choosing a Platform

The platform decision should come after the process is mapped. It is surprisingly common for businesses to choose a platform first and then discover that their processes need more (or less) than the platform can handle. Our workflow mapping guide covers how to document your processes in a way that makes the platform decision obvious rather than agonised.

Our advisory service includes platform recommendations based on your specific process map, team capabilities, and growth plans. We work with all three approaches and recommend based on fit, not preference.

Not sure which platform fits your situation? Get a platform recommendation based on your actual requirements.