Quick answer: Make and n8n are the two best general-purpose automation platforms for small businesses in 2026. Make is cloud-only, visual, and built for non-technical teams — pricing starts at around £9/month for 10,000 operations, with 1,000+ app integrations. n8n is open-source and self-hostable (free with unlimited executions) or available as managed cloud from €24/month, with full JavaScript in every node and around 400+ integrations. Pick Make if you want to launch automations fast without managing infrastructure; pick n8n if you have a developer, care about data privacy, or run high-volume workflows where per-operation pricing would hurt. Most small businesses should start on Make and move to n8n when volume justifies the self-hosting overhead.
What is Make?
Make (formerly Integromat) is a cloud-hosted visual automation platform founded in 2012. You connect modules on a drag-and-drop canvas, configure them with clicks and dropdowns, and Make executes the workflow on its servers. It is designed to be accessible to non-technical users while still supporting complex multi-step scenarios.
In August 2025, Make moved from "operations" to a "credits" model, where standard module executions cost one credit and AI features consume more. As of 2026, Make offers more than 1,000 native app integrations — one of the largest catalogues of any automation platform.
What is n8n?
n8n is an open-source workflow automation tool you can self-host on your own server for free, or run via n8n Cloud. Unlike Make, it is built with developers in mind: you can write custom JavaScript inside any node, the data model is transparent, and the source code is public on GitHub.
n8n charges per execution — one complete run of a workflow, regardless of how many nodes it contains or how much data it moves. The self-hosted Community Edition has no execution limits. Cloud plans start at €24/month for 2,500 executions and scale to €800/month for business tiers with single sign-on.
Make vs n8n: head-to-head comparison
| Feature | Make | n8n |
|---|---|---|
| Hosting | Cloud only | Self-hosted or cloud |
| Pricing model | Per operation/credit | Per execution (or unlimited self-hosted) |
| Entry price | ~£9/month (Core, 10,000 ops) | Free (self-hosted); €24/month (Cloud Starter) |
| Ease of use | Beginner-friendly | Requires some technical knowledge |
| Custom code | Limited (HTTP module, functions) | Full JavaScript in any node |
| Native integrations | 1,000+ | 400+ (growing) |
| Data control | Data passes through Make servers | Full control if self-hosted |
| Scalability cost | Costs grow with volume | Near-zero marginal cost when self-hosted |
| AI features | Make AI (built-in) | AI nodes, LangChain integration |
| Users | Per workspace | Unlimited on every plan |
How pricing actually works (and where people get caught)
Make: the credit trap
Make moved from "operations" to "credits" in August 2025. The change matters because a single workflow run can consume many credits — every module in the scenario counts as one credit, and that includes triggers, filters, and iterators. A 5-module scenario processing 10 records burns 50 credits, not 10.
A typical small business running lead routing, email sequences, and weekly reports will land on the Core plan (~£9/month for 10,000 operations) and stay there. The pain starts when you add AI features — Make's AI modules consume credits at a higher rate than standard modules, so a workflow that summarises support tickets with GPT can drain your monthly allocation in days rather than weeks.
n8n: execution simplicity (with a self-hosting catch)
n8n's pricing unit is the execution: one complete run of a workflow, regardless of how many nodes it has. A 200-node workflow processing 10,000 rows of data counts as one execution. This makes costs predictable in a way Make's cannot match.
The catch is self-hosting. The free Community Edition has unlimited executions, but you pay for the infrastructure to run it — a small VPS costs around £10-50/month depending on volume. You also own updates, backups, and the occasional late-night debugging session when a node fails. For a small business without in-house technical staff, that overhead is real.
When to choose Make
Make is the right choice when:
- Your team is non-technical — marketers, ops managers, or founders building automations themselves without coding
- You need fast time-to-value — Make's visual builder and pre-built templates let you launch in hours, not days
- Your volume is moderate — if your automations run tens of thousands of operations per month (not millions), Make's pricing is reasonable
- You rely on niche SaaS apps — Make's 1,000+ integrations cover tools that n8n does not yet support natively
If you can describe what you want in a sentence, you can usually build it in Make without touching documentation. For teams new to automation, our AI Kickstart workshop walks you through building a real workflow in 90 minutes.
When to choose n8n
n8n is the better choice when:
- You have a developer — even a part-time one. Custom JavaScript transforms n8n from good to exceptional
- Volume is high — if you are processing hundreds of thousands of records, per-operation pricing kills the business case for Make
- Data privacy matters — self-hosting means sensitive business data never touches a third-party server
- You want to avoid vendor lock-in — n8n's open-source nature means you own your automations and can migrate hosting freely
For UK businesses handling client data under UK GDPR, the self-hosting option is not just a cost play — it is a compliance lever. Data stays in your jurisdiction, under your control.
Real example: a UK marketing agency
A 5-person marketing agency in Leeds managed 12 clients across Meta Ads, Google Ads, HubSpot, and Mailchimp. Every Monday, an account manager spent 4 hours exporting reports from each platform and pasting them into a client-ready template.
They evaluated both tools:
Make approach: Built a scenario pulling data from all four platforms, generating a formatted report in Google Docs, and emailing it to each client. The scenario used 8 modules and ran 12 times per week (once per client). Monthly credit consumption: around 3,800. Plan: Core at ~£9/month. Time saved: 16 hours/month. No infrastructure to manage.
n8n approach: The same workflow in n8n would count as 12 executions per week (around 50/month). Self-hosted on a £15/month VPS, the marginal cost was effectively zero. The trade-off: one of the agency's developers spent a weekend setting up Docker, configuring backups, and writing a small JavaScript function to format currency correctly — work that took 12 hours upfront.
Outcome: They chose Make. The 12-hour setup investment for n8n was not worth it for 50 executions a month. They would revisit n8n if their client roster doubled and Make's credit consumption crossed 30,000/month.
This is the decision most small businesses will reach. The crossover point where n8n's economics beat Make is usually around 30,000-50,000 operations per month — beyond what most SMBs run in their first year of automation.
Common mistakes to avoid
1. Picking n8n because it is "free"
The Community Edition is free, but your time is not. If no one on your team can comfortably manage a Docker deployment, the "free" option becomes an expensive one the first time a workflow breaks at 11pm before a client deadline.
2. Picking Make and ignoring the credit model
Before committing to a plan, estimate your real monthly credit consumption: number of workflows × average modules per workflow × runs per month. Add 30% for AI features and unexpected iterations. If that number exceeds 10,000, budget for the Pro plan (~£16/month) from day one.
3. Treating the choice as permanent
Both platforms export workflows as JSON, and most common patterns (webhook → transform → API call → notification) translate between them with minor rework. Start with whichever fits your team today, and reassess when your volume doubles.
4. Not having a recovery plan
Automations become invisible infrastructure. If a workflow that routes leads or sends invoices stops running, the impact is silent until someone notices. Whichever tool you pick, set up failure notifications and review execution logs weekly for the first month.
How CortexLeap helps: We design and build the automation for you — on whichever platform fits your team and volume — so you avoid the 12-hour setup trap and the credit-burn surprise. Our Custom Automation service handles the full build: tool selection, workflow design, AI integration, testing, and handover with documentation your team can maintain. If you want to run automations yourself long-term, the AI Practitioner course trains your team on multi-tool workflows over four weeks.
Frequently asked questions
Is Make or n8n cheaper for a small business?
For low-to-moderate volume (under 30,000 operations per month), Make is cheaper and simpler — the Core plan at ~£9/month covers most SMB use cases. For high volume or when you can self-host, n8n's economics win because the Community Edition has unlimited executions and you only pay server costs (~£10-50/month).
Can I use n8n without coding?
Yes, for simple workflows. n8n has a visual node-based editor and most integrations need no code. But the platform's real power — custom data transforms, error handling, complex logic — requires JavaScript. If no one on your team writes code, Make is the safer choice.
Does Make or n8n have better AI features?
Both integrate well with AI models. Make has built-in AI modules (Make AI) that let you call GPT and other models without configuration. n8n offers deeper AI capability through its LangChain integration and AI agent nodes, allowing you to build multi-step AI workflows and custom agents. For simple AI tasks (summarise, classify, translate), both are equal. For agentic AI, n8n is more capable.
Which has more integrations, Make or n8n?
Make has over 1,000 native integrations; n8n has around 400+. Both support custom HTTP requests to any API, so the gap is smaller than it looks for technical teams. If you rely on niche SaaS apps, check both catalogues before choosing.
Can I migrate from Make to n8n (or vice versa)?
Yes, but not automatically. Both export workflows as JSON, and common patterns translate with minor rework. Plan for a few hours of rebuilding per complex workflow. Most businesses treat the first platform as a learning step and rebuild deliberately when they switch.
Do I need a server to use n8n?
No — n8n Cloud runs on managed infrastructure from €24/month. You only need a server if you choose to self-host the Community Edition, which is free but requires you to manage updates, backups, and uptime yourself.
Pick the tool that fits your team, not the one with the best marketing
The automation platform decision is not really about features — both Make and n8n are production-grade and actively developed. It is about your team. A non-technical team will ship faster on Make and avoid infrastructure headaches. A technical team will pay less and build more on n8n. Either way, the cost of doing nothing — manual exports, copy-paste reports, leads sitting in an inbox — is higher than either subscription.
Get your automation built by CortexLeap — we pick the right tool for your volume, design the workflow, integrate AI where it earns its place, and hand it over with documentation your team can maintain.
Not sure where to start? Book a free discovery call and we will map your workflows and recommend the platform that fits your team and budget.
Last updated: 2026-07-23