n8n vs Make for HR teams (September 2026): an independent comparison

n8n or Make: the verdict

n8n vs Make: comparison table

Compare all fields side by side, including certifications and hosting →

n8n vs Make: what it costs per year

Pricing: n8n bills each workflow run, Make bills each action and trigger check

n8n counts executions: one execution is a run of the whole workflow, whatever the number of steps. Cloud prices are in euros, billed annually, as of September 2026: Starter €20 a month for 2,500 executions, Pro €50 for 10,000 and the self-hosted Business plan €667 for 40,000. Enterprise is on request, and the self-hosted Community Edition is free.

Make bills in credits. Each module action uses 1 credit by default, and so does each check a trigger makes for new data; Make's AI Provider costs more, based on tokens. The pricing page lists a Free plan with up to 1,000 credits a month, then Core from $9, Pro from $16 and Teams from $29 a month for 10,000 credits. Paying annually saves 15 percent or more, but the page does not say which billing cycle these prices assume. Annual Pro and Teams plans release the year's credits at once, and extra credits cost 25 percent more.

The two models meet at 3 actions. A flow started once per new hire with 1 trigger and 3 actions uses 4 credits in Make and 1 execution in n8n, so Core's 10,000 credits and Starter's 2,500 executions both cover 2,500 runs. At 8 actions, Make Core covers about 1,100.

Polling costs on both. A flow that checks Personio for new employees every 15 minutes runs 2,880 times in a 30-day month: 2,880 credits in Make even when nobody joined, and 2,880 executions for an n8n Schedule Trigger, more than Starter's 2,500.

HR workflows: Make has more ready HR apps, n8n reaches the rest through APIs

Make lists more than 3,000 apps. Its Personio app is built and verified by Make, with triggers for new employees and new positions and actions such as Create an employee, Update an employee and Upload a document, so an HR generalist can build a Personio onboarding flow without code. For Personio itself, see Personio vs HiBob.

n8n lists 2,274 integrations, but its documentation has no Personio node, so a Personio flow runs through the HTTP Request node and Personio's API. Recruiting shows the reverse: n8n has a built-in Workable Trigger node for new and moved candidates, while Make's Workable app is maintained by community users rather than by Make.

Teams running both get help from Make: since July 2026, Make Grid, which maps automations and their dependencies, can also map n8n workflows. Both platforms are among the 9 tools in our workflow automation category.

AI agents: both build them, Make still labels its agents beta

n8n builds agents from nodes: an AI Agent node with memory and tools, human-in-the-loop approval steps, guardrails and evaluations. It works with any language model, including local models through its Ollama node, so prompts with employee data can stay on your servers when you self-host. MCP nodes connect n8n workflows and agents to assistants such as Claude and to outside tools.

Make's current agent app, Make AI Agent (New), is in open beta, and Make says functionality and pricing may change. Agents use modules, scenarios and MCP tools and show each decision in a reasoning panel; Make says you can add manual approvals. Make's AI Provider works on all plans and bills tokens as credits, while your own OpenAI, Claude or Gemini connection needs a paid plan. Its MCP server lets Claude and ChatGPT run scenarios.

Both AI builders are pre-release. n8n Assistant, in preview, builds and troubleshoots workflows from a chat and may process selected execution and error details, which can contain employee data. Maia by Make, in public beta on paid plans, builds and fixes scenarios, and Make says Maia never uses your data for training. Test both with made-up employee records.

Employee data and country fit: n8n can stay in-house, Make lets you pick the EU or the US

n8n Cloud stores data in the EU, on servers in Frankfurt run on Microsoft Azure, and an independent auditor reviews its SOC 2 program every year. Self-hosting keeps employee data on your own infrastructure, under a license for internal business use, but someone must patch it: n8n published ten security advisories on GitHub on September 16, 2026, eight rated High. Workflow settings can stop n8n from saving successful executions, and Enterprise plans can redact execution data.

Make runs only in its own cloud on AWS. Customers choose an EU or a US data center when they create an organization and cannot change it later. Make has an ISO 27001-certified security program, completed SOC 2 Type II and SOC 3 audits, and a data processing agreement with Standard Contractual Clauses; its parent Celonis is registered under the EU-US Data Privacy Framework. A Keep data confidential setting stops Make from keeping scenario data in its execution logs.

Our directory lists Germany as n8n's strong market and none for Make, whose Czech home is outside our 23 countries. The rules follow the use, not the tool: a flow that filters applications or evaluates candidates counts as high risk under the EU AI Act, with duties from December 2, 2027, the UK's Data (Use and Access) Act 2025 requires safeguards such as human intervention for significant solely automated decisions, and New York City's Local Law 144 requires a bias audit before an automated tool substantially assists hiring or promotion decisions.

Who should pick n8n and who Make

The bottom line

Choose by who builds the flows and where the data may go. If HR staff will build short flows themselves and a cloud service in the EU or the US is acceptable, start with Make and pick the data center with care. If flows run many steps per person or employee data must stay in-house, start with n8n and decide first who installs its updates.