# Zapier vs Make vs n8n vs Gumloop pricing the same workflow

Canonical URL: https://industrycontents.com/compare-make-vs-zapier/
Language: en-GB
Published: 2026-10-13
Structure updated: 2026-10-25

> Dated article snapshot. Canonical web page may receive later updates. Vendor figures retain their stated source-check dates. Illustrations are not product benchmarks.

Published: 2026-10-13 · Structure updated: 2026-10-25. Vendor prices and source checks retain the dates stated below.

## Quick answer

Zapier generally meters tasks, Make credits, n8n workflow executions and Gumloop credits plus an orchestration fee. Count the events generated by the same workflow, then add AI services, recovery and maintenance to compare total cost.

## Key terms

**Zapier task**: Zapier task is a billable successful action under Zapier’s rate rules; some actions have multipliers or exclusions.

**Workflow execution**: Workflow execution is one complete workflow run under the platform’s execution rules.

**Orchestration**: Orchestration is the coordination of steps, tools and model calls in a workflow.

## Zapier vs Make vs n8n vs Gumloop, billing models

| Comparison | Make | Zapier | n8n | Gumloop |
| --- | --- | --- | --- | --- |
| Primary meter | Credits, generally per module action | Tasks, generally per successful billable step, with some AI and code actions costing more | Workflow executions, regardless of step count | Credits based on models and tools, plus an orchestration fee |
| Free offer | 1,000 credits a month | 100 tasks a month | Self hosted Community Edition. Cloud trial with 1,000 executions and no card | 14 day free trial on Pro |
| Published entry price | Core $12 a month at 10,000 credits | Professional from $19.99 a month, with the task tier varying | Starter €20 a month, annual billing, 2,500 executions | Pro from $37 a month, 20,000 credits plus an 8% orchestration fee |
| Higher tier example | Pro $21 a month at 10,000 credits | Team from $69 a month, with the task tier varying | Pro €50 a month, annual billing, 10,000 executions | Pro credit slider up to higher volumes. Enterprise custom |
| AI costs | Some AI connections use token based credits. Your API key means a separate model bill | AI by Zapier model tiers use 1, 3 or 5 tasks per step, and tools may add tasks | Provider API charges can sit outside execution fees | Credit use depends on model and tools |
| Operational consideration | Loops and repeated modules multiply credits | Shared task pool and higher priced actions | Execution, concurrency, memory and hosting limits | Credit consumption, concurrency and model selection |

Caption: Start with how each automation tool charges. The table sets out Comparison, Make, Zapier, n8n, Gumloop.

USD and EUR figures appear as listed, without conversion. Starting prices buy different capacity on each tool. See each company's linked pricing page for billing cycle selections and tax treatment.

## Five-step workflow, credits, tasks and executions compared

The five-step scenario produces different units on Make, Zapier and n8n because each platform applies its own trigger and action rules.

Consider a form submission followed by four ordinary app actions. For this illustrative five module Make scenario, assume five credits per successful run. For a Zap with one trigger and four standard billable actions, assume four tasks. For n8n, one successful complete workflow run is one execution. Gumloop cannot be assigned a comparable credit count until you choose the specific nodes, tools and model.

| Successful runs a month | Make credits | Zapier tasks | n8n executions | Gumloop credits |
| --- | --- | --- | --- | --- |
| 100 | 500 | 400 | 100 | Measure in product |
| 1,000 | 5,000 | 4,000 | 1,000 | Measure in product |
| 10,000 | 50,000 | 40,000 | 10,000 | Measure in product |

Caption: Put the same five step workflow through the numbers. The table sets out Successful runs a month, Make credits, Zapier tasks, n8n executions, Gumloop credits.

Read these as unit estimates, since they carry no prices. Polling, loops, retries, AI steps and plan restrictions can change the outcome. At 1,000 runs the illustrative Make workload fits within 10,000 credits, while Zapier's $19.99 entry price may fall short of 4,000 tasks. n8n's Starter allowance covers 1,000 executions, subject to other plan limits. At 10,000 runs the n8n count equals the 10,000 executions on its Pro plan, and Make would need a tier of 50,000 credits.

## Choose the workflow builder your team can maintain

Make suits people who like a visual scenario builder and can estimate usage module by module. Zapier makes sense when your workflows depend on its app integrations, provided you model task use, including AI and code steps. n8n is another option for multi step workflows, where paying per full execution can work out cheaper, especially if your team can handle setup or self hosting. Gumloop is aimed at AI heavy agent workflows, so measure its credit use on your chosen model before comparing bills.

## Check how Make and Zapier count each step

Make generally meters module actions as credits. Zapier generally meters successful billable steps as tasks, and many built in tools are excluded. The gap shifts when you add branching or AI. Zapier's published task rate table lists different task multipliers for model tiers and certain actions, so assuming one task for every Zapier action will understate the bill. [Check Zapier's task rate table](https://zapier.com/pricing/rates).

## See when n8n execution billing helps

n8n's cloud plans count a full workflow execution rather than charging separately for every step. A ten step flow running 1,000 times is roughly 1,000 executions on n8n, while a Make flow with ten one credit modules would use roughly 10,000 credits. That gap in units stops short of a ten to one price saving. Platform prices, execution limits, hosting, model APIs and failures all affect the bill. Self hosted n8n also requires server operations, upgrades and backups.

## Gumloop vs Zapier and n8n, AI costs to measure

Gumloop's listed Pro starting price is $37 a month for 20,000 credits plus an 8% orchestration fee; confirm which spend attracts that fee. For Gumloop vs Zapier, run the same AI task and record both credit usage and Zapier's model-tier task multiplier. For Gumloop vs n8n, include model/API charges alongside credits or executions. A cheap credit does not establish a cheap completed workflow.

## Include setup and maintenance in the decision

Compare the same real workflow in each trial, covering app authentication, data mapping, error recovery, retry behaviour, execution logs, permissions and maintenance time. If your team needs approvals, shared credentials or audit logs, check the exact tier, because the base plan may leave them out. Before switching, inventory triggers, branches, API keys, webhooks, logs and any custom code.

## Workflow branch costs

The five-step example assumes every record follows one path. Real workflows filter, loop and recover. Map the branches before multiplying the average step count: a hundred records entering a loop are different from one record passing through a hundred nodes.

For an invented 1,000-run workload, assume two ordinary billable actions always run and eight more run for 20% of inputs. That is 2,000 + 1,600 = 3,600 actions, rather than 10,000. If 50 records repeat all ten, add 500 actions. Platform-specific triggers, excluded tools, AI multipliers and recovery rules still need to be applied; this is a design count, not a cross-platform invoice.

### Separate the model

On Make, a custom model connection produces operation credits plus a separate token bill. Built-in AI can fold token costs into credits instead. On Zapier, apply the actual model-tier and tool-call task rates. On Gumloop, ask what spending attracts the orchestration fee. On n8n, count external services alongside executions. Do not count the same model charge twice or assume it is included everywhere.

## Find the migration payback

Use monthly savings after maintenance, not the subscription gap. A hypothetical move taking 12 hours at $75 costs $900. Saving $60 a month after extra operations gives a 15-month simple payback. Saving only $15 makes it five years. Include parallel running, rebuilding credentials and checking historical data.

### Test recovery first

Use a duplicate event, an API timeout after a successful write, an expired credential and a malformed record. The important question is whether retrying creates a duplicate invoice or message. Record billed usage and time to repair, then test a peak burst and a full usage allowance. A platform that looks cheaper in the happy path can cost more to keep running.

## Measure cost per successful workflow

Cost per successful workflow includes platform fees, extra usage, external services and maintenance divided by accepted completed runs.

**Effective cost per successful run = (platform subscription + usage overages + model and API spend + estimated maintenance cost) ÷ successful completed runs.** Use your workload. A 100 run trial should record both billed units and failed or manually repaired outcomes. Compare a normal month and a peak month, then check what happens when usage caps are reached.

## Questions to settle before choosing

### Make against Zapier on cost

Make's published entry price is lower, yet that alone leaves the cheaper workflow undecided. Count Make credits and Zapier tasks for your actual steps, then compare the required plan tiers and any AI usage.

### n8n against Make for complex automations

n8n charges per full workflow execution on its cloud plans, while Make generally charges per module action. More steps can therefore increase Make usage faster, though hosting, concurrency and plan limits may change total cost.

### Zapier charges for AI steps

Task consumption varies by AI model tier and tool calls. Zapier's published rate table lists Standard, Advanced and Premium model tiers with different multipliers.

### Gumloop pricing against n8n

The two use different units. Gumloop uses credits and an orchestration fee, while n8n uses workflow executions. Benchmark the same AI workflow, including model and API charges, to estimate a meaningful cost per completed run.

### Best platform for a small business

Start with the integrations your team uses every day. Run one representative workflow, calculate the billed units and check support, permissions and error handling on the plan you would buy.

### How should we test the same workflow?

Use identical inputs and success criteria. Record billed steps, executions, retries and model charges, then add setup and maintenance time. Compare total cost per successful run.

[Compare n8n and Make hosting choices](/n8n-vs-make-european-teams/). Use the dated sources in this article for price context; the comparison method is an editorial framework, not a hands-on product benchmark.

Our [Cloudflare Code Mode explainer](https://industrycontents.com/cloudflare-code-mode-mcp-explained/) examines a different way to expose tools to an agent. It helps distinguish the orchestration design from the workflow platform’s billing unit.

Depth update, 25 October 2026: rechecked Zapier’s task-rate page and Gumloop pricing, with Make credit rules and n8n pricing reviewed in this series. Branch and migration calculations are hypothetical. Earlier entry-tier figures remain dated; no hands-on benchmark was performed.

## Sources

- [Make pricing](https://www.make.com/en/pricing) and [Make credit rules](https://help.make.com/credits), checked 14 October 2026.

- [Zapier pricing](https://zapier.com/pricing), [task rates](https://zapier.com/pricing/rates) and [AI model tiers](https://help.zapier.com/hc/en-us/articles/46425475442829-AI-by-Zapier-model-tier-pricing), checked 14 October 2026.

- [n8n pricing](https://n8n.io/pricing/), checked 14 October 2026.

- [Gumloop pricing](https://www.gumloop.com/pricing) and [credit tier guidance](https://support.gumloop.com/articles/9772571567-how-do-i-get-more-credits-without-enabling-overages), checked 14 October 2026.

Unit scenarios are illustrative. Plan prices as listed on 14 October 2026.

IC Lab · AI Agent Pricing Tracker

Take the pricing rows into your budget.

Compare source-linked rows in the tracker. [Open the tracker and find the CSV export](/ic-lab/ai-agent-pricing-tracker/). The tracker offers a CSV export with a TalkHead signup. TalkHead is our weekly newsletter on AI in business.

**Related** [All comparisons](/ic-lab/compare/) · [Workflow automation pricing](/guides/workflow-automation-pricing/) · AI Agent Pricing Tracker

For the hosting decision, read our [n8n vs Make comparison for European teams](/n8n-vs-make-european-teams/), covering cloud hosting, self-hosting and data checks.

### Related reporting and benchmarks

**[The cost of automation after deployment](https://industrycontents.com/uipath-and-warmwind-agentic-automation/)** UiPath and Warmwind take different approaches. Our analysis examines the trade-offs beyond subscriptions.

**[Faster experimentation in practice](https://industrycontents.com/ai-experimentation-workflow-fyxer/)** Fyxer's team used AI to run hundreds of growth tests. Read the process behind the number.

If self-hosting is part of the decision, browse [the GitHub repo directory](/ic-lab/github-repos/) for n8n and other workflow projects.