What should you build with — and what will it actually cost?

We project a starting volume from your company size, business type, role and the workflow types you select — each workflow type carries a typical monthly runs baseline, and picking several scales the estimate sublinearly rather than just adding them up. The auto-estimate is only a starting point: you can override the runs, workflows and budget sliders in the Volume & budget step at any time.

From each tool's official public pricing page, verified by hand. Every change we record is dated and published in our price changelog, and the footer shows the month prices were last verified.

Enterprise and custom tiers have no public price. When your volume goes beyond the biggest public plan, we estimate the custom tier by scaling the largest public plan's price sublinearly, calibrated against published price points. Estimated prices are marked with ~ and a disclaimer: treat them as a ballpark and confirm the real quote with the vendor's sales team.

n8n, Make, Zapier, Pipedream, Activepieces, Automatisch, Node-RED, plus Pabbly Connect, Zoho Flow, Relay.app, Latenode and IFTTT — cloud plans and self-hosted options alike. You can also put them side by side on the compare page.

Each tool gets a 0–100 rating on seven features — no-code UI, code access, AI nodes, self-hosting, GDPR, integrations and ecosystem maturity. We weight those to your selected role, then blend feature fit (52%), price against your budget (26%) and ecosystem maturity (12%), plus a small workflow-fit bonus. The per-feature ratings are WizardCost's own editorial assessment, based on public documentation and sources we record internally — they are considered judgements, not measured benchmarks or vendor-supplied numbers. Treat the scores as guidance for comparison, not exact metrics.

No — we're editorially independent, and we currently run no affiliate links at all. Every tool is ranked by the same formula, and the per-feature scores are our own editorial judgement applied equally to all of them. See the affiliate disclosure.

Because saved hours aren't money until you realize them, and each way of realizing them has different economics. If it's your own time, hours × rate is roughly right. If you redeploy employees, freed time fragments — 10 people saving 15 minutes a day rarely becomes 2.5 people's worth of output, so we apply a realization factor. If you cut headcount, savings only exist in whole positions, and most countries make exits expensive: in Czechia, for example, a statutory 2-month notice period (still on payroll) plus severance of 1–3 monthly salaries — so the wizard shows the one-time exit cost and the month your savings actually break even. The cleanest case is avoiding the next hire: no severance, no notice, the position simply never happens. Pick your realization mode in the "Is it worth it?" panel — the employment-law presets are editorial orientation for 2026, not legal advice.

Because n8n can be self-hosted — the Community Edition software is free, so the only cost is the infrastructure it runs on. That is server cost, not a tool fee, and it scales with volume: roughly $8/mo for a small VPS at low volume, rising to ~$45/mo at 100k runs (you'll need Postgres and Redis by then) and ~$200/mo at very high volume. Two caveats: this assumes lightweight workflows — heavy data processing needs a bigger box — and it does NOT include your time to set up, patch, monitor and back up the server. If self-hosting isn't realistic for you, switch Hosting preference to “Cloud only” in the Volume & budget step and the ranking recalculates with cloud plans only.