Make.com Release | MAY'26
What do the MAY’26 Make.com updates really mean for your team?
We break down what matters, what to act on, and where it impacts your operations.
https://help.make.com/module-output-search-raw-data-view-undo-and-redo-gemini-flash
You get faster debugging and governance for complex automations: output search and raw data view let your team inspect, copy, and share exact input/output bundles when troubleshooting AI, CRM, and data pipelines. Undo and redo reduce rebuild risk during scenario edits, improving iteration speed and safety. Gemini 3.5 Flash strengthens rapid prototyping for multi-step workflow alternatives, but you should still enforce approval steps and validate outputs to prevent “autonomous” mistakes in production. Value is highest for maintainability, supportability, and controlled scaling.
https://help.make.com/introducing-make-skills
Make Skills lets you give AI assistants file-based Make know-how so they can generate scenarios, configure modules, map data flows, and work with the Make MCP server and APIs to create or modify connections and data structures. For your team, this can speed up builds and reduce errors in repeatable patterns, but you must treat it like a governed change system: validate outputs, enforce naming/standards, and test in non-prod. Expect biggest value on scaffolding and refactoring, less on complex edge-case logic.
https://help.make.com/unified-canvas-navigation
If you build large, complex automations, aligning canvas pan and zoom with common visual-tool gestures reduces retraining and “navigation friction,” especially for teams moving across devices. The guided first-visit hints lower onboarding time and help prevent mistakes caused by switching between navigation modes. For scaling, this improves builder adoption and speed, but it’s not a governance feature, so you still need consistent conventions, versioning, and documentation for maintainability.
https://help.make.com/2026
Stronger scenario operability for B2B teams: unified canvas and grid improvements, undo/redo, output search and raw data view, plus scenario recovery reduce build and troubleshooting time. Nested if else and merge enable more maintainable decision logic without spaghetti flows. Dynamic labels and new module/search and migration/CLI features improve governance and reduce broken automations. Expect meaningful admin/security impact from 2FA enforcement, credential request changes, and OpenAI model deprecations, so review skills, credentials, and dependencies regularly.
https://help.make.com/dynamic-labels-for-http-and-sleep-modules
Dynamic labels on HTTP and Sleep modules now reflect the request method and endpoint path, or the configured delay time directly on the canvas. For your team this improves faster audits, debugging, and handoffs, especially in multi-HTTP scenarios or staggered timing. It reduces “open-to-understand” friction but doesn’t change runtime behavior, so focus on governance via naming, consistent conventions, and version control to keep maintainability high.
https://help.make.com/new-version-of-api-endpoint-for-credential-requests-scenario-usage-tracking-connection-filtering-and-app-updates
You should plan a fast migration from the deprecated credential requests API to v2 by June 10, 2026, since parameter requirements are reduced and broken automations are the main risk. Expanded scenario usage tracking (beyond connections) improves governance, cost attribution, and impact analysis before refactors. Connection filtering helps you manage sprawl and ownership. App/module updates add better time-context handling, clearer token-based credit visibility, and new personalization modules, but require you to revalidate prompts, AI behavior, and cost controls.
https://help.make.com/new-mms-options-app-updates
You gain new Make Managed Services controls: you can create/manage child organizations and adjust credit allocations via API, plus monitor child data transfer and credit consumption. Operationally, this supports scaling across business units with clearer cost governance and better resilience, but it increases admin complexity and audit needs. App updates improve model access (Anthropic Claude), reduce integration friction (Google Business Profile rename, Typeform regional selection, Zoho multi-region), and harden connections (Zendesk OAuth refresh token; Zendesk legacy deprecation April 30, 2026). Plan migrations and validate prompts/limits to avoid silent behavior changes.
https://help.make.com/openai-gpt-54-nano-and-mini-now-available-in-make
You now have faster, cheaper GPT-5.4 mini and nano options in Make for multi-step and agentic workflows. Use mini for strong reasoning at lower cost, nano for high-volume repetitive tasks like extraction, classification, and routing, and reserve larger models for orchestration. This can materially improve throughput and cost in CRM lead triage, support ticket routing, and document processing, but you should set guardrails, validate outputs, and monitor accuracy to avoid scaling errors.
https://help.make.com/deprecated-openai-audio-preview-models
You need to audit every OpenAI module in your Make scenarios for the deprecated gpt 4o audio preview and gpt 4o mini audio preview models. New or updated modules can’t select them, and existing ones may fail at runtime. Replace them with gpt audio 1.5 and gpt audio mini, then re-test critical voice-to-action workflows. Also review pricing and usage, since audio model changes can shift costs and affect reliability.
https://help.make.com/new-make-functions-app
You can now build data transformations as standalone, connectable modules instead of nested IM (mapping) functions, which makes complex mapping logic easier to read, test, and maintain in RevOps-grade workflows. For your org, this improves governance because inputs and outputs are explicit, and changes are localized. Use transform module chaining for multi-step data prep. Watch for duplicated logic across scenarios and enforce conventions for module reuse naming and versioning.
https://help.make.com/nested-if-else-and-merge-is-now-available
You can now model multi-level business rules in one scenario using nested if-else and merge points, which makes complex routing easier to review and safer to change. This is especially useful when related conditions must stay together for governance and traceability. Operationally, it should reduce “router sprawl” and improve maintainability, but you still need discipline: keep branches readable, avoid deep nesting without clear merge strategy, and test decision paths thoroughly.
https://help.make.com/introducing-mcp-toolboxes
MCP toolboxes give you tighter, auditable control over which Make-connected AI tools (scenarios) an AI model can access, with per-toolbox keys for secure sharing across teams or clients. Strategically, this improves governance, cost predictability, and reliability by limiting tool sprawl and making tool usage visible. For maintainability, standardize toolbox design and tool annotations. Watch for complexity if you create too many toolboxes or don’t manage keys and permissions.
https://help.make.com/whatsapp-business-cloud-connections-bluesky-media-handling-new-xai-and-mondaycom-modules-and-more
You get several operational upgrades: clearer connection setup and stricter trigger typing for WhatsApp and GoHighLevel reduces connection-related failures; Netsuite and QuickBooks enhancements improve data correctness (erasing nulls, higher search/selection limits, invoice links) which strengthens CRM and billing automations; Bluesky media sizing reduces post-publishing errors; XAI adds tool-enabled AI responses for more capable, governed assistants; Monday aggregate table and Salesmate linking fields simplify scalable reporting and activity attribution. Main risk: module behavior changes and new connection requirements demand scenario QA and governance updates.
https://help.make.com/introducing-new-openai-skills-and-skill-versions-modules
You can now centralize repeated OpenAI instructions as reusable “skills” and version them in Make.com, then attach specific skills to model responses. This improves consistency across marketing, sales, and service automations, and makes updates safer because you change shared instructions once and control which version is the default. Operationally, treat skills like governed assets: use version pointers, review diffs before switching, and test downstream scenarios to avoid silent behavior changes. Value is high for repeated AI tasks; it’s limited for one-off prompts.