Make

Make is a visual automation platform for connecting apps, data sources and AI models into workflows built by drag-and-drop, by prompt or via MCP. Formerly Integromat and now part of Celonis, it serves over 400,000 organisations and adds AI Agents that reason and act on the same canvas as deterministic scenarios.

Make is a visual automation platform, launched as Integromat in 2015 and now part of process-mining company Celonis, that positions itself as 'the visual AI automation platform' for connecting 'any app, data source, or AI model' into workflows built visually, in code, or by prompt. Its differentiating primitive is the canvas: scenarios are drawn as branching graphs with routers, filters, iterators and aggregators rather than the linear trigger-then-action chain most competitors use, which is why it is repeatedly chosen over Zapier for multi-branch logic. The app catalogue lists more than 3,000 pre-built integrations — the directory showed 3,546 apps at time of writing — spanning Salesforce, HubSpot, NetSuite, Microsoft 365, Google Workspace, Slack, Notion, Airtable, Shopify and Stripe, with a generic HTTP module for anything with an API. On the AI side Make integrates OpenAI, Anthropic Claude, Google Vertex AI, Mistral, Azure OpenAI, Hugging Face, Perplexity and ElevenLabs, and its AI Agents feature — with a next-generation release announced 11 February 2026 and availability on all plans — lets agents reason about a situation and choose their next step while running on the same canvas as deterministic scenarios, so teams can build, run and debug both without switching tools. The company reports a global team of 350-plus and more than 400,000 organisations across 200-plus countries as customers, naming BNY Mellon, Stellantis, BambooHR and Bolt among them. Security posture is enterprise-grade: SOC 2 Type II and SOC 3 audits completed, an ISO 27001-certified information security program for the enterprise platform, GDPR adherence, AES-256 encryption at rest, TLS 1.2/1.3 in transit, AWS KMS key management and a 99.5% cloud uptime commitment for Enterprise.

Category

Automation & Workflows

Pricing

Freemium, Subscription, Usage-based

Target Market

CIOs, Operations Leaders, Marketing Teams, RevOps Teams

Deployment

Cloud-only

Founded

2015

Team Size

201-500

Customers

400,000+ organisations across 200+ countries

• ✓
  Visual branching canvas

  Scenarios are drawn as graphs with routers, filters, iterators and aggregators, expressing multi-path logic that linear trigger-action tools cannot.

• ✓
  3,000+ app integrations

  A catalogue of over 3,500 listed apps covering Salesforce, HubSpot, NetSuite, Microsoft 365 and Google Workspace, plus an HTTP module for anything else.

• ✓
  AI Agents on the same canvas

  Agents reason about context and choose next steps rather than following fixed logic, and are built, run and debugged alongside existing scenarios.

• ✓
  Multiple build modes

  The same automation can be assembled by drag-and-drop, generated from a prompt, or driven through Model Context Protocol from an AI client.

• ✓
  Multi-provider AI model access

  Native modules for OpenAI, Anthropic Claude, Google Vertex AI, Mistral, Azure OpenAI, Hugging Face, Perplexity and ElevenLabs avoid single-vendor lock-in.

• ✓
  Make API and developer platform

  Available from the Core plan up, letting engineering teams manage scenarios programmatically and build custom apps for the catalogue.

• ✓
  Enterprise security and compliance

  SOC 2 Type II and SOC 3 audits, an ISO 27001-certified security program, AES-256 at rest, TLS 1.2/1.3 in transit and AWS KMS key management.

• ✓
  Execution logging and search

  Full-text execution log search on Pro and above, with 30-day retention by default and longer retention available for Enterprise customers.

• •
  CRM and marketing stack synchronisation

  Keeps Salesforce or HubSpot records in step with billing, support and marketing tools without a custom integration service to maintain.

• •
  Lead routing with conditional logic

  Routers and filters send each inbound lead down a different path by score, region or product, then notify the right owner.

• •
  AI-assisted content and support workflows

  An LLM module drafts a reply or summary mid-scenario, with the deterministic steps around it handling retrieval and delivery.

• •
  E-commerce order and fulfilment automation

  Shopify or WooCommerce orders trigger inventory updates, Stripe reconciliation and customer notifications across systems that do not talk natively.

• •
  Agentic operations across many apps

  An AI Agent assesses an incoming request, decides which of 3,000-plus connected apps to act in, and executes the steps.

List pricing is published for every tier except Enterprise: Free at 1,000 credits a month, then Core at $12, Pro at $21 and Teams at $38 a month, each including 10,000 credits with annual billing saving about 15%. The meaningful cost driver is credits, not seats — every module run in a scenario consumes them — so a workflow that loops over records burns budget far faster than the plan price suggests, and Capterra reviewers cite exactly this escalation. Overage protection, SSO, custom functions and extended log retention are all gated behind Enterprise, which is quote-only.

This page was written from 7 sources, 2 on domains other than make.com.

1. 1.make.com — envendor
2. 2.make.com — pricingvendor
3. 3.make.com — securityvendor
4. 4.make.com — ai agentsvendor
5. 5.make.com — integrationsvendor
6. 6.capterra.com — Integromat
7. 7.hn.algolia.com — hn.algolia.com