Make (Integromat)
It's genuinely 3-5x cheaper per operation than Zapier — but a single Zapier task can equal 3-8 Make operations, so the real savings are smaller than the headline number suggests. Still meaningfully cheaper, just not quite that dramatic.
What is Make?
Make began life in Prague, Czech Republic as Integromat, a bootstrapped visual automation startup, before Celonis, the process-mining giant, acquired it in 2020 and officially rebranded it to Make.com in February 2022. The Integromat name is fully retired now — but for anyone still searching for it, you're effectively shopping for Make.com, since old accounts and scenarios migrated automatically with zero effort. The platform's core idea is a genuine strength: rather than a linear, step-by-step list of actions like some competitors, Make uses a visual, drag-and-drop canvas where you connect modules into branching "scenarios" — with routers, filters, iterators, and custom logic, plus the ability to run JavaScript or Python directly via the Make Code app for advanced users. That visual approach genuinely makes complex, multi-branch automations easier to understand and debug than a purely linear interface, and Make connects more than 3,000 apps, with AI Agents, an AI toolkit, and connections to 350+ AI apps letting you add intelligence directly into a scenario.
The genuinely important, honest nuance worth understanding clearly, especially if you're comparing Make directly against Zapier, already covered elsewhere in this directory: Make is genuinely 3-5 times cheaper per operation than Zapier, but "operations" and "tasks" aren't measured the same way — a single Zapier task can equal 3-8 Make operations for an equivalent multi-step workflow, meaning the real-world savings, while genuinely real, are smaller than the headline price-per-unit comparison suggests. It's also worth knowing about a real, current structural change: in August 2025, Make transitioned its billing unit from "operations" to "credits" — one credit still equals one module action, and routers and error handlers cost nothing, but AI modules now charge a variable number of credits depending on the tokens, files, or pages they process, adding a real layer of complexity that's caught some users off guard with workflows consuming more budget than expected. It's also fair to flag a genuinely specific, honest limitation relative to a developer-first competitor: Make's AI integration is largely built around basic HTTP requests, while n8n, also covered elsewhere in this directory, offers first-class AI nodes for LangChain, OpenAI, and vector databases specifically.
The nuanced operations-vs-tasks correction is drawn directly from a detailed independent 2026 pricing analysis (PxlPeak). The August 2025 credits transition and its complexity is drawn from an independent breakdown (Thinkpeak AI) and corroborated by AutomationAtlas's pricing page. The rebrand history is confirmed by ToolChase's independent review. The cancellation gotcha is drawn from AI PM Tools' documented experience.
Key features
Visual scenario builder
Drag-and-drop canvas with branching logic, easier to debug than a linear interface.
3,000+ app integrations
Including OpenAI, Salesforce, HubSpot, monday.com, and Slack.
AI Agents & AI toolkit
Connects to 350+ AI apps for adding intelligence directly into scenarios.
Make Code (JavaScript/Python)
Run custom code directly inside a scenario for advanced logic.
Routers, filters & iterators
Genuinely flexible data manipulation and branching for complex workflows.
Make ONE (enterprise governance)
SSO, audit logs, and data residency for larger organizations.
Pricing
Free
- Enough for roughly 200-300 simple workflow runs
- Not a trial — available indefinitely, but genuinely limited for business use
- 15-minute minimum scheduling interval
Core
- A 4-step workflow running 50 times/day uses roughly 6,000 credits/month
- Good fit for 2-3 active workflows of moderate complexity
- Monthly billing runs roughly 30% more than annual
Pro / Teams / Enterprise
- Enterprise adds Make ONE governance (SSO, audit, data residency)
- Scaling businesses can hit walls forcing five-figure Enterprise tiers
- Confirm current credit allowances directly given frequent plan changes
Cancellation requires confirming a link sent by email after you submit the request — missing that confirmation results in being billed again for the next cycle. Prices reflect Make's published pricing as of July 2026.
Available models
Integrations & platforms
Pros, cons & best for
Pros
- Genuinely cheaper than Zapier for equivalent workflows, real savings even if smaller than headline figures
- Visual canvas is a real, tangible advantage for understanding complex branching logic
- Old Integromat scenarios and templates still work with zero migration effort
Cons
- Steep learning curve for non-technical users, especially with complex data structures
- August 2025 credits transition adds real billing complexity, especially for AI modules
- AI integration is more basic (HTTP requests) compared to n8n's first-class AI nodes
Best for
- Operations, IT, and marketing teams comfortable mapping fields and conditions
- Teams needing complex, branching multi-step workflows on a real budget
- Not the pick for non-technical teams wanting the absolute simplest interface
Take a look inside
Alternatives
For the simplest possible interface or deeper native AI nodes instead:
Our verdict
Make's visual, branching scenario builder is genuinely one of the best interfaces in the category for understanding and debugging complex, multi-path automations, and the cost advantage over Zapier is real — just worth understanding precisely rather than taking the "3-5x cheaper" headline at pure face value, since operations and tasks aren't measured identically across the two platforms. The August 2025 shift to credit-based billing adds a real layer of complexity worth planning around, particularly for AI-heavy workflows where token-based module costs can be harder to predict than flat per-action pricing. For teams comfortable with a genuine learning curve who want serious branching logic and real, tangible savings over Zapier without dropping to a fully developer-first tool like n8n, Make remains a genuinely excellent, well-balanced choice.
FAQ
Is Make really 3-5x cheaper than Zapier?
Per operation, yes — but a single Zapier task can equal 3-8 Make operations for an equivalent multi-step workflow, so the real-world savings, while genuine, are smaller in practice than the headline per-unit price comparison suggests.
Is Make the same thing as Integromat?
Yes — Integromat was acquired by Celonis in 2020 and officially rebranded to Make.com in February 2022, with all accounts and scenarios migrated automatically and zero data loss; the Integromat name and brand are now fully retired.
Did Make's pricing structure change recently?
Yes, significantly — in August 2025, Make transitioned its billing unit from "operations" to "credits," where AI modules now charge a variable number of credits based on tokens, files, or pages processed, adding real complexity some users found their workflows consumed more budget than expected under the new system.
How does Make's AI integration compare to n8n's?
Make's AI integration is largely built around basic HTTP requests, while n8n offers first-class AI nodes specifically for LangChain, OpenAI, and vector databases — a real, specific advantage for AI-heavy, developer-oriented workflows.
Is there anything to watch out for when canceling a Make subscription?
Yes — cancellation requires confirming a link sent by email after you submit the request, and missing that confirmation means you'll be billed again for the next cycle.
Is Make's free plan enough for real business use?
It's genuinely usable for testing (1,000 credits a month, roughly 200-300 simple workflow runs), and it's not a time-limited trial, but it's limited to just 2 active scenarios, making it thin for ongoing production business needs.