Make
Make is a visual automation platform that connects over 3,000 apps, data sources and AI models. Teams assemble workflows and AI agents by drag-and-drop, by prompt or in code, on a free plan that never expires.
What is Make?
Make is a visual automation and integration platform that presents itself as "the visual AI automation platform". It lets an organisation connect applications, data sources and AI models, then build and run automated processes across them. The product was previously called Integromat; the site keeps a dedicated page documenting that transition. Make states it was founded in 2015 and is now part of Celonis, the process-mining company its about page describes as valued at 13 billion dollars. The platform claims more than 400,000 client organisations across 200-plus countries and territories, and a team of over 350 people.
The working unit is the scenario: a chain of modules, each representing one action inside an application. Every module action consumes one credit, which is also the billing unit. Scenarios can be assembled in three ways, and this flexibility is central to the pitch: dragging and dropping modules in the visual editor, describing the intent in natural language to the assistant Maia, or going through MCP. The catalogue covers more than 3,000 pre-built applications, including over 300 GenAI apps, with named connectors for OpenAI, Salesforce, HubSpot, Slack, monday.com, NetSuite, Canva, Perplexity AI and DeepSeek. When an application is not covered, HTTP modules call any public API directly, and custom apps can be built.
The AI layer goes beyond connectors. Make offers reusable AI agents that adapt in real time and can be shared across several workflows, ready-made AI Tools modules for tasks such as categorisation or summarisation that need no prompt expertise, and a cloud-hosted MCP server that exposes scenarios to any AI interface. Make Grid adds a real-time visual map of the automation landscape, and an analytics dashboard tracks how workflows perform over time.
Make targets seven functional areas: IT, operations, marketing, sales, finance, customer experience and people. On the compliance side it holds SOC 2 Type II, SOC 3 and ISO 27001, claims GDPR compliance, and offers SSO, encryption and selectable data residency. It runs as a web application with an API available from the Core plan onwards, plus on-premise agents. Public ratings shown on the site are 4.8 on Capterra, 4.7 on G2, 4.8 on GetApp and 4.6 on Gartner.
What it does
- Connect more than 3,000 applications and any custom API into a single automated workflow
- Build automation scenarios visually by drag-and-drop, in natural language with Maia, or through MCP
- Deploy reusable AI agents that adapt in real time and operate across several workflows
- Call any public API and receive external events through HTTP modules and webhooks
- Transform, filter and route data between systems with built-in functions and conditional logic
- Monitor usage and performance of every workflow through the analytics dashboard and execution logs
- Govern access with team roles, per-team limits and shareable scenario templates
When to use Make / When not to
A quick filter to help you decide if Make is the right fit.
When to use Make
- Operations managers and business analysts who need to connect siloed SaaS systems without writing code
- Marketing and sales teams automating lead routing, CRM synchronisation and campaign reporting
- IT and DevOps teams automating monitoring, incident response and internal provisioning
- Finance and procurement teams running quote-to-cash and procure-to-pay processes across several tools
- Startup founders, freelancers and consultants who want a serious automation layer at a low entry price
When not to use Make
- Teams who need a mobile app: Make is a web platform, with no iOS or Android application
- Organisations that require a named support email or a phone line, since support runs solely through the Help Center, the community and web forms
- Non-English-speaking teams who need a localised interface, documentation or support
- European organisations that require German data residency but are not on an Enterprise contract, as other customers are hosted in Ireland
- Teams with high or unpredictable volumes who cannot absorb a per-action credit model in which unused credits never roll over
How to use Make
A typical end-to-end flow, from setup to results.
- Create an account free of charge; no credit card is required and the free plan has no time limit
- Pick the applications to connect from the catalogue of more than 3,000 pre-built apps
- Authorise each application by connecting the relevant account to Make
- Create a scenario and place the trigger module that will start it, such as a webhook or a scheduled check
- Add the action modules that follow, either by drag-and-drop in the visual editor or by describing the goal to Maia
- Insert routers and filters so that each branch runs only under the conditions you set
- Map and transform the data passed between modules using the built-in functions
- Use an HTTP module to reach any public API that is not covered by an existing connector
- Run the scenario, then read the execution log to confirm each step behaved as expected
- Schedule the scenario, then monitor credits and performance from the analytics dashboard
Pros & Cons
Pros
- A visual, no-code builder that keeps a technical escape hatch through HTTP modules, custom apps and the Make API
- A permanently free plan with no credit card and no time limit, then a low paid entry point at 9 USD per month
- A broad catalogue of more than 3,000 applications, including over 300 GenAI apps
- A serious, publicly documented compliance file: SOC 2 Type II, SOC 3, ISO 27001, DPA, SCC and a versioned sub-processor list
- Selectable data residency, with European customer data kept on servers located in the European Union
- Backing by Celonis, an established group, which reduces the risk of the platform disappearing
- A dense learning ecosystem: Make Academy, community, Help Center, Developers Hub and partner directory
Cons
- The credit model is hard to forecast, since every single module action consumes a credit
- Unused credits expire at the end of the term and never roll over
- Scenarios stop running as soon as credits run out, until credits are added again
- No support email is published: help goes through the Help Center, the community or a form
- No mobile application, and the site and documentation are available in English only
- Key capabilities are gated by plan: the Make API from Core, teams and roles from Teams, German hosting from Enterprise
- The vendor takes no general public position on whether customer data is used to train AI models
Pricing & Plans
Make offers a permanently free plan, with no credit card required and no time limit, providing up to 1,000 credits per month. The lowest paid entry point is the Core plan at 9 USD per month for 10,000 credits per month. Annual billing is advertised as saving 15% or more, and the Enterprise plan is quoted on request.
- 0 USD per month
- up to 1
- 000 credits per month
- including the no-code visual builder
- 3
- 000+ apps
- routers and filters
- customer support
- 9 USD per month for 10
- 000 credits per month
- adding unlimited active scenarios
- scheduling down to the minute
- higher data transfer limits and access to the Make API
- 16 USD per month for 10
- 000 credits per month
- the plan recommended on the site
- adding priority scenario execution
- custom variables and full-text search in the execution log
- 29 USD per month for 10
- 000 credits per month
- adding teams
- team roles and the ability to create and share scenario templates
- custom pricing on request
- adding custom functions
- enterprise app integrations
- 24/7 support
- access to the Value Engineering team
- overage protection and advanced security features
Data, GDPR & hosting
A consolidated view of how Make handles your data.
GDPR overview
GDPR compliance is claimed explicitly on the homepage and the security page, and it is backed by documents rather than by claims alone. Make publishes a Data Processing Agreement (May 2024), Standard Contractual Clauses under Module Two (controller to processor), an International Data Transfer Addendum for the United Kingdom, its Technical and Organisational Measures, and a versioned list of sub-processors going back to February 2024. Celonis Inc. and Celonis Labs LLC are certified under the EU-U.S. Data Privacy Framework, its UK Extension and the Swiss-U.S. Framework. The privacy contact is privacy@celonis.com and the notice was last updated on 20 July 2026. Two gaps remain: an external data protection officer is mentioned for Celonis SE without any contact details, and no EU representative under Article 27 is named in the documents consulted.
Who owns the data?
The contracting entity is Celonis, Inc., based in New York, and the customer keeps ownership of the data pushed through the platform. The published Data Processing Agreement places Celonis in the role of processor: it processes personal data only on the customer's documented instructions. The DPA states that Celonis will not sell personal data, will not retain, use or disclose it for any commercial purpose, and will not combine it with data from another source. Processing may involve six named Celonis affiliates in Germany, the Czech Republic and the United States, plus eighteen publicly listed sub-processors. The Master Services Agreement guarantees that customer data can be exported or downloaded at any time during the subscription term.
Reuse rights
Customers keep full practical control over the data they run through Make. The Master Services Agreement states that customer data remains available for export or download at any point during the subscription term, so it can be reused, migrated or archived elsewhere without asking permission. On the vendor side the Data Processing Agreement is restrictive: Celonis acts only on documented instructions and explicitly undertakes not to sell the data, not to retain, use or disclose it for any commercial purpose of its own, and not to enrich it with information from other sources. Where AI features are used, data is passed to named third-party providers such as OpenAI, Anthropic, Google, Microsoft and Groq. One point stays open: beyond a statement that Google Workspace API data is not used to develop, improve or train generalised AI or ML models, the site sets out no general position on model training.
Data retention & training
Hosting summary
Make runs on private Amazon AWS EC2 instances inside an Amazon VPC, with the cluster deployed across two zones for availability. Enterprise customers get a separately managed, isolated AWS environment. Hosting location is selectable: the United States, Germany for enterprise customers, or Ireland for everyone else. The scenario database, MongoDB, follows the same rule for EU customers and sits in the United States for non-EU customers, as does the Supabase database behind Make and Grid, which uses Ireland for EU customers. Some sub-processors host elsewhere: support ticketing in Germany, product analytics in the Netherlands, and one AI provider in France. Cloudflare serves as a global CDN, but the sub-processor list states that data of European customers is processed only via servers located in the European Union. Data is encrypted at rest with AES-256 and AWS KMS, and in transit with TLS 1.2 and 1.3. Internal access is restricted to the private network over VPN.
Things to keep in mind
Risks and trade-offs to weigh before adopting Make.
- Billing is per module action and credits expire at the end of the term, so an unoptimised scenario can quietly consume a monthly allowance
- Automation stops dead when credits run out, which can silently break a business process unless automatic purchasing or Enterprise overage protection is in place
- Using the AI features sends data to third-party providers such as OpenAI, Anthropic, Google, Microsoft and Groq, most of them hosting in the United States
- No general public position is taken on whether customer data is used to train AI models, and no training opt-out is documented
- German data residency is reserved for enterprise customers, while other European customers are hosted in Ireland
- No minimum age is stated in the contractual documents, and no EU representative under Article 27 is designated
- Automating a process end to end can erode the team's understanding of it: when a scenario breaks, whoever built it may no longer be there to explain the logic
Setup & Integrations
Technical difficulty
Getting started is easy: registration is free, needs no credit card, and simple scenarios can be built entirely by drag-and-drop, by prompt through Maia, or from ready-made templates. The site claims moving from idea to execution takes hours rather than weeks. Difficulty rises with ambition. Routers, filters, data-transformation functions and especially HTTP modules assume you understand APIs, authentication and data formats. Make Academy, the Help Center, the community and the Developers Hub cover the learning curve, and a partner directory plus a managed service exist for complex deployments.
Deployment
Integrations
Supported languages
Behind Make
Fundraising
Social
Resources
All the official URLs gathered for verification and reference.
Alternatives
Tools that compete with or complement Make.
Frequently asked questions
Is there a free plan, and does it expire?
How much does the cheapest paid plan cost?
What exactly is a credit?
What happens if I run out of credits?
Does Make provide an API?
How many applications can Make connect to?
Where is my data hosted?
Is Make GDPR compliant?
Is my data used to train AI models?
Which alternatives does Make itself compare against?
Should you pick Make?
Make occupies a clear position: a visual automation platform that stays approachable for people who do not write code, without capping what technical users can do. The drag-and-drop builder, the natural-language assistant and the catalogue of more than 3,000 applications cover the common cases, while HTTP modules, custom apps and the Make API handle everything the catalogue misses. The AI layer, with reusable agents and a hosted MCP server, is genuinely integrated rather than bolted on.
The commercial terms are unusually open for this category. A free plan that never expires, no credit card at signup and a paid entry point at 9 USD per month make evaluation cheap. The compliance file is equally substantial and, more importantly, public: SOC 2 Type II, SOC 3, ISO 27001, a Data Processing Agreement, Standard Contractual Clauses, Technical and Organisational Measures and a versioned list of eighteen sub-processors. Being part of Celonis adds a degree of durability that smaller automation vendors cannot offer.
Three reservations deserve attention before committing. First, the credit model bills every module action and credits do not roll over, so a verbose scenario can become expensive and an exhausted balance halts automation outright. Second, support has no published email address and runs entirely through the Help Center, the community and forms, which suits self-service teams better than those needing a named contact. Third, the vendor states no general position on whether customer data trains AI models, and the AI features route data to third-party providers hosted mainly in the United States.
For teams connecting several SaaS tools and willing to model their credit consumption, Make is a strong and well-documented choice. The site itself invites comparison with Zapier and n8n.
- Choosing a selection results in a full page refresh.
- Opens in a new window.