n8n vs Make: self-host and code, or polished cloud scenarios?

Short answer: choose Make when you want a polished, managed visual builder for cloud scenarios — routers, filters, thousands of apps, and a credit-based bill your ops lead can explain without standing up infrastructure. Choose n8n when you need self-host or data-residency control, code nodes that do not feel like a dead end, execution-based economics on complex multi-step flows, or AI agent pipelines you can inspect on a canvas. If you are still deciding whether Zapier is good enough, stop and read Zapier vs Make first. This page is the next fork after that.

This is a studio and RevOps decision, not a feature laundry list. Integration counts move. Plan names move. The decision that lasts is whether your bottleneck is builder accessibility on a managed cloud, or control, code, and where the workflow actually runs.

Both are on Unloc: n8n, Make. Zapier stays in the cluster when linear Zaps are still the comparison: Zapier. Browse more under Automation on /tools.

Quick verdict

If you cannot say whether the artefact is a marketing-friendly scenario or a controllable agent pipeline on your infrastructure, you will pick the wrong platform and then blame the nodes. Name the artefact and the residency requirement first.

What Make is

Make (make.com) is a managed visual automation platform: you build scenarios as graphs of modules, with routers, filters, error handlers, and a large catalogue of app connectors. The product pitch for studios is speed to a working cloud workflow without standing up servers — lead routing, content ops, CRM sync, Slack alerts, and increasingly AI modules and agents inside the same canvas.

Public Make pricing describes Free, Core, Pro, Teams, and Enterprise. Billing is credit-based: as of Make’s public note (27 August, year on their pricing FAQ), credits are the unit — most module actions consume credits, with some AI-provider features using more. Free includes a monthly credit allowance and limited active scenarios; paid tiers raise credits, scheduling granularity, collaboration, and governance. Read Make pricing the week you buy. We are not printing sticky dollar amounts.

Make’s durable advantage is the polished cloud experience and breadth of ready-made apps. The durable constraint is that you are on Make’s cloud and Make’s economics: complex, high-volume, multi-module scenarios burn credits per action, which is a different shape from “one run = one billable unit.” Model that before you migrate a chatty Zapier estate.

Make is not self-host Make. If data residency or air-gapped control is a hard requirement, you are already leaning toward n8n’s self-host story (or another on-prem automation stack). Make’s on-prem agent and Enterprise security features help some enterprise network cases — they are not the same as owning the runtime.

What n8n is

n8n is an open-source-friendly workflow automation platform aimed at technical teams: build visually, drop into JavaScript or Python when you need to, connect HTTP and custom APIs, and run AI agents with inspectable steps. You can use n8n Cloud or self-host the Community edition (and licensed Business / Enterprise features on your own infrastructure).

Public n8n Cloud pricing is organised around Starter, Pro, Business, and Enterprise, with pricing based on monthly workflow executions — a full run of a workflow counts as one execution regardless of how many steps it contains, per n8n’s public pricing explanation. Self-host Community is the path when you want the runtime on your machines; Business and Enterprise add collaboration, SSO, environments, queue mode, and governance. Read n8n pricing and the product pages the week you buy.

n8n’s durable advantage is control: code when you need it, self-host when residency matters, and execution economics that favour complex multi-step workflows compared with per-action credit models. The durable cost is ownership — someone has to run, monitor, and upgrade a self-hosted instance, or you pay for Cloud and accept n8n’s hosting. “Open source” is not the same as “free of ops.”

n8n markets AI agent nodes, MCP-related triggers/clients, evaluations, and human-in-the-loop patterns. Treat those as product capabilities to verify on your plan, not as a promise that every LangChain tutorial will paste cleanly into production.

Side-by-side

Do not collapse this into which has more logos on a marketing page. Catalogue size helps day one. Code and hosting decide whether you are still happy at month six.

Pricing snapshot

Publish note: plan names and unit economics below reflect official pages checked 2 October 2026. Numbers move. We name tiers and billing shapes only — no sticky dollars in this article.

Make (pricing): Free, Core, Pro, Teams, Enterprise. Credits are the billing unit for module actions in scenarios. Free is limited (credits, active scenarios, schedule interval). Paid tiers unlock unlimited active scenarios (per their table), finer schedules, collaboration, and Enterprise governance. Overage and extra-credit behaviour is defined on Make’s FAQ — read it before you scale a noisy scenario.

n8n (pricing): Cloud Starter, Pro, Business, Enterprise — execution-based, unlimited users/workflows/integrations on the public Cloud framing, with concurrency and insight retention rising by tier. Community Edition for standard self-host. Business/Enterprise license keys apply when you need those features on your own infra.

Conceptual ops-vs-execution (why bills surprise people):

• In Make, a scenario that touches ten modules ten times can consume many credits in one run. Branching and iterators multiply actions.
• In n8n Cloud, one successful end-to-end run is typically one execution even if the graph is deep — which is why complex AI pipelines often look cheaper on paper once volume is high. Self-host shifts cost to infrastructure and people.
• Neither model is “always cheaper.” Cheap is the workflow volume × complexity × failure retries you actually run. Prototype three real scenarios and estimate units on both pricing pages before you migrate.

Who Make is best for

Make is for studios and operators who want automation without becoming platform engineers. Marketing ops connecting form → CRM → Slack. Content teams pushing approved copy into three channels. RevOps building visible scenarios a non-developer can adjust. Teams that liked Zapier’s accessibility but need routers, more control flow, and a visual graph — and who already accepted that Zapier vs Make is the prior decision (read that page).

It is also the honest pick when leadership will not fund self-hosting, and when the data in the scenarios is already living in SaaS tools with similar residency posture.

Not ideal if

• Client or regulated data must stay on your infrastructure. Managed cloud is the product.
• Your builders are developers who will outgrow no-code modules in a week and resent credit burn on chatty graphs.
• You need deep custom code as the default, not an exception.
• You assumed Make can be self-hosted like n8n Community. It cannot, as a full product equivalent.

Who n8n is best for

n8n is for technical studios, product ops, and security-conscious teams. Self-host when residency or VPC placement is a hard requirement. Cloud when you want n8n’s execution model without running Docker yourself. Code nodes when the integration catalogue is not enough. AI agent pipelines when you need to see every tool call on the canvas and keep humans in the loop.

It is the honest pick when Make’s credit model would punish a multi-step enrichment pipeline, or when Zapier’s linear model already failed you and the next question is control — not another polished catalogue.

Not ideal if

• Nobody on the team can own a workflow that breaks at 17:00 on a Friday. Self-host multiplies that risk.
• The only builders are marketers who need templates and a gentle UI above all else — start with Make (or Zapier) unless residency forces n8n.
• You wanted “free forever” without counting engineer time. Community is free of license fees, not free of ownership.
• You expected every SaaS logo from Make’s homepage to exist as a first-class node on day one. HTTP nodes close gaps; they are still work.

Studio workflows

Lead routing (form → CRM → Slack)

Make is usually faster to a reliable first version if your CRM and Slack connectors are first-class and the graph is shallow. n8n is better when enrichment steps multiply, when you need custom scoring code, or when form payloads must not leave your network. Either way: log failures somewhere humans see, and do not silently drop leads when credits or executions run out.

Content ops (brief → draft queue → publish checklist)

Make fits editorial ops who already live in Airtable / Notion / Drive style stacks and want visible scenarios. n8n fits when publishing needs custom API calls, branchy approvals, or an AI step you must evaluate before it posts. Keep a human approval node before anything customer-facing ships.

AI agent pipeline (research → tool calls → ticket)

n8n’s agent and evaluation story is the better conceptual fit when the pipeline is the product. Make’s AI modules and agents can cover lighter cases inside scenarios. Whichever you pick: constrain tools, log prompts and outputs, and do not let an agent write to production CRM fields without a review path.

Client-data residency path

If the client contract says data stays in your VPC or region you control, prefer n8n self-host (or another on-prem stack) and document where credentials live. Make Enterprise features and on-prem agents may help specific network access cases — verify with Make and your counsel; do not assume parity with self-hosting the automation runtime.

Ownership and on-call

Automation without an owner is a silent outage. Name a primary and a backup for every production scenario or workflow. Write where credentials live, what “success” means for a run, and who gets paged when the CRM write fails. Make’s visual history helps non-developers spot a red module. n8n’s execution log helps technical owners replay a step. Neither replaces a human who understands the business rule. If your studio cannot staff that role, prefer fewer workflows over a more powerful platform.

Templates versus systems

Both vendors ship templates. Templates are demos with your logo risk attached. Promote a template to production only after you have changed the error path, added alerting, and tested the empty-payload case. Studios that “install twelve templates on day one” inherit twelve unknown failure modes. Start with one revenue-critical path. Earn complexity.

Migrating from Zapier — what actually breaks

If you are leaving Zapier, read Zapier vs Make for the Zapier-versus-Make decision. This section is only what breaks when you move toward Make or n8n after Zapier.

What usually breaks:

• Mental model: Zaps are linear steps. Make scenarios and n8n workflows are graphs. Routers and branches are powerful; they also hide paths you forget to test.
• Billing units: Zapier tasks, Make credits, and n8n executions are not interchangeable. Rebuild cost estimates from scratch on three real workflows. Do not multiply old Zapier task counts by a blog’s rule of thumb.
• Auth and connections: Every app connection is re-authorised. Shared Zapier connections do not magically become Make or n8n credentials. Plan a connection inventory.
• Built-in apps and Paths / Interfaces / Tables: Zapier product surfaces beyond basic Zaps may not have one-to-one equivalents. Replace the job, not the feature name.
• Error handling: Zapier’s failure email is not a strategy. Make error handlers and n8n error workflows must be designed on purpose or you will discover breaks from a client.
• Owner risk: The person who built the Zapier estate may not be the person who can debug n8n code nodes. Train or hire before cutover.

A clean migration pattern: freeze new Zaps, rebuild the top ten by business impact in the new tool, run dual-write or shadow runs for a week, then cut over with a kill switch. Do not big-bang fifty Zaps on a Sunday night.

Questions people actually search

Is n8n better than Make in 2026?

For self-host, code-heavy, or complex multi-step execution economics, n8n is often the better fit. For polished managed scenarios and non-developer builders, Make is often better. Better means better at your artefact and residency needs, not a permanent winner.

Make vs n8n pricing — which is cheaper at high volume?

It depends on whether your workflows are action-chatty or deep-but-infrequent. Make credits scale with module actions; n8n Cloud executions scale with runs; self-host shifts cost to infra. Prototype real volumes on both official pricing pages. We do not invent a crossover dollar figure here.

Can I self-host Make like n8n?

No — not as a full Make equivalent. Make is a managed cloud product with Enterprise networking options. n8n offers a real self-host Community path plus licensed features for self-hosted Business/Enterprise. If self-host is mandatory, n8n is the fork on this page.

n8n vs Make vs Zapier — how do they differ?

Zapier optimises for simple linear automation and the largest habit surface. Make optimises for visual cloud scenarios with richer logic. n8n optimises for technical control, code, and optional self-host. Start with Zapier vs Make if you are still on Zapier; use this page once Zapier is out of the shortlist.

Which is easier for non-developers, Make or n8n?

Make, for most marketing and ops builders. n8n is approachable for technical ops but expects more comfort with APIs, data shapes, and debugging. Ease is not the same as correctness under failure.

Is n8n good for AI agents and LangChain workflows?

n8n is explicitly positioned for AI agents you can inspect, with nodes, MCP-related pieces, and evaluations on the current product story. LangChain and similar frameworks may still live in code you call from n8n. Verify the exact nodes on your version; do not assume every notebook pastes in.

Does Make have more integrations than n8n?

Make’s marketing emphasises thousands of apps; n8n emphasises nodes plus HTTP/code for everything else. Catalogue size favours Make for click-to-connect day one. Extensibility favours n8n when the app is missing. Count the connectors you actually need, not the homepage number.

Should a studio switch from Make to n8n?

Switch when residency, code depth, agent pipelines, or credit burn on complex graphs become the bottleneck — and when someone can own n8n. Do not switch because a Twitter thread called n8n more powerful. Power without an owner is an outage.

How to decide this week

Write down three workflows that matter: trigger, systems touched, whether client data is in the payload, and who must be able to edit the graph. Rebuild the smallest one in Make and in n8n (Cloud trial or local Community). Estimate credits versus executions using each vendor’s definitions. If residency is non-negotiable, drop Make from the residency-sensitive path immediately.

Pick Make when the team will actually maintain visual cloud scenarios. Pick n8n when control, code, or self-host is the job. Keep Zapier only if the estate is still simple — and be honest when it is not. Cancel the dual-running automation platform you stopped opening.

Make and n8n are on Unloc under automation. Organic listings are not ranked by who paid. This piece does not use affiliate links. Links go to Make, n8n, Zapier, and the listings on this site.