Decision tool Prices pending verification — A value has been proposed from the official source but no editor has verified it yet.
What would this workflow cost on Make, n8n and Zapier?
They bill different things: Make counts credits (every module run), n8n counts executions (one per whole run) and Zapier counts tasks (every successful action). The same workflow is a different number on each, so enter your workload once and read each count and price below. It shows the arithmetic and does not name a winner.
200 runs a month, 3 action steps, 1 item per run, polling every 1 min, 1 active workflow, annual billing.
Make
Counts credits
43,800 credits / month
Core · 80,000 credits
$55/month · billed annually
annual billing: a 12-month commitment, shown as a monthly equivalent. On monthly billing: $64.71/month.
smallest published size with at least 43,800 credits (80,000 included)
Trigger (polling)
43,200 checks (every 1 min × 30 days) — each costs 1 credit even when nothing is found = 43,200
Actions
200 runs × 1 item × 3 actions × 1 credit = 600
Total
43,800 credits per month
Rule: A module run is one operation and costs one credit; a module that handles several items runs once per item. The trigger runs once per check — a polling trigger spends a credit on every check, even when it finds nothing.
Make's lowest regular scheduling interval is one minute (Make Academy). We could not confirm on Make's live pages on 5 November 2026 which interval each plan allows or whether a plan limits active scenarios: check these in Make. The page offers Core up to 300,000 credits a month.
Rule: One task per successful action step. Triggers (polling included), filters, paths, formatter, delay and looping steps are free. An action that runs once per loop iteration is one task per iteration.
1-minute polling, 25 seats, shared Zaps, folders and app connections.
The links above open each vendor’s official website. At present they carry no partner code, and which products appear here does not depend on any commercial relationship. Our affiliate disclosure says what would change if that ever does.
How to read this
The three totals are in different units, so they are never compared with each other — only the price of the published plan each needs.
Make and Zapier are both priced in US dollars: at this workload Make needs $55 a month and Zapier needs $69 a month. That is the price at these assumptions, not a verdict — change one input below and it can reverse.
Where a product has no published plan for your volume, the page says so instead of guessing a price. Overage rates that a vendor does not publish are not modelled.
What would change the answer
The same workload with one thing changed. Looping many items per run, a fast polling trigger and the number of runs move the three products very differently.
Plan and price on each product when one input changes
If there were…
Make
n8n
Zapier
Your workload
$55/monthCore
€20/monthCloud Starter
$69/monthTeam
twice the runs
$55/monthCore · 44,400 credits
€20/monthCloud Starter · 400 executions
$69/monthTeam · 1,200 tasks
10 items per run in a loop
$55/monthCore · 49,200 credits
€20/monthCloud Starter · 200 executions
$169/monthTeam · 6,000 tasks
an instant (webhook) trigger
Free plan, conditionalFree · 800 credits
€20/monthCloud Starter · 200 executions
$19.99/monthProfessional · 600 tasks
monthly instead of annual billing
$64.71/monthCore · 43,800 credits
€24/monthCloud Starter · 200 executions
$103.50/monthTeam · 600 tasks
How the price moves with volume
The same workflow shape at different numbers of runs per month, on annual billing. Each product steps up at its own sizes; where a plan is missing the page says so. No product is named cheaper.
Plan and price on each product at different monthly run counts
What this calculator does not cover, and where the numbers come from
Make
Not modelled
Extra credits can be bought when a plan's credits run out (see Make's help on extra credits); this calculator shows the next plan size instead.
Modules that cost more than one credit (AI modules billed by tokens, the Code app at 2 credits per second) are not modelled; AI features follow separate rules on Make's credits page.
The Teams plan has the same credits as Pro at a higher price (it adds team roles and shared templates) and is not offered by the selector. Enterprise is custom-priced.
Make sells fixed credit sizes: its pricing slider stops only at the sizes in this table (10,000, 20,000, 40,000, 80,000, 150,000, 300,000 and, on Pro, up to 8,000,000; checked on 2026-10-19 after the page's count-up animation had settled). A count just above a size therefore moves to the next size.
Credits expire at the end of the term (month or year); extra credits expire after a month, or at the end of the year on annual Pro and Teams. A router or error-handler module costs no credit (Make pricing FAQ).
Make renamed “operations” to “credits”; older articles use the old word.
Make's page says annual billing saves “15% or more”; both billing prices are shown for every size.
VAT/tax treatment is not stated on the pricing page.
n8n Cloud is priced up to 50,000 executions a month (Pro); above that the page prices only the self-hosted Business licence (up to 500,000) and Enterprise (contact sales). No overage price is published for Cloud Starter or Pro.
Self-hosting n8n Community Edition has no licence fee under the Sustainable Use License (internal business or non-commercial use only); the server, database, backups and your time are your cost — enter them in the self-hosting assumptions below, they are not part of the plan prices.
n8n's documentation (docs.n8n.io, Understand executions, read 2026-10-19) says only production executions count toward the quota: a sub-workflow run is not counted separately, and manual test runs, error-workflow runs and polls that find no data do not count. Not modelled here (the calculator counts one execution per run that processes work).
The Cloud free trial (no credit card) is Pro-level with 1,000 executions; Business has a 14-day trial that needs a card (n8n pricing FAQ). Trials are not priced here.
Prices are shown in euro and are not converted (n8n also shows other currencies in some countries; only the euro prices were read).
VAT/tax treatment is not stated on the pricing page.
A schedule-triggered workflow runs (and counts) on every interval; an app trigger polled by n8n runs only when there is new data — the model counts only executions that process work.
When a plan's tasks run out, Zapier either stops new runs or, with pay-per-task billing on, bills extra tasks up to 3× the plan's task limit in total and then stops the rest (Zapier help). Zapier's help says the rate "depends on your plan and billing cycle type" and is shown in the billing settings; no per-task price was found on the pages read, so overage is not modelled — the next size is shown instead.
Zapier's task usage rates page (read 2026-10-19) lists what costs more than one task: Lead Router 5 tasks per lead, MCP 2 tasks per tool call, AI by Zapier 1, 3 or 5 tasks per AI step on its standard, advanced and premium model tiers (1 task with your own model), and Code by Zapier extended runtime 1 task per 30 seconds beyond the plan allowance. These are not modelled; the page also says rates may change.
Enterprise is contact-sales; the slider's “Custom task limit” position has no price.
VAT/tax treatment is not stated on the pricing page.
Prices and unit rules were read from the vendors’ official pages on and read again, by a different method, on . A third reading on confirmed the entry prices and corrected the n8n size tables. They have not yet been verified by an editor. Plans change, so confirm on the vendor’s page before buying. See how we verify.