Make (Integromat)

Make provides a canvas-based workflow editor where users connect pre-built modules (triggers, actions, filters) by dragging connectors between nodes. Each module encapsulates API calls, data transformations, or conditional logic; the platform compiles the visual graph into executable workflows that execute sequentially or in parallel based on connection topology. The builder validates module compatibility (input/output schema matching) in real-time and generates execution plans without requiring code.

Make maintains a library of 1,500+ pre-configured connectors that abstract away API authentication, pagination, rate limiting, and response parsing for popular SaaS platforms (Salesforce, HubSpot, Slack, Google Workspace, etc.). Each connector is a module template with pre-mapped fields, error handling, and OAuth/API key management built-in. The platform handles credential storage in encrypted vaults and automatically refreshes tokens, eliminating manual API integration work.

Make supports team workspaces where multiple users can collaborate on scenarios, with role-based access control (admin, editor, viewer). Scenarios can be shared within teams, and changes are tracked with basic audit logs. The platform allows teams to manage shared API credentials, set workspace-level quotas, and organize scenarios into folders. Collaboration features include scenario locking (to prevent simultaneous edits) and execution history visibility across team members.

Make offers a free tier enabling users to build and execute unlimited workflows without providing a credit card or payment information. The free tier includes access to the visual builder, all 3,000+ connectors, and unlimited scenario executions (subject to fair-use policies). Limitations on the free tier are not documented but typically include reduced API rate limits, limited team members, or reduced execution priority compared to paid tiers. The free tier enables users to prototype and learn Make before committing to paid plans.

Capability enabling workflows to handle errors gracefully through conditional branching based on error types or execution outcomes. Users configure error handlers (alternative paths) that execute when a node fails, enabling workflows to retry, skip, or take corrective action. Conditional branching supports decision logic based on previous node outputs, enabling workflows to route around failures or implement fallback logic. Specific error handling mechanisms (automatic retries, exponential backoff, dead-letter queues) are not documented.

Make provides filter and router modules that evaluate conditions on data flowing through the workflow (e.g., 'if email domain is @company.com, route to Slack channel A, else route to channel B'). Conditions are built using a visual condition builder supporting AND/OR logic, comparison operators, and data field references. The platform evaluates conditions at runtime and directs execution to different downstream modules based on results, enabling dynamic workflow behavior without code.

Make integrates AI modules (powered by OpenAI, Anthropic, or other LLM providers) that accept text prompts and data inputs, then generate or transform content within workflows. Users configure prompts with variable placeholders (e.g., 'Summarize this customer feedback: {{feedback}}'), and the module substitutes runtime data, sends the request to the LLM API, and returns generated text. This enables AI-powered content creation, summarization, translation, and data enrichment without leaving the workflow builder.

Make supports multiple trigger types: scheduled timers (run every hour/day/week), webhook endpoints (run when external system POSTs data), app event subscriptions (run when Salesforce record is created), and manual triggers (run on-demand). Triggers are configured as the first module in a scenario; the platform manages trigger registration, polling intervals, and event delivery. Scheduled triggers use cron-like syntax; webhooks generate unique URLs that external systems can call; app event triggers subscribe to native APIs and receive real-time notifications.

Make provides a real-time execution dashboard showing workflow runs with step-by-step status (success, error, skipped), execution duration, and detailed logs for each module. Users can inspect input/output data for each step, view error messages, and replay failed executions. The platform stores execution history and provides filtering/search to find specific runs. Error handling includes automatic retries, fallback paths, and error notification modules that alert users when workflows fail.

Make provides data transformation modules that map, filter, and reshape data flowing through workflows. The platform includes a visual data mapper that allows field-to-field mapping (e.g., map 'firstName' from source to 'first_name' in destination), array operations (flatten, group, filter), and custom functions for complex transformations. Users can write JavaScript expressions (e.g., '{{firstName}} {{lastName}}' to concatenate) or use built-in functions (date formatting, string manipulation, math operations) without leaving the workflow builder.

Make generates unique webhook URLs for each scenario that external systems can POST data to, triggering workflow execution. Additionally, the platform provides HTTP request modules that allow workflows to make GET/POST/PUT/DELETE calls to any REST API, with support for headers, authentication (API key, OAuth, basic auth), request body templating, and response parsing. This enables integration with custom or niche APIs not covered by pre-built connectors.

Make provides a library of pre-built scenario templates for common use cases (e.g., 'sync Salesforce leads to HubSpot', 'send Slack notifications for new emails'). Users can clone templates, customize them for their specific apps and data, and save custom scenarios as reusable templates. The platform supports scenario versioning and allows teams to share templates across workspaces, reducing the need to rebuild common workflows from scratch.

Make enforces execution quotas based on pricing plan (free plan: 1,000 operations/month, paid plans: higher limits). The platform tracks operation counts across all scenarios and prevents execution when quotas are exceeded. Users can monitor quota usage in the dashboard and receive warnings as they approach limits. The platform also implements rate limiting per module to prevent overwhelming third-party APIs, with configurable delays between requests.