n8n, Make or Zapier: picking the tool and the person who builds it
Short answer
Zapier suits a team that wants simple automations running with no technical help. Make suits a team that wants visual workflows with branches at a low price per step. n8n suits a team with some technical confidence that wants to self-host, keep data in-house or run long workflows cheaply. Whichever you pick, the build matters more than the tool: error handling, a log and a named owner.
By Jakub Cambor, founder of AI for Marketing, updated
n8n, Make or Zapier: which should my business use?
Use Zapier if nobody on the team is technical and the workflows are short, Make if someone is happy to learn a visual builder and the workflows branch, and n8n if you have a developer or must keep the data on your own servers. All three connect the same everyday tools, such as HubSpot, Google Sheets, Slack, Gmail and the ad platforms, so the choice comes down to how they charge, where they run and who will maintain them.
The way each one counts usage is the difference most buyers miss. Zapier charges by the task, which is each action a workflow completes. Make charges by the credit, and most actions use one. n8n charges by the execution, which is one full run of a workflow however many steps it has. A short workflow costs much the same anywhere; a long one that runs thousands of times a month can cost several times more on a per-step plan than on a per-run plan.[1][2][3]
| Compared on | Zapier | Make | n8n |
|---|---|---|---|
| How usage is counted | Per task: each action a Zap completes | Per credit: most actions use one | Per execution: one full run, any number of steps |
| Free way in | Free plan with 100 tasks a month and two-step Zaps | Free plan with 1,000 credits a month | Community edition, free to run on your own server |
| Paid plans listed from | $19.99 a month | $9 a month for 10,000 credits | $20 a month, billed annually, for 2,500 executions on n8n's cloud |
| Where it runs | Zapier's cloud | Make's cloud, on AWS in the EU or North America | Your own server, or n8n's cloud with data held in Frankfurt |
| AI steps | AI steps, code and MCP, all drawing on one task allowance | Make AI Agents and AI Toolkit on every plan, with Make's AI provider or your own key | AI agent workflows with models from OpenAI, Anthropic, Google and others |
| Limits worth knowing | Free plan checks for new data every 15 minutes | Free plan runs at most every 15 minutes | Cloud Starter runs 5 workflows at the same time |
| Learning curve | Gentlest: fill in steps, over 9,000 app connections | Moderate: a visual canvas with routers and filters, over 3,000 apps | Steepest: closer to code, with JavaScript and Python steps |
Sources: Zapier, Plans and pricing (opened September 2026); Make, Pricing (opened September 2026); n8n, Plans and pricing (US dollar prices, opened September 2026); n8n Docs, Self-host n8n; n8n Docs, Build AI workflows
Prices are as each vendor lists them in US dollars at the time of writing, September 2026; n8n shows euro prices to European visitors, and all three change their plans often, so check the pricing page before you commit.[3][2][1]
- A small team with no developer, automating a handful of hand-offs such as new form entry to CRM to Slack alert: Zapier. It is the quickest to learn and the most likely to have a ready connector for an obscure app.
- A marketing or operations team with workflows that loop over lists, branch on conditions or call an AI model: Make. The canvas shows the whole flow, and the per-credit price stays low while volumes are modest.
- A team with a developer, strict rules on where customer data lives, or workflows with many steps running all day: n8n, self-hosted or on its cloud.
- A team already paying Zapier for tens of thousands of tasks: price the same workflows on n8n before renewing, because a long workflow counts once per run there.
Search demand is heaviest for n8n by name, in the UK and even more in the US; far fewer people search for Zapier or Make automation.
| Search | UK | US |
|---|---|---|
| n8n | 49,500 | 165,000 |
| n8n automation | 480 | 1,600 |
| zapier automation | 140 | 1,000 |
| n8n agency | 170 | 720 |
| make.com automation | 30 | 260 |
| All of these | 50,320 | 168,580 |
Monthly Google searches for each phrase, in each country.
Source: Google Ads monthly search volume, via DataForSEO (September 2026)
Who can build n8n or Make automations for my company?
Three kinds of people build them: someone on your own team who learns the tool, a freelancer or specialist found through the vendors' own expert directories, or a marketing partner who builds the automation as part of running your marketing. We are the third kind. We build custom automations in Make, n8n and Zapier as an add-on to our managed marketing service, never as a separate offer, so each one is tied to a result in your marketing and has someone keeping it running.
The most useful test of a builder is a workflow they have already put into production. Ours ran on Make.com for Allegiance Industries, a facilities services company whose sales team used to research prospects by hand. Here is that workflow in plain steps.
- 1
Add a company
The sales team types a target company into a Google Sheet, one for each industry they sell to.
- 2
Start the run
Each morning a scheduled Make.com scenario reads any new companies and passes them on one at a time.
- 3
Decide who to find
A Relevance AI agent picks the job titles worth contacting for that industry and builds the search.
- 4
Find the people
Apollo returns decision makers with verified email, direct phone, LinkedIn profile and company details.
- 5
Write the record
Zoho CRM receives a complete lead after a duplicate check, so the sales team starts the day with it waiting.
- 6
Flag the gaps
A company with no valid contact goes to a person for review instead of being dropped quietly.
Source: Automated lead prospecting: AI for Marketing case study
Up to 41
Automated lead prospecting
Up to 41 companies researched per daily run, 17 data points per lead, and an estimated $9,000 to $10,000 a year saved (the client's own estimate).
Allegiance Industries, a $100M+ facilities company, built by AfM in 2026 with Make.com, Relevance AI, Apollo and Zoho CRM.
Read the case studyWhoever builds yours, ask for the parts that keep a workflow alive after launch: retries and an alert when a step fails, a log of every run, credentials held in your own accounts, a short written map of what the workflow does, a human check anywhere AI writes something a customer will read, and one named person responsible for it. A workflow without those works on the day it is shown to you and fails quietly a few months later. The full Allegiance Industries case study shows the error handling, and our comparison of n8n and Relevance AI covers when an AI agent beats a fixed workflow.
What does a good automation cost to build and run?
Running a good automation usually costs little and predictably: the tool's plan, any AI model usage and the apps it connects. Building it is normally the bigger cost, and it depends on how many apps are involved, how many exceptions the workflow must handle and whether AI makes decisions inside it. The cost people forget is upkeep, because apps change and a workflow nobody owns breaks without telling anyone.
To see how the running cost scales, take a lead-routing workflow: a trigger when a new lead arrives, then seven actions to enrich it, score it, write it to the CRM and alert the right salesperson. At 500 new leads a month, each vendor's own counting rule gives very different totals.[1][2][3]
- Zapier tasks3,500
- Make credits4,000
- n8n executions500
Worked out with each vendor's published rule. Zapier counts each completed action; Make charges one credit for most actions, the trigger included; n8n counts each full run. AI steps can use more than one task or credit.
Sources: Zapier, Plans and pricing (opened September 2026); Make, Pricing (opened September 2026); n8n, Plans and pricing (US dollar prices, opened September 2026)
At that volume the workflow fits inside the 10,000 credits of Make's Core plan and the 2,500 executions of n8n's cloud Starter plan, while on Zapier it needs a plan with at least 3,500 tasks a month. A weekly reporting digest, by contrast, runs four or five times a month and uses only a few dozen tasks or credits on any of the three.[2][3][1]
- Apps without a ready connector need custom API calls, which take longer to build and test.
- Branches and exceptions, such as what happens to a lead with no email, add most of the build time and most of the value.
- AI steps need prompts, test cases and a person checking the output before anything reaches a customer.
- A self-hosted n8n needs a server, updates and backups, which is a cost in time even when the software is free.
- Data held under UK GDPR or in regulated industries may decide where the workflow can run.
We scope custom automations on a call, as an add-on to the managed service. The free breakdown below is the place to start: it shows where your marketing loses money first, and whether an automation is the right fix at all.
Proof from our own work
1,600+
Inbound lead automation
1,600+ leads processed in the peak quarter; follow-up went from 1 to 3+ days to the same day.
Global technology platform (anonymised), in the same engagement as the paid media result: lead capture, enrichment and CRM entry automated. Delivered by Jakub as a freelance consultant before or while founding AfM.
Read the case study4
Brand-trained SEO content agent
An agent with 4 capabilities built for a premium sake brand, delivered to the client.
HEAVENSAKE, a premium sake brand. Delivered to the client, not yet deployed on their site.
Read the case study
What we do
We run your marketing for you and make it pay for itself.
- First comes a free written breakdown: we look at your site and any numbers you choose to share, with no logins, and show you where the money leaks and the first thing we would fix.
- Then we fix and run the connected commercial journey: the work that earns attention, the pages that turn it into a lead or sale, and the follow-up that creates the next valuable conversion.
- Everything is actively managed and improved, with a weekly report showing what changed and what it made, and you approve everything before it goes out.
- Rolling monthly. No lock-in, cancel any time.
What it is not
- Not a single-channel agency: we do whatever makes you money next, not ads forever or SEO forever.
- Not a junior team on a six-month lock-in: you work with Jakub directly, rolling monthly.
- Not AI left to run on its own: every piece is reviewed by a person and approved by you.
We also build custom automations, such as reporting digests, lead enrichment and routing, and support triage in tools like n8n, Make and Zapier. They are an add-on to the managed service, not a separate offer.
More questions
Is n8n really free?
The self-hosted Community edition costs nothing to license: n8n's own docs say that without a licence key it runs as the free Community edition. You still pay for the server it runs on and for someone to keep it updated, backed up and secure, and the paid Business and Enterprise editions add features such as single sign-on and version control.[4][3]
Can Zapier, Make or n8n use ChatGPT or Claude inside a workflow?
Yes. All three can call a language model as a step, for example to classify a support ticket, summarise a week of results or draft a reply. Make and n8n let you use your own API key, and Zapier counts AI steps against the same task allowance as everything else, so check how much each AI step uses before you run one on every record.[2][5][1]
Should we move from Zapier to n8n or Make?
Only when there is a reason: the task bill has grown, you need workflows Zapier handles awkwardly, or customer data must stay on your own servers. Moving means rebuilding and retesting every workflow, so start with the most expensive or most fragile one and keep the rest where they are until it has run cleanly for a month.
Related answers
- AI marketing automationAI marketing automation is marketing automation with an AI model inside the workflow, so a step can read, judge or write instead of only following a fixed rule.
- AI automation agencyAn AI automation agency designs, builds and looks after workflows that connect your business tools and use AI models for the steps that need reading or judgement, such as summarising, classifying or enriching.
- What AI automation isAI automation is the use of AI models inside automated workflows, so software can handle tasks that need reading, judgement or writing, such as sorting emails, pulling details out of documents or summarising results.
- AI marketing reportingAI can write your weekly marketing report once the numbers reach it cleanly.
- AI marketing toolsThe best AI marketing tools are the ones that do a job you already have, inside the systems you already run.
- AI lead generationB2B companies use AI for lead generation in four places: researching which companies and people fit, enriching each lead with verified contact and company data, drafting outreach that refers to something real about the prospect, and routing new enquiries to the right person at once.
Sources
- 1.Zapier, Plans and pricing (opened September 2026)zapier.com
- 2.Make, Pricing (opened September 2026)make.com
- 3.n8n, Plans and pricing (US dollar prices, opened September 2026)n8n.io
- 4.n8n Docs, Self-host n8ndocs.n8n.io
- 5.n8n Docs, Build AI workflowsdocs.n8n.io
- 6.Google Ads monthly search volume, via DataForSEO (September 2026)dataforseo.com
- 7.Automated lead prospecting: AI for Marketing case study