Zapier vs Make: How the Pricing Models Differ
Zapier bills successful action steps and gives the trigger away for free. Make bills every module run, trigger included, and counts each bundle separately, so the same workflow records more units on Make while each unit costs a small fraction as much.
The pricing pages look comparable. The billing meters are not. Zapier charges you for successful action steps and gives away the trigger. Make charges for module runs, and the trigger is a module run. Build the same workflow on both and Make will usually record more billable units than Zapier does for the identical job. Make's units are also roughly twenty times cheaper each, which is why the arithmetic almost always lands in Make's favor and why comparing "price per unit" between the two is meaningless.
What follows is drawn from each vendor's own pricing and help pages, checked September 2026. Prices move. The dates are there so you know how stale this is when you read it.
What Zapier counts as a task
Zapier publishes a specific list of what does and does not burn a task. From Zapier, How is task usage measured in Zapier?, as of September 2026:
Counts as a task: every successful action step. Successful steps inside error-handler paths. Steps that rerun during a full Zap replay. Action steps inside sub-Zaps, plus the Call and Return actions. Search actions configured to "proceed if nothing found." Some products bill at their own rate: Code by Zapier is one task per run, Lead Router is five tasks per lead routed, Zapier MCP is two tasks per successful tool call, and AI by Zapier varies by model tier.
Does not count: all trigger steps. Filter and Paths steps. Failed or halted action steps. Steps skipped by a filter. Search actions set not to proceed if nothing is found. Steps from Formatter, Delay, Looping, Digest, Zapier Manager, Storage, Tables, and Forms. Polling — Zapier checking an app every few minutes to see if anything changed is free.
Zapier's polling is free. A Zap that checks a mailbox every two minutes and finds nothing for a week costs zero tasks. You pay only when something actually moves.
What Make counts as an operation
Make's definition is shorter and much broader. Make's help article on operations defines one as "a single module run to process data or check for new data." Checking counts. The trigger counts.
Two consequences follow from that sentence, and they are where budgets go wrong.
First, the trigger is billable. A three-module scenario — trigger, then two actions — is three operations, where the equivalent Zapier Zap is two tasks.
Second, bundles multiply. Make's documentation states that modules process each bundle separately, so each bundle triggers its own module run. Adding five contacts to a sheet is five runs of that module, so five operations. The trigger module is the exception: it runs once to fetch data regardless of how many bundles it returns, so it counts as one.
One naming change to know about. On January 19, 2026, Make renamed its billing unit from operations to credits. Existing balances converted at 1:1 — 1,000 operations became 1,000 credits — and Make's announcement states that plan pricing and the cost of buying more were unchanged. If you read an older guide that talks about operations and a current invoice that says credits, they are the same thing at the same rate. The one caveat Make flags is that consumption for its built-in AI features may rise to match actual costs.
The same workflow, counted twice
Take a plain one: a new row lands in a spreadsheet, you create a customer record in a CRM, then you send a notification email. Nothing exotic.
| Step | Zapier tasks | Make credits |
|---|---|---|
| New spreadsheet row (trigger) | 0 | 1 |
| Create CRM record | 1 | 1 |
| Send email | 1 | 1 |
| Per run | 2 | 3 |
That is arithmetic from the published rules, not a benchmark. At 1,000 runs a month you are looking at 2,000 Zapier tasks and 3,000 Make credits. Now price them.
The prices, dated
From Zapier's pricing page, as of September 2026, on monthly billing: Free is $0 for 100 tasks a month, limited to two-step Zaps with a 15-minute polling interval. Professional starts at $19.99 a month for 750 tasks and unlocks multi-step Zaps, unlimited premium apps, and a 2-minute polling interval. Team starts at $69 a month for 2,000 tasks, includes 25 users, shared app connections, and a 1-minute interval. Zapier advertises a 33% saving on annual billing. Higher task tiers cost more within each plan.
From Make's pricing page, as of September 2026, on monthly billing at the 10,000-credit tier: Free is $0 for 1,000 credits a month, capped at two active scenarios with a 15-minute minimum interval. Core is $12. Pro is $21 and adds high-priority scenario execution and full-text execution log search. Teams is $38 and adds team roles and shareable scenario templates. Core and above all get a 1-minute minimum interval and unlimited active scenarios. Make advertises 15% or more off on annual billing.
Divide it out at list price, September 2026: Zapier Professional is about 2.7 cents per task at its 750-task entry tier. Make Core is about 0.12 cents per credit at 10,000. Make can spend three units where Zapier spends two and still cost a fraction as much.
Where each one actually wins
Make wins on volume. If your automations fire hundreds or thousands of times a month, the unit price gap is not close and the extra operation for the trigger does not rescue it. This is the whole reason people migrate.
Make wins on run frequency for the money. A 1-minute minimum interval starts at Make's $12 Core plan. On Zapier, the 1-minute interval sits on Team at $69. If you need something checked every minute and you do not need 25 seats, that is a real gap.
Zapier wins on predictability. Task counting is easier to forecast because it maps to one thing you can see: successful actions. Make's bundle behavior means a search step that returns more rows than you expected quietly multiplies your bill. Zapier has the same multiplication for actions run repeatedly, but the trigger and the routing steps stay free, so the count tracks intuition more closely.
Zapier wins on free-tier shape, sometimes. Make's free tier gives ten times the units but caps you at two active scenarios. Zapier's free tier gives 100 tasks but does not cap the number of Zaps — it caps them to two steps each. If you want six tiny automations running for nothing, Zapier's shape fits. If you want two busy ones, Make's does.
Where it is a tie
Below roughly 100 workflow runs a month, this comparison does not matter. Both free tiers cover it, both cap you at a 15-minute interval, and the difference between $12 and $19.99 is not worth a migration. Pick on which builder you can stand looking at, and whether the specific apps you need are supported. Connector coverage decides more small-business automation projects than pricing does, and neither vendor's app directory is a subset of the other's.
It is also a tie on the thing people expect to be a differentiator: neither vendor charges you for a workflow that sits idle. Zapier does not bill polling. Make bills the trigger module run, but a scheduled scenario you have switched off is not running at all.
What to check before you commit
Count your real steps first. Open the workflow you actually plan to build, list the modules, and apply the two rules above. Two numbers, not one.
Then check the tier above your estimate on both pricing pages. Both vendors sell units in steps, not continuously, and the plan that looks cheapest at your estimate is often not the cheapest at 1.5x your estimate. Automation volume goes up, not down.
Last, check whether any step you need falls into a special rate. Zapier's per-task multipliers for Lead Router, MCP calls, and AI models are published on the help page linked above and can turn a two-task Zap into a ten-task one. Make flags AI feature consumption as the area most likely to change. Read those two paragraphs before you build the AI-heavy version of anything.