n8n vs Zapier vs Make is a choice between three ways of counting work. Zapier bills tasks, usually one per successful action, and connects 9,000+ apps. Make bills credits per module action. n8n bills one execution per full workflow run, or nothing on its self-hosted Community Edition. Breadth favors Zapier, visual logic favors Make, volume and control favor n8n.
Most comparisons rank the three tools on ease of use and stop there. The numbers that decide a purchase sit one layer down: the billing unit, the AI meter, the hosting model and the person who will maintain the workflows a year from now. This guide works through each layer, with a worked example you can rerun on your own workflows.
n8n vs Zapier vs Make starts with how each one counts work
The billing unit is the biggest difference between the three platforms. Zapier charges a task each time a Zap successfully completes an action, and some apps or actions use more than one; triggers, polling, filters, paths and Formatter steps do not count (Zapier pricing). Zapier now publishes a rate for every step type, so its usage report is the place to check real consumption (Zapier task usage rates).
Make charges credits each time a module performs an action, and that includes reading data from an app or webhook and transforming data with its built-in tools. Routers and error handlers are free, and unused credits expire at the end of the term (Make pricing). n8n bills a workflow execution, which it defines as a single run of the entire workflow, however many steps it has (n8n pricing).
A worked cost example at three volumes
The fastest way to compare is to convert one real workflow into each vendor's unit. Take a lead-routing workflow with one trigger and five standard actions: enrich the lead, score it, create the CRM record, assign an owner and post a Slack alert. The table below counts what that workflow consumes at three monthly volumes.
| Runs per month | Zapier tasks | Make credits | n8n executions | Smallest n8n cloud plan that fits |
|---|---|---|---|---|
| 500 | 2,500 | 3,000 | 500 | Starter, 20 euros a month (2,500 executions) |
| 2,000 | 10,000 | 12,000 | 2,000 | Starter, 20 euros a month (2,500 executions) |
| 10,000 | 50,000 | 60,000 | 10,000 | Pro, 50 euros a month (10,000 executions, no headroom) |
Two patterns show up. Zapier and Make units grow with every step you add, so a six-step workflow at 10,000 runs needs a 50,000-task Zapier tier or about 60,000 Make credits, while n8n still counts 10,000 executions. At low volume the gap matters less, and Zapier's free logic steps can make it the cheapest option for short workflows. Zapier also sells tiers from 100 to 2 million tasks a month, so read the tier price for your number on its pricing page before deciding.
List prices at the time of writing: Zapier Professional starts from $19.99 a month billed yearly ($29.99 month to month) and Team from $69 billed yearly ($103.50 month to month), with Free and Professional limited to one user. n8n Starter costs 20 euros a month for 2,500 executions, Pro 50 euros for 10,000 and the self-hosted-only Business plan 667 euros for 40,000, all billed annually, with unlimited users on every plan. Make's Free plan includes 1,000 credits a month, and since November 6, 2025, extra Make credits cost 25 percent more than the credits in your plan (Make Community announcement). Price twelve months at your real volume, overage included.
AI workflow automation adds a second meter
AI steps carry their own cost on every platform, and each vendor meters them differently. Zapier prices AI by Zapier steps by model tier: Standard at 1x, Advanced at 3x and Premium at 5x, with Advanced as the default for new steps and a pause for approval if one step reaches 75 tasks in a single run (Zapier Help Center). A Zapier MCP tool call uses two tasks, and Zapier Agents sit on a separate plan, with Agents Pro at $33.33 a month billed annually for 1,500 activities.
n8n includes its AI Agent node, MCP server and client, and human approval for tool calls on every cloud plan, so an agent run still counts as one execution. Model tokens are a separate cost, paid to your AI provider or drawn from the gateway credits included on Starter and Pro. Make opened bring-your-own AI provider connections to all paid plans in November 2025, so teams can pay OpenAI or Anthropic directly. If you are deciding how to budget agents more broadly, our breakdown of AI agent pricing models covers seat, usage and outcome pricing.
Self-hosting turns the choice into a data control decision
Hosting is where n8n separates from both rivals. A standard self-hosted Community Edition is free on GitHub, n8n's hosted plans store data on servers in Frankfurt, Germany, and self-hosted data lives wherever you put it. The trade is operational: your team owns uptime, upgrades and backups. Paid self-hosted licenses also ping n8n's license server daily with production execution counts, and telemetry is on by default unless you disable it.
Zapier and Make run as hosted services, so governance comes from plan features. Zapier reserves custom data retention and advanced admin and app controls for Enterprise, while n8n puts SSO, LDAP, Git version control and multiple environments in its self-hosted Business plan. Whichever tool you choose, write down who can connect which apps before automations or agents touch customer records. The checklist in our guide to AI agent governance applies to automation platforms as much as to agents.
Zapier alternatives and app coverage
n8n has drawn heavy investor interest. The Berlin company raised a $180 million Series C at a $2.5 billion valuation in October 2025, led by Accel with Nvidia's NVentures participating, and said it had over 3,000 enterprise clients (Tech.eu). Its pricing page shows more than 200,000 GitHub stars for the source-available Community Edition.
Funding does not settle fit. Zapier still claims the widest catalog at 9,000+ app integrations, and Make advertises 3,000+ native apps plus custom app connections for any public API. A marketing team that lives in niche SaaS tools often gets more from catalog depth than from a lower per-run cost. A team with engineers on hand will find that n8n's HTTP request and code steps cover most gaps. Smaller teams just starting out can use our AI automation guide for small businesses to pick their first workflows before choosing a platform.
A fit test for choosing between n8n, Zapier and Make
Pick the platform your team can maintain at your real volume. Choose Zapier when non-technical marketers will build and own most workflows and app coverage matters most. Choose Make when operations staff want visual branching and are comfortable sizing credit tiers. Choose n8n when you run long, high-volume or agentic workflows, have engineering support, or need data to stay on your own servers.
Before you commit, list your three busiest workflows, convert each into tasks, credits and executions the way the table above does, add the AI steps, and price twelve months including overage. Then name the person who will own the workflows after launch, because an automation nobody owns breaks quietly. If you want help mapping workflows and costs before you buy, our AI solutions practice runs that exercise with revenue teams.
