Checked October 4, 2026

One product, in depth

Make review and alternatives

By SaaS Think Tank
Checked 4 October 2026
Sources & limits ↓

Our view

Make suits workflows with different paths.

Make suits app workflows with different paths. Our test saved six valid requests. A later automatic retry saved one request twice after a lost reply.

Consider Make for

An app workflow with several paths and someone who checks saved results when a reply fails.

Look closely at an alternative if

You need to run the workflow service yourself, inspect its code or keep backups under your control.

See where the alternatives fit ↓
Try Make
The workflow we testedMake · 2 Oct 2026
Request queue. Six saved requests, one invalid and one repeated request.Frame from our published recording

Make against Zapier and n8n.

  1. Make vs Zapier

    42 of 57Capabilities Make and Zapier both had
    2Confirmed differences
    13Capabilities with an unconfirmed answer

  2. Make vs n8n

    91 of 167Capabilities Make and n8n both had
    0Confirmed differences
    76Capabilities with an unconfirmed answer

Make and the named competitor both had the capability in 133 of 224 comparison checks. A capability checked against two competitors counts twice. See the dated differences ↓

The recorded result

The test saved six valid requests.

The published Request queue workflow, recorded on 2 October 2026. Choose a moment to watch.

What this proves

The clip shows Make’s run history; saved results are in the comparison. A later automatic retry saved one request twice after a lost reply.

What you get

What Make can do.

167 capabilities from the latest comparison. These are that comparison’s answers, not capabilities checked twice. Search above for the rest, including pricing and product details.

167 Make capabilities from the latest published comparison
CapabilityMake
Build in a visual editorArrange and connect workflow steps on a visual canvas. YesScenario Builder canvas; ease of use untested.Checked 2 October 2026Make source ↗
Receive a webhook eventStart a workflow when data arrives at a web address supplied by the workflow. YesCustom and app webhooks; queue and rate limits can affect delivery.Checked 2 October 2026Make source ↗
Start on a scheduleStart a workflow at a chosen time or interval. YesMinimum interval depends on plan.Checked 2 October 2026Make source ↗
Filter an invalid itemStop an item that fails a stated condition before later steps. YesA filter on a module connection stops nonmatching bundles.Checked 2 October 2026Make source ↗
Route an item by a conditionSend an item down a branch chosen by its values. YesRouter filters and fallback; routes process sequentially.Checked 2 October 2026Make source ↗
Transform a field without codeChange a field with built-in visual tools rather than custom code. YesVisual Functions modules; specific transform and effort untested.Checked 2 October 2026Make source ↗
Run JavaScript or Python in a stepExecute JavaScript or Python inside a workflow step. YesMake Code is paid-plan only and the published charge is two credits per second; runtime and library limits vary by plan.Checked 2 October 2026Make source ↗
Call a third-party APISend an HTTP request to another service through its API. YesHTTP Make a request module; v4 requires secure HTTPS with a verified certificate, and target API permissions are separate.Checked 2 October 2026Make source ↗
Start from a workflow templateLoad a saved starting point and reconnect its apps. YesScenario templates; required connections and settings still need setup.Checked 4 October 2026Make source ↗
Repeat a step for each list itemSend each item through the same set of steps. YesIterator splits an array into bundles.Checked 4 October 2026Make source ↗
Combine items into one resultCollect multiple items before the next step. YesAggregator combines bundles in a selected structure.Checked 4 October 2026Make source ↗
Process items in batchesControl how many items reach a step at once. YesRepeater and flow controls build the loop; choose its stop conditions.Checked 4 October 2026Make source ↗
Pause before the next stepHold a run until a delay or response ends. YesSleep delays a run; the checked guide caps a delay at 300 seconds.Checked 4 October 2026Make source ↗
Call another workflowReuse a smaller workflow and pass data back. YesCall a scenario, Start scenario and Return output modules.Checked 4 October 2026Make source ↗
Define workflow inputs and outputsName the data a reusable workflow receives and returns. YesTyped scenario inputs and outputs work with subscenarios, APIs and agent tools.Checked 4 October 2026Make source ↗
Add notes to the canvasKeep explanations beside the workflow steps. YesScenario notes document the setup.Checked 4 October 2026Make source ↗
Organize workflows in foldersGroup related workflows for easier browsing. YesSubfolders organize scenarios.Checked 4 October 2026Make source ↗
Start a workflow from a formCollect field values through a form submission. UnconfirmedConnected form apps can trigger scenarios; an equivalent native form builder is unconfirmed here.Checked 4 October 2026Make source ↗
Publish or activate a workflowMake a saved workflow available to its trigger. YesActivate a scenario after saving and configuring its schedule.Checked 4 October 2026Make source ↗
Run a workflow manuallyStart a workflow from its editor. YesRun once starts a scenario manually; connected actions can write real data.Checked 4 October 2026Make source ↗
Edit code in the built-in editorWrite and change code inside the workflow editor. YesMake Code includes syntax highlighting, completion and error highlighting; paid plans only.Checked 4 October 2026Make source ↗
Validate incoming workflow fieldsCheck input fields before later steps use them. YesScenario inputs can be required and typed; Make validates collection fields against their specification. This does not validate every external payload automatically.Checked 4 October 2026Make source ↗
Map data between stepsUse earlier output as an input to another step. YesMapping fields select earlier bundle values.Checked 4 October 2026Make source ↗
Use expressions in a fieldCalculate a field value while the workflow runs. YesMake functions run in mapping fields.Checked 4 October 2026Make source ↗
Reuse configuration variablesRead named settings across workflows. YesCustom variables need Pro, Teams or Enterprise; team/organization values are stored in plain text, so do not store secrets.Checked 4 October 2026Make source ↗
Store records inside the platformSave data for later workflow runs. YesData stores persist data across scenarios; storage allowances apply.Checked 4 October 2026Make source ↗
Create, read, update and delete stored recordsManage records in the platform’s own storage. YesData stores provide record operations.Checked 4 October 2026Make source ↗
Pass files between stepsRetrieve a file and send it to another step. YesFile data and name map between modules; maximum file size varies by plan.Checked 4 October 2026Make source ↗
Browse the integration directoryFind connectors before choosing the platform. YesOfficial Community Apps guide says community apps appear with public apps on the Integrations page; individual actions need their own check.Checked 4 October 2026Make source ↗
Build a custom connectorWrap an API in reusable workflow steps. YesCustom Apps development defines an app for a service without a ready-made connector.Checked 4 October 2026Make source ↗
Install community-built connectorsAdd connectors made outside the vendor’s core set. YesCommunity apps install into an organization through an invite link; partners set their own price and terms.Checked 4 October 2026Make source ↗
Choose a connection at runtimeUse a different saved app login for each run. YesDynamic connections are Enterprise-only.Checked 4 October 2026Make source ↗
Use a command-line clientManage platform resources from a terminal. YesMake CLI calls the platform API and can support external Git or deployment scripts.Checked 4 October 2026Make source ↗
Manage workflows through an APIRead or change workflows from another program. YesScenario API endpoints create, manage and execute scenarios; paid plan and selected token scopes apply.Checked 4 October 2026Make source ↗
Import a cURL requestTurn a pasted API request into step settings. UnconfirmedThe HTTP guide reviewed previously does not establish this exact import action.Checked 4 October 2026Make source ↗
Paginate an API responseFetch further pages of results from a service. UnconfirmedExact automatic pagination setup remains unconfirmed in the reviewed HTTP reference.Checked 4 October 2026Make source ↗
Remove duplicate itemsKeep one copy of matching items in a list. YesArray functions remove duplicate items by value; durable write protection needs a separate design.Checked 4 October 2026Make source ↗
Create reusable custom functionsSave code that can be called from multiple workflows. YesEnterprise JavaScript custom functions extend mapping.Checked 4 October 2026Make source ↗
Send error notifications to another serviceNotify an operator when configured error handling runs. YesBuild a notification action on an error route; destination app setup remains separate.Checked 4 October 2026Make source ↗
Give an API key selected permissionsLimit which platform resources a program may access. YesScenario API requires selected token scopes such as scenarios:read; endpoint and paid-plan access apply.Checked 4 October 2026Make source ↗
Load field choices dynamicallyWhen building a custom connector, fetch selectable field options from its service. YesCustom app authors use a dynamic-options RPC to populate a select field from the service API. This is connector development, not a setting on every supplied app.Checked 4 October 2026Make source ↗
Extract file metadataRead a stored knowledge file’s descriptive fields for later steps. YesKnowledge > Get knowledge returns name, description, date and ID for a selected knowledge file. This does not promise arbitrary file-format metadata extraction.Checked 4 October 2026Make source ↗
Change stored data in bulkApply a record change to multiple stored items. UnconfirmedData stores support record operations, but a comparable multi-record update/delete operation is not established in this guide; iteration is a separate route.Checked 4 October 2026Make source ↗
Reuse stored output in later workUse a previous output as input to another run. YesData stores persist records that later scenarios can retrieve; storage allowance depends on plan.Checked 4 October 2026Make source ↗
Clean incoming dataTrim or replace text and remove unwanted values before the next step. YesText functions include trim, replace and case conversion; configure the rule for the incoming field.Checked 4 October 2026Make source ↗
Start a workflow from another programSend a request from another program to start a workflow. YesMake’s scenario API includes a run endpoint; paid endpoint access, token scopes and scenario inputs apply.Checked 4 October 2026Make source ↗
Build an agent that chooses toolsLet a model choose an allowed action to perform. YesAI Agent (New) chooses connected tools; open beta can change.Checked 4 October 2026Make source ↗
Call an agent from another agentDelegate part of a request to a second agent. UnconfirmedThe reviewed agent reference does not establish this exact nested-agent configuration.Checked 4 October 2026Make source ↗
Connect a model providerUse a model service in the workflow. YesMake’s own provider works on all plans; custom agent provider connections need a paid plan.Checked 4 October 2026Make source ↗
Give an agent toolsLimit which actions an agent can call. YesModules, scenarios and remote MCP tools can be attached.Checked 4 October 2026Make source ↗
Keep conversation memoryCarry earlier messages into the next agent response. YesConversation ID joins messages; history size is configurable.Checked 4 October 2026Make source ↗
Return a structured agent answerAsk an agent to return named fields. YesResponse format accepts a defined data structure or JSON.Checked 4 October 2026Make source ↗
Add files to an agent’s knowledgeMake reference material available to an agent. YesKnowledge modules upload, update, delete and search reference files.Checked 4 October 2026Make source ↗
Search a vector databaseFind stored content by meaning. UnconfirmedDedicated vector-store control is unconfirmed by the reviewed knowledge-file guide.Checked 4 October 2026Make source ↗
Run a model on your own infrastructureCall a model service you operate yourself. UnconfirmedExact local-model access through the managed agent provider remains unconfirmed.Checked 4 October 2026Make source ↗
Call tools on an MCP serverUse tools published by another AI service. YesMCP client tools attach to agents; account permissions and remote transport apply.Checked 4 October 2026Make source ↗
Expose a workflow as an MCP toolLet an AI client call the workflow. YesActive on-demand scenarios expose defined inputs and outputs; Free and paid authentication differ.Checked 4 October 2026Make source ↗
Chat with an agent during testingSend sample requests before publishing the agent. YesAgent chat tests behavior in the editor.Checked 4 October 2026Make source ↗
Provide a hosted chat entry pointLet other people send a chat message to the workflow. UnconfirmedThe reviewed test-chat reference does not establish an equivalent public hosted chat page.Checked 4 October 2026Make source ↗
Require approval before an agent tool runsPause a proposed tool action until a person approves it. UnconfirmedThe reviewed agent reference does not establish an equivalent native approval gate.Checked 4 October 2026Make source ↗
Build or edit using natural languageAsk an assistant to create or change workflow steps. YesMaia is in public beta on paid plans; Free has a 30-day trial from signup. Inspect generated changes.Checked 4 October 2026Make source ↗
Rerank retrieved search resultsReorder results by how well they match a query. UnconfirmedThe reviewed knowledge guide does not establish a native reranking control.Checked 4 October 2026Make source ↗
Ground answers with a knowledge graphUse relationships in a graph as an answer source. UnconfirmedThe reviewed knowledge-file guide does not establish a native graph-grounding control.Checked 4 October 2026Make source ↗
Filter retrieved documents by metadataLimit the reference material returned to an agent. UnconfirmedThe reviewed knowledge guide does not establish arbitrary metadata filters.Checked 4 October 2026Make source ↗
Combine keyword and meaning-based searchUse two search methods when retrieving reference content. UnconfirmedNative hybrid search is unconfirmed in the reviewed knowledge guide.Checked 4 October 2026Make source ↗
Inspect AI costs by modelBreak down model usage into monetary costs. UnconfirmedAgent token totals do not establish a monetary cost breakdown by model.Checked 4 October 2026Make source ↗
Search the web through an AI toolAsk a model-backed tool for web results. YesMake AI Web Search exposes a search module; model cost and accuracy remain separate.Checked 4 October 2026Make source ↗
Choose between an AI chain and an agentSelect fixed model steps or model-selected tool calls. UnconfirmedThe reviewed platform guides do not establish this exact operation. Verify the chosen component and access.Checked 4 October 2026Make source ↗
Combine model choices with fixed rulesUse both model-selected actions and fixed conditions in one workflow. YesRun an agent is a scenario module and can use module or scenario tools; fixed scenario steps remain separate. Open beta.Checked 4 October 2026Make source ↗
Run an action after an AI answerUse an agent’s answer to trigger the next workflow action. YesThe Run an agent module emits output for later scenario modules; confirm the target app action separately. Open beta.Checked 4 October 2026Make source ↗
Process data with an AI modelSend workflow data to a model and use its response. YesRun an agent module sends the prompt to the selected model; beta and provider access apply.Checked 4 October 2026Make source ↗
Process unstructured contentTurn content without fixed fields into workflow data. UnconfirmedThis exact operation remains unconfirmed. Related features don’t establish support.Checked 4 October 2026Make source ↗
Ingest and index reference dataLoad reference content into a searchable store. UnconfirmedThis exact operation remains unconfirmed. Related features don’t establish support.Checked 4 October 2026Make source ↗
Switch between model providersChange the model provider without rebuilding the workflow. YesAgent connection selects a provider; Make provider is on all plans, custom providers require paid access. Open beta.Checked 4 October 2026Make source ↗
Configure a selected modelChoose model settings for a workflow step. YesChoose the provider model, reasoning effort and maximum output length. Open beta; custom providers require paid access.Checked 4 October 2026Make source ↗
Extract structured fields from textTurn text into named fields for later steps. UnconfirmedThis exact operation remains unconfirmed. Related features don’t establish support.Checked 4 October 2026Make source ↗
See the inputs and outputs of a saved runInspect the data entering and leaving the steps of a saved run. YesHistory details show module bundles, with input/output fields; plan retention and saved-log settings matter.Checked 2 October 2026Make source ↗Make source 2 ↗
Search or filter earlier runsNarrow a list of past runs with the available search or filter controls. YesScenario-run search is Pro or higher; Free/Core search scenarios only.Checked 2 October 2026Make source ↗
Retry an incomplete or failed runResume or repeat work from a run that did not finish successfully. YesRetries failed module with prior configuration; scenario must be active.Checked 2 October 2026Make source ↗
Replay a completed run from its triggerStart the current workflow again with saved trigger data from a completed run. YesAll plans can replay old trigger data through every module of the current scenario, using current variables and consuming credits; retention and repeated writes matter. Poll checks with no data and isolated non-trigger module runs are excluded.Checked 2 October 2026Make source ↗
Route an error for separate handlingSend a failed step to a separate recovery path. YesSkip, Retry, Resume, Commit and Rollback paths have different effects.Checked 2 October 2026Make source ↗
Restore an earlier workflow versionReturn a workflow definition to an earlier saved version. YesManually saved versions up to 60 days; save again after restore. An older no-undo statement conflicts with newer undo docs.Checked 2 October 2026Make source ↗
Reuse saved data while testingTest changes without asking the source for fresh data. YesRun replay uses saved trigger data; current variables and later modules can run again.Checked 4 October 2026Make source ↗
Create mock data for a testSupply fictional inputs before connecting live systems. UnconfirmedA dedicated mocking control remains unconfirmed in the reviewed replay guide.Checked 4 October 2026Make source ↗
Run a set of test inputs through an AI workflowCompare several test inputs before relying on the workflow. UnconfirmedAn equivalent built-in evaluation suite remains unconfirmed; agent chat is a different scope.Checked 4 October 2026Make source ↗
Score AI outputs across test runsTrack whether changed AI output meets the same checks. UnconfirmedAn equivalent metric-based evaluation tab remains unconfirmed.Checked 4 October 2026Make source ↗
Inspect an agent’s execution stepsSee the tool calls and token use behind a response. YesAgent output includes steps and token totals; available reasoning depends on model.Checked 4 October 2026Make source ↗
Retry automatically after a temporary failureAttempt failed work again under configured retry rules. YesIncomplete-execution retries depend on error type and scenario state; prior writes may repeat.Checked 4 October 2026Make source ↗
Review workflow activity over timeSee aggregate runs, failures and recorded usage. YesGrid and scenario history show activity and credit use.Checked 4 October 2026Make source ↗
Map dependencies across workflowsSee which automations depend on shared resources. YesGrid maps automations and resources; access and connected-provider scope matter.Checked 4 October 2026Make source ↗
Stream platform events to an external log serviceSend platform events to another monitoring system. UnconfirmedEquivalent configurable event streaming remains unconfirmed by the audit-log guide.Checked 4 October 2026Make source ↗
Inspect errors with an AI assistantAsk an assistant to explain a workflow problem. YesMaia reads run history and logs to explain or fix errors; public beta access has plan limits.Checked 4 October 2026Make source ↗
Delete saved execution dataRemove stored run data from the platform. UnconfirmedManage incomplete executions lets you delete stored incomplete records; full history-deletion scope is unconfirmed.Checked 4 October 2026Make source ↗
Set execution history retentionChoose how long execution data stays available. UnconfirmedRetention follows subscription terms; the reviewed history guide does not establish a buyer-set duration.Checked 4 October 2026Make source ↗
Inspect a step while testingSee a step’s input and output in the editor. YesView module run data shows bundles during testing.Checked 4 October 2026Make source ↗
Inspect worker statusSee the workers processing queued workflow runs. UnconfirmedThis exact operation remains unconfirmed. Related features don’t establish support.Checked 4 October 2026Make source ↗
Send execution traces to a monitoring serviceFollow a run and its steps in an external tracing tool. UnconfirmedThis exact operation remains unconfirmed. Related features don’t establish support.Checked 4 October 2026Make source ↗
Label execution traces for attributionAdd identifying information to a recorded execution trace. UnconfirmedThis exact operation remains unconfirmed. Related features don’t establish support.Checked 4 October 2026Make source ↗
Inspect workflow changesSee what changed between saved workflow versions. UnconfirmedThis exact operation remains unconfirmed. Related features don’t establish support.Checked 4 October 2026Make source ↗
Monitor changes in AI behaviorWatch for changes in AI outputs after deployment. UnconfirmedThis exact operation remains unconfirmed. Related features don’t establish support.Checked 4 October 2026Make source ↗
Track reported time savedRecord the time a workflow is expected to save. UnconfirmedThis exact operation remains unconfirmed. Related features don’t establish support.Checked 4 October 2026Make source ↗
Filter usage insightsNarrow the activity shown in the dashboard. UnconfirmedThis exact operation remains unconfirmed. Related features don’t establish support.Checked 4 October 2026Make source ↗
Export a workflow in its own product formatDownload a workflow definition in the product’s own format. YesExports Make blueprint JSON; imports don’t include working account connections.Checked 2 October 2026Make source ↗
Import a workflow in its own product formatLoad a workflow export back into the same product. YesImports Make blueprint JSON below 2 MB; recipient recreates connections.Checked 2 October 2026Make source ↗
Share a workflow with a colleagueGive a named colleague access to the same workflow. YesTeam-owned scenarios are accessible to team members; permissions and plan matter.Checked 2 October 2026Make source ↗
Share or reconnect app credentialsLet an authorized colleague use a shared connection or reconnect one after import. YesTeam members can use/manage team connections; blueprint import requires new connections.Checked 2 October 2026Make source ↗Make source 2 ↗
Assign different collaborator rolesGive collaborators different permissions to view or edit work. YesTeam role management requires Teams plan or higher.Checked 2 October 2026Make source ↗
Keep workflow versions in GitConnect workflow versions to a Git repository. UnconfirmedMake CLI may work alongside Git/CI/CD; direct built-in Git version control is unverified. The linked guide gives related context, not confirmation of the exact operation.Checked 2 October 2026Make source ↗
Add custom workflow propertiesAttach named metadata to a workflow. YesCustom scenario properties support field types and required values; organization setup is needed.Checked 4 October 2026Make source ↗
Share an agent with a colleagueLet another person reuse the same configured agent. UnconfirmedThe new-agent guide establishes creation and scenario tools, but does not confirm a colleague’s direct reuse of the same agent independently of scenario access.Checked 4 October 2026Make source ↗
Create a reusable workflow templatePackage a starting workflow so it can be loaded for another build. YesAnyone can create a scenario template on all plans; community-app templates cannot be publicly published.Checked 4 October 2026Make source ↗
Share a workflow templateLet another person load the same starting workflow. YesTeam templates can be shared outside the team through a public link; all plans support templates.Checked 4 October 2026Make source ↗
Manage a shared projectOrganize workflows and members in a shared project. YesTeams own scenarios, connections and data; multiple teams and role management require Teams or Enterprise.Checked 4 October 2026Make source ↗
Edit a workflow with colleaguesGive colleagues access to change the same workflow. YesTeam Member role can create, edit and delete team scenarios; this does not claim simultaneous editing.Checked 4 October 2026Make source ↗
Use stored app authenticationSave a connection for later workflow runs. YesConnections hold app authorization; app permissions still apply.Checked 4 October 2026Make source ↗
Connect an external secrets vaultLoad credential values from a managed secret store. UnconfirmedA native external-vault integration remains unconfirmed in the reviewed connection guide.Checked 4 October 2026Make source ↗
Define custom permission rolesChoose which actions a collaborator may perform. YesEnterprise-only custom organization/team roles; permissions apply to all resources of a type in a team.Checked 4 October 2026Make source ↗
Separate team resourcesLimit a colleague’s access to one team or project. YesTeam resources follow the organization/team role model.Checked 4 October 2026Make source ↗
Sign in through a company identity providerUse a company-managed login for platform access. YesSAML SSO is an Enterprise feature; identity-provider setup is required.Checked 4 October 2026Make source ↗
Sign in using LDAPUse a directory service for platform user login. UnconfirmedPlatform-login LDAP is unconfirmed; an LDAP app connector would be a different scope.Checked 4 October 2026Make source ↗
Use a second sign-in factorRequire a second code for a personal login. YesPersonal 2FA uses an authenticator; account recovery codes need safekeeping.Checked 4 October 2026Make source ↗
Enforce second factors for organization membersRequire members to use two-factor sign-in. YesEnterprise enforcement covers native email/password login; external providers enforce their own factors.Checked 4 October 2026Make source ↗
Inspect administrator activitySee who changed platform resources and when. YesEnterprise audit logs retain events for 12 months; data-store module actions are excluded.Checked 4 October 2026Make source ↗
Keep payloads out of saved run logsReduce sensitive data retained after a run. YesKeep data confidential removes retained payloads and limits repair options.Checked 4 October 2026Make source ↗
Authenticate an incoming webhookCheck who may call the workflow endpoint. YesWebhook authentication must be configured; exact requested scheme needs checking.Checked 4 October 2026Make source ↗
Provision new company usersAdd company members and assign access. YesEnterprise domain claim and APIs can provision members into teams.Checked 4 October 2026Make source ↗
Keep a personal workspace privateSeparate personal workflows from shared team work. YesPrivate spaces work on all plans; admin visibility requires Enterprise.Checked 4 October 2026Make source ↗
Limit a personal workspace’s usageSet an allowance for a member’s private work. YesAdmins can set monthly or yearly private-space credit limits.Checked 4 October 2026Make source ↗
Claim a company email domainDirect members using a company domain to its login. YesEnterprise domain claim routes new logins to the configured SSO.Checked 4 October 2026Make source ↗
Set a platform session timeoutLimit how long a signed-in session lasts. YesOrganization owners/admins can configure timeout; browser restore settings can affect it.Checked 4 October 2026Make source ↗
Map identity-provider groups to project rolesAssign platform access from company directory groups. UnconfirmedEquivalent automatic group-to-team-role mapping is unconfirmed in the reviewed SSO guide.Checked 4 October 2026Make source ↗
Enforce organization-wide sharing rulesControl which resources members may share. UnconfirmedThe custom-role guide documents permissions; equivalent organization-wide sharing policy is unconfirmed.Checked 4 October 2026Make source ↗
Require company SSO for hosted chatUse company identity to control a public chat entry. UnconfirmedA native public chat protected by company SSO remains unconfirmed.Checked 4 October 2026Make source ↗
Require company SSO for a hosted formUse company identity to control a workflow form. UnconfirmedEquivalent native hosted form and SSO gate remain unconfirmed.Checked 4 October 2026Make source ↗
Revoke a connected app’s authorizationRemove permission for a connected account. UnconfirmedCredential-request documentation discusses connection authorization; revocation in the external provider still needs checking.Checked 4 October 2026Make source ↗
Authenticate access to an agentControl who can call an agent entry point. UnconfirmedThis exact operation remains unconfirmed. Related features don’t establish support.Checked 4 October 2026Make source ↗
Separate stored data by projectKeep one project’s stored records apart from another’s. YesTeam-owned data stores follow team permissions; this is not per-record isolation.Checked 4 October 2026Make source ↗
Authenticate a platform API callProve which account may call the platform API. YesMake API requests use an API token; plan and endpoint permissions apply.Checked 4 October 2026Make source ↗
Assign administrator accessGive someone administrative control over platform resources. YesOrganization owners/admins manage teams; Team Admin controls team resources. Team-role management requires Teams or Enterprise.Checked 4 October 2026Make source ↗
Inject encrypted credential valuesPass protected credential values to a workflow step. UnconfirmedThis exact operation remains unconfirmed. Related features don’t establish support.Checked 4 October 2026Make source ↗
Use the vendor-hosted platformBuild and run workflows on infrastructure operated by the vendor. YesMake’s scenario service creates and runs scenarios on its documented production zones, including eu1.make.com; ordinary access is through the hosted service. Separate White Label access is not implied.Checked 2 October 2026Make source ↗
Run the workflow engine on your own infrastructureInstall and operate the workflow engine on infrastructure you control. UnconfirmedA separate White Label/OEM on-premise engine is documented, but availability under the ordinary Make subscription buyer's plan is unverified. The Enterprise on-prem agent connects a local network; it is not the workflow engine. The linked guide gives related context, not confirmation of the exact operation.Checked 2 October 2026Make source ↗Make source 2 ↗Make source 3 ↗
Connect a cloud workflow to a private networkLet a vendor-hosted workflow reach a service on a private network. YesEnterprise On-prem agent gives cloud scenarios access to local HTTP/API systems; target fit untested.Checked 2 October 2026Make source ↗
View billable usage for a runSee the chargeable usage units attributed to one saved run. YesCredit usage table shows credits per scenario run on all plans.Checked 2 October 2026Make source ↗
Run work in a worker queueSeparate receiving work from workers that execute it. UnconfirmedEquivalent subscriber-operated worker queues are unconfirmed on ordinary Make plans.Checked 4 October 2026Make source ↗
Store execution files in your own S3 bucketKeep execution files in separately managed object storage. UnconfirmedEquivalent native execution-storage configuration remains unconfirmed.Checked 4 October 2026Make source ↗
Import custom code librariesUse packages beyond the platform’s included libraries. YesMake Code custom imports require Enterprise; ordinary paid plans include selected libraries.Checked 4 October 2026Make source ↗
Run a shell command on the engine’s hostExecute a command where the workflow engine runs. UnconfirmedMake Code is a sandbox; direct access to the engine’s host shell is unconfirmed.Checked 4 October 2026Make source ↗
Read and write local engine filesUse files on the machine running the workflow engine. UnconfirmedDirect engine-filesystem access is unconfirmed on ordinary Make subscriptions.Checked 4 October 2026Make source ↗
Separate development and production environmentsMove approved changes between distinct environments. UnconfirmedEquivalent built-in environment promotion is unconfirmed; external CLI scripts are a separate route.Checked 4 October 2026Make source ↗
Run workflows concurrentlyProcess more than one workflow run at once. YesInstant webhooks run in parallel unless process-in-order is enabled.Checked 4 October 2026Make source ↗
Set a higher execution priorityMove selected work ahead in the execution queue. UnconfirmedEarlier Make pricing listed priority execution on Pro. Current access remains unconfirmed.Checked 4 October 2026Make source ↗
Inspect and modify the engine sourceRead or change the code running the platform. UnconfirmedOrdinary managed subscriptions do not establish source access; OEM terms remain unverified.Checked 4 October 2026Make source ↗
Install the engine with DockerRun the workflow service in a container. UnconfirmedOrdinary subscriptions do not establish container deployment; OEM access is separate and unverified.Checked 4 October 2026Make source ↗
Deploy the engine on KubernetesOperate the workflow service in a container cluster. UnconfirmedSeparate White Label Helm documentation exists; ordinary subscriber eligibility remains unverified.Checked 4 October 2026Make source ↗
Configure the engine with environment variablesSet instance options outside the workflow editor. UnconfirmedOrdinary cloud subscriptions do not establish engine configuration access.Checked 4 October 2026Make source ↗
Keep the engine running without internet accessRun workflows on an isolated network. UnconfirmedAn on-prem agent requires cloud connectivity; isolated OEM operation remains unverified.Checked 4 October 2026Make source ↗
Embed the workflow editor in your own productOffer the editor as part of another application. YesWhite Label/OEM route is documented separately; commercial eligibility and terms need checking.Checked 4 October 2026Make source ↗
Use multiple main instances for high availabilityKeep another main instance ready to serve work. UnconfirmedComparable buyer-controlled multi-main operation remains unconfirmed on ordinary subscriptions.Checked 4 October 2026Make source ↗
Back up a self-hosted instanceSave what is needed to restore the engine. UnconfirmedOrdinary Make subscription access does not establish subscriber-operated engine backups.Checked 4 October 2026Make source ↗
Start from a message queueReceive work from a configured queue service. UnconfirmedThe current RabbitMQ app guide lists Send to Exchange and Send to Queue, but does not establish a native receive-message trigger. This does not rule out other broker or custom-app routes.Checked 4 October 2026Make source ↗
Set policies across multiple automationsApply governance rules across multiple automations. UnconfirmedThe reviewed platform guides do not establish this exact operation. Verify the chosen component and access.Checked 4 October 2026Make source ↗
Promote changes to productionMove an approved workflow from development to production. UnconfirmedThis exact operation remains unconfirmed. Related features don’t establish support.Checked 4 October 2026Make source ↗
Compare development and production versionsInspect changes before moving them between environments. UnconfirmedThis exact operation remains unconfirmed. Related features don’t establish support.Checked 4 October 2026Make source ↗
Run an automated deployment pipelineApply workflow changes through a repeatable deployment job. YesMake CLI can be used with external Git and deployment scripts; it is not built-in environment promotion.Checked 4 October 2026Make source ↗
Deploy with a Helm chartInstall the engine from a Helm deployment package. UnconfirmedThis exact operation remains unconfirmed. Related features don’t establish support.Checked 4 October 2026Make source ↗
Set spending or usage limitsSet a boundary on the platform’s consumption. YesEnterprise admins can cap each team’s credits; reaching the limit pauses that team’s scenarios.Checked 4 October 2026Make source ↗

Checks from 4 October 2026. Source links appear where the record includes a vendor source.

The alternatives

Check the differences that matter to your workflow.

These counts keep the original comparison dates. Later findings appear beside earlier answers. An unconfirmed answer doesn’t mean a product lacks the capability.

01 / COMPARISON DATED 26 September 2026

Make vs Zapier

You need a less-common app connection. Check Zapier’s catalog and the exact action before changing a working setup.

9Zapier had these capabilities; Make had a no or an unconfirmed answer0 confirmed differences · 9 unconfirmed answers
6Make had these capabilities; Zapier had a no or an unconfirmed answer2 confirmed differences · 4 unconfirmed answers

Zapier had these at the check

  • Choose a shorter automatic history retention periodZapier: Yes · Make: UnconfirmedZapier: Enterprise admins can set Zap history retention to 7–30 days. Deletion is permanent and applies account-wide. Make: History duration depends on plan. A comparable configurable account-wide deletion period was not confirmed.
  • Require approval before publishing workflow changesZapier: Yes · Make: UnconfirmedZapier: Enterprise administrators can require approval before a Zap is published. This is publishing control, not an approval step inside a running workflow. Make: A comparable built-in publish-approval gate was not confirmed. Team roles alone do not establish this scope.
  • Restrict the apps people can connectZapier: Yes · Make: UnconfirmedZapier: Enterprise app policies control apps and actions. Managed apps can limit who creates or shares connections. Make: A same-scope organization policy for allowed apps and actions was not confirmed.
  • Create separate workspaces in an organizationZapier: Yes · Make: UnconfirmedZapier: Owners and Super admins can create organization workspaces. Each has its own members, assets and task budget. Minimum plan and organization entitlement remain unconfirmed. Make: A comparable organization-workspace creation scope remains unconfirmed. Shared workflow access alone does not establish separate workspaces.
  • Build a form inside the platformZapier: Yes · Make: UnconfirmedZapier Forms provides a form builder. Form features and limits depend on the plan. Make: A comparable native form builder was not confirmed. Connecting a third-party form is a different scope.
  • Show form fields according to earlier answersZapier: Yes · Make: UnconfirmedZapier: Paid Professional or higher; up to five conditions. This feature is not available on the Pro trial. Make: Native form-field logic was not confirmed. Scenario filters act after data arrives and do not establish this feature.
  • Collect uploaded files through a native formZapier: Yes · Make: UnconfirmedZapier: Free allows files up to 5 MB; paid plans up to 25 MB. File type and count controls are available. Make: Native form uploads were not confirmed. A connected form provider has its own limits.
  • Embed a native form in another websiteZapier: Yes · Make: UnconfirmedZapier: Paid plans support script or iframe embeds. Forms with restricted access cannot be embedded. Make: A comparable native form embed was not confirmed.
  • Restrict who can open a native formZapier: Yes · Make: UnconfirmedZapier: Paid plans support restricted access. Professional supports magic-link login; Team and Enterprise also support Google login. Make: A comparable native form access system was not confirmed. Connected form services must be checked separately.

Make had these at the check

  • Use conditional branches on the Free planZapier: No · Make: YesMake: Free includes routers and filters. Usage and active-scenario limits still apply.Zapier: Paths is unavailable on Free. A paid plan removes this restriction.Zapier Paths requires a paid plan. Make Free includes routers and filters, subject to usage and active-scenario limits.
  • Create reusable custom JavaScript functionsZapier: Unconfirmed · Make: YesZapier: A comparable reusable custom-function library remains unconfirmed. A Code step alone does not establish this scope. Make: The organization license must enable customFunctions. The documented management API uses an API token. Minimum named plan and authoring role remain unconfirmed. Some JavaScript features are unsupported.
  • Create a role with selected permissionsZapier: Unconfirmed · Make: YesZapier: Default account and product roles are documented; a comparable custom-role builder was not confirmed. Make: Enterprise only, excluding private instances. Permissions apply to all resources of a type within a team; use separate teams for narrower boundaries.
  • Connect to a private system through an installed agentZapier: Unconfirmed · Make: YesZapier: A comparable installed-agent route was not confirmed. Public API or static-IP access does not establish this scope. Make: Enterprise only. The on-premise agent currently supports the HTTP Agent application; it is not general self-hosting.
  • Wait for one webhook run to finish before starting the nextZapier: Unconfirmed · Make: YesZapier: Delay After Queue spaces out actions but its documentation warns that replays or infrastructure delays can cause overlap. Strict whole-run serialization was not confirmed. Make: Process data in order waits for the prior run to finish. Instant webhook runs otherwise execute in parallel.
  • Customize the response to an incoming webhookZapier: No · Make: YesZapier: Catch Hook and Catch Raw Hook responses cannot be customized. This mark applies to these built-in triggers, not every possible custom integration. Make: The Webhook Response module sets a custom status, headers and response body.
See the dated evidence ↗
02 / COMPARISON DATED 4 October 2026

Make vs n8n

You need your own server and code access. Check n8n’s license, plan limits and the upkeep you’ll take on.

42n8n had these capabilities; Make had a no or an unconfirmed answer0 confirmed differences · 42 unconfirmed answers
15Make had these capabilities; n8n had a no or an unconfirmed answer0 confirmed differences · 15 unconfirmed answers

n8n had these at the check

  • Start a workflow from a formn8n: Yes · Make: UnconfirmedMake: Connected form apps can trigger scenarios; an equivalent native form builder is unconfirmed here.n8n: Form Trigger serves a form and begins a workflow; configure fields and access.
  • Import a cURL requestn8n: Yes · Make: UnconfirmedMake: The HTTP guide reviewed previously does not establish this exact import action.n8n: HTTP Request supports importing cURL; credentials and unsupported options still need checking.
  • Paginate an API responsen8n: Yes · Make: UnconfirmedMake: Exact automatic pagination setup remains unconfirmed in the reviewed HTTP reference.n8n: HTTP Request supports pagination modes and expressions.
  • Change stored data in bulkn8n: Yes · Make: UnconfirmedMake: Data stores support record operations, but a comparable multi-record update/delete operation is not established in this guide; iteration is a separate route.n8n: Data Table node updates or deletes one or more rows using matching conditions; storage and project limits apply.
  • Call an agent from another agentn8n: Yes · Make: UnconfirmedMake: The reviewed agent reference does not establish this exact nested-agent configuration.n8n: Human-review documentation covers an AI Agent used as another agent’s tool.
  • Search a vector databasen8n: Yes · Make: UnconfirmedMake: Dedicated vector-store control is unconfirmed by the reviewed knowledge-file guide.n8n: Vector stores use embeddings, retrievers and document loaders; database setup remains separate.
  • Run a model on your own infrastructuren8n: Yes · Make: UnconfirmedMake: Exact local-model access through the managed agent provider remains unconfirmed.n8n: Self-hosted AI starter kit includes local Ollama; it is for testing and needs hardening before production.
  • Provide a hosted chat entry pointn8n: Yes · Make: UnconfirmedMake: The reviewed test-chat reference does not establish an equivalent public hosted chat page.n8n: Chat Trigger offers hosted or embedded chat; each message uses an execution.
  • Require approval before an agent tool runsn8n: Yes · Make: UnconfirmedMake: The reviewed agent reference does not establish an equivalent native approval gate.n8n: Human review can guard selected agent tools through chat and connected approval channels.
  • Choose between an AI chain and an agentn8n: Yes · Make: UnconfirmedMake: The reviewed platform guides do not establish this exact operation. Verify the chosen component and access.n8n: Chains follow predefined calls; the Agent node chooses tools. These are different workflow components.
  • Process unstructured contentn8n: Yes · Make: UnconfirmedMake: This exact operation remains unconfirmed. Related features don’t establish support.n8n: Agent knowledge bases accept uploaded files to search and read; preview and self-hosted sandbox requirements apply.
  • Ingest and index reference datan8n: Yes · Make: UnconfirmedMake: This exact operation remains unconfirmed. Related features don’t establish support.n8n: Default Data Loader loads binary/JSON documents and splits text for a connected vector store; embedding and store setup are separate.
  • Extract structured fields from textn8n: Yes · Make: UnconfirmedMake: This exact operation remains unconfirmed. Related features don’t establish support.n8n: Information Extractor accepts text and attribute descriptions, a JSON example or JSON Schema; schema format does not prove factual accuracy.
  • Create mock data for a testn8n: Yes · Make: UnconfirmedMake: A dedicated mocking control remains unconfirmed in the reviewed replay guide.n8n: Code, Edit Fields and sample sets of test inputs can create development test data.
  • Run a set of test inputs through an AI workflown8n: Yes · Make: UnconfirmedMake: An equivalent built-in evaluation suite remains unconfirmed; agent chat is a different scope.n8n: Light evaluations use data tables or Google Sheets; Cloud all plans and registered Community are documented.
  • Score AI outputs across test runsn8n: Yes · Make: UnconfirmedMake: An equivalent metric-based evaluation tab remains unconfirmed.n8n: Metric evaluations score outputs; plan gates and a single-workflow allowance apply.
  • Stream platform events to an external log servicen8n: Yes · Make: UnconfirmedMake: Equivalent configurable event streaming remains unconfirmed by the audit-log guide.n8n: Log streaming forwards selected events; Enterprise access and destination setup apply.
  • Delete saved execution datan8n: Yes · Make: UnconfirmedMake: Manage incomplete executions lets you delete stored incomplete records; full history-deletion scope is unconfirmed.n8n: Execution deletion is documented; saved-data settings and retention still matter.
  • Set execution history retentionn8n: Yes · Make: UnconfirmedMake: Retention follows subscription terms; the reviewed history guide does not establish a buyer-set duration.n8n: Self-hosted pruning settings control saved execution data; Cloud plan retention differs. Cloud Enterprise also lists extended retention; a plan entitlement differs from self-hosted configuration.
  • Send execution traces to a monitoring servicen8n: Yes · Make: UnconfirmedMake: This exact operation remains unconfirmed. Related features don’t establish support.n8n: Self-hosted OpenTelemetry sends workflow and node traces to a collector; configuration is required.
  • Inspect workflow changesn8n: Yes · Make: UnconfirmedMake: This exact operation remains unconfirmed. Related features don’t establish support.n8n: Workflow history shows saved versions; accessible history and differences depend on plan.
  • Track reported time savedn8n: Yes · Make: UnconfirmedMake: This exact operation remains unconfirmed. Related features don’t establish support.n8n: Insights uses your fixed estimate or Time Saved nodes, not an independent measurement; parent workflows only.
  • Filter usage insightsn8n: Yes · Make: UnconfirmedMake: This exact operation remains unconfirmed. Related features don’t establish support.n8n: Insights dashboard supports plan-dependent time periods and per-workflow metrics; Cloud Pro/Enterprise or self-hosted Business/Enterprise.
  • Keep workflow versions in Gitn8n: Yes · Make: UnconfirmedMake CLI may work alongside Git/CI/CD; direct built-in Git version control is unverified. The linked guide gives related context, not confirmation of the exact operation.n8n: Git source control requires Business/Enterprise and administrator setup.
  • Share an agent with a colleaguen8n: Yes · Make: UnconfirmedMake: The new-agent guide establishes creation and scenario tools, but does not confirm a colleague’s direct reuse of the same agent independently of scenario access.n8n: Published agents are project artifacts available to users and callable from that project’s workflows or sub-agents. Preview: Cloud all plans; self-hosted requires setup and excludes Enterprise. This does not promise a public share link.
  • Connect an external secrets vaultn8n: Yes · Make: UnconfirmedMake: A native external-vault integration remains unconfirmed in the reviewed connection guide.n8n: External secret stores are Enterprise features; values resolve in credential fields.
  • Sign in using LDAPn8n: Yes · Make: UnconfirmedMake: Platform-login LDAP is unconfirmed; an LDAP app connector would be a different scope.n8n: Cloud Enterprise or self-hosted Business/Enterprise; instance owner and an existing LDAP server are required.
  • Run the workflow engine on your own infrastructuren8n: Yes · Make: UnconfirmedMake: A separate White Label/OEM on-premise engine is documented, but availability under the ordinary Make subscription buyer's plan is unverified. The Enterprise on-prem agent connects a local network; it is not the workflow engine. The linked guide gives related context, not confirmation of the exact operation.n8n: Community/paid self-hosted editions run on buyer infrastructure; operation and feature gates remain.
  • Run work in a worker queuen8n: Yes · Make: UnconfirmedMake: Equivalent subscriber-operated worker queues are unconfirmed on ordinary Make plans.n8n: Self-hosted queue mode uses Redis and workers; shared encryption and database setup are required.
  • Store execution files in your own S3 bucketn8n: Yes · Make: UnconfirmedMake: Equivalent native execution-storage configuration remains unconfirmed.n8n: External binary/execution storage requires self-hosted Business or Enterprise; Cloud is excluded.
  • Run a shell command on the engine’s hostn8n: Yes · Make: UnconfirmedMake Code is a sandbox; direct access to the engine’s host shell is unconfirmed.n8n: Execute Command is self-hosted only and disabled by default from n8n 2.0.
  • Read and write local engine filesn8n: Yes · Make: UnconfirmedMake: Direct engine-filesystem access is unconfirmed on ordinary Make subscriptions.n8n: Cloud can access only /home/node/ and files are temporary. Self-hosted access follows process permissions and configured path restrictions.
  • Separate development and production environmentsn8n: Yes · Make: UnconfirmedMake: Equivalent built-in environment promotion is unconfirmed; external CLI scripts are a separate route.n8n: Git-based environments need Business or Enterprise and administrator setup.
  • Inspect and modify the engine sourcen8n: Yes · Make: UnconfirmedMake: Ordinary managed subscriptions do not establish source access; OEM terms remain unverified.n8n: Self-hosted source is available under n8n’s license; commercial reuse has conditions.
  • Install the engine with Dockern8n: Yes · Make: UnconfirmedMake: Ordinary subscriptions do not establish container deployment; OEM access is separate and unverified.n8n: Docker and Docker Compose are documented self-hosted options; server upkeep remains yours.
  • Deploy the engine on Kubernetesn8n: Yes · Make: UnconfirmedMake: Separate White Label Helm documentation exists; ordinary subscriber eligibility remains unverified.n8n: A Google Kubernetes deployment guide is documented; cluster operation and feature gates remain.
  • Configure the engine with environment variablesn8n: Yes · Make: UnconfirmedMake: Ordinary cloud subscriptions do not establish engine configuration access.n8n: Self-hosted instances expose documented environment-variable settings.
  • Use multiple main instances for high availabilityn8n: Yes · Make: UnconfirmedMake: Comparable buyer-controlled multi-main operation remains unconfirmed on ordinary subscriptions.n8n: Queue mode documents multi-main setup as a self-hosted Enterprise feature.
  • Back up a self-hosted instancen8n: Yes · Make: UnconfirmedMake: Ordinary Make subscription access does not establish subscriber-operated engine backups.n8n: Self-hosted backup/restore guide distinguishes CLI exports from a full instance backup.
  • Start from a message queuen8n: Yes · Make: UnconfirmedMake: The current RabbitMQ app guide lists Send to Exchange and Send to Queue, but does not establish a native receive-message trigger. This does not rule out other broker or custom-app routes.n8n: Built-in RabbitMQ Trigger uses a connected broker and RabbitMQ credentials; this external event source is separate from n8n’s internal worker queue.
  • Promote changes to productionn8n: Yes · Make: UnconfirmedMake: This exact operation remains unconfirmed. Related features don’t establish support.n8n: Git-based promotion between instances requires source-control access and administrator setup.
  • Compare development and production versionsn8n: Yes · Make: UnconfirmedMake: This exact operation remains unconfirmed. Related features don’t establish support.n8n: Workflow diffs compare local and remote versions; source-control and plan access apply.

Make had these at the check

  • Validate incoming workflow fieldsn8n: Unconfirmed · Make: YesMake: Scenario inputs can be required and typed; Make validates collection fields against their specification. This does not validate every external payload automatically.n8n: The reviewed input-definition guide does not establish automatic type and required-field validation for every incoming workflow payload; add explicit checks for the particular route.
  • Create reusable custom functionsn8n: Unconfirmed · Make: YesMake: Enterprise JavaScript custom functions extend mapping.n8n: Equivalent instance-wide custom-function registration is unconfirmed; reusable workflows and Code are different scopes.
  • Extract file metadatan8n: Unconfirmed · Make: YesMake: Knowledge > Get knowledge returns name, description, date and ID for a selected knowledge file. This does not promise arbitrary file-format metadata extraction.n8n: The checked file and knowledge guides do not establish the same stored-knowledge metadata operation; file payload handling is separate.
  • Search the web through an AI tooln8n: Unconfirmed · Make: YesMake AI Web Search exposes a search module; model cost and accuracy remain separate.n8n: Equivalent native AI web search is unconfirmed in the reviewed AI-component overview; connected search services are a different route.
  • Replay a completed run from its triggern8n: Unconfirmed · Make: YesMake: All plans can replay old trigger data through every module of the current scenario, using current variables and consuming credits; retention and repeated writes matter. Poll checks with no data and isolated non-trigger module runs are excluded.n8n: Reviewed docs show copying successful execution data to the editor for a manual rerun, not the exact production replay operation. The linked guide gives related context, not confirmation of the exact operation.Make reruns all current modules and uses credits. The exact n8n production replay operation is unverified.
  • Map dependencies across workflowsn8n: Unconfirmed · Make: YesMake: Grid maps automations and resources; access and connected-provider scope matter.n8n: An equivalent dependency map across workflows remains unconfirmed in the reviewed Insights guide.
  • Add custom workflow propertiesn8n: Unconfirmed · Make: YesMake: Custom scenario properties support field types and required values; organization setup is needed.n8n: An equivalent arbitrary typed workflow-property control remains unconfirmed; execution custom data is a different scope.
  • Enforce second factors for organization membersn8n: Unconfirmed · Make: YesMake: Enterprise enforcement covers native email/password login; external providers enforce their own factors.n8n: The reviewed guide covers an individual account’s second factor. Equivalent organization-wide enforcement remains unconfirmed.
  • Keep a personal workspace privaten8n: Unconfirmed · Make: YesMake: Private spaces work on all plans; admin visibility requires Enterprise.n8n: The credential guide establishes personal credential placement, not privacy of every workflow. Workflow privacy for this exact space is not confirmed.n8n’s shared projects and access vary by edition.
  • Limit a personal workspace’s usagen8n: Unconfirmed · Make: YesMake: Admins can set monthly or yearly private-space credit limits.n8n: An equivalent per-member execution allowance is unconfirmed in the reviewed pricing list.
  • Claim a company email domainn8n: Unconfirmed · Make: YesMake: Enterprise domain claim routes new logins to the configured SSO.n8n: Equivalent domain claim behavior remains unconfirmed in the reviewed SAML guide.
  • Set a platform session timeoutn8n: Unconfirmed · Make: YesMake: Organization owners/admins can configure timeout; browser restore settings can affect it.n8n: The exact comparable platform control remains unconfirmed in the reviewed user settings.
  • Connect a cloud workflow to a private networkn8n: Unconfirmed · Make: YesMake: Enterprise On-prem agent gives cloud scenarios access to local HTTP/API systems; target fit untested.n8n: Postgres credential docs show an app-specific SSH tunnel, while self-hosting is a different route; a general n8n Cloud-to-private-network link is unverified. The linked guide gives related context, not confirmation of the exact operation.Make’s on-prem agent is Enterprise-only. A general n8n Cloud private-network route is unverified.
  • View billable usage for a runn8n: Unconfirmed · Make: YesMake: Credit usage table shows credits per scenario run on all plans.n8n: Run lists and aggregate quota rules exist, but no per-run billable flag was found; vendor docs conflict on failed-run quota treatment. The linked guide gives related context, not confirmation of the exact operation.
  • Set spending or usage limitsn8n: Unconfirmed · Make: YesMake: Enterprise admins can cap each team’s credits; reaching the limit pauses that team’s scenarios.n8n: This exact operation remains unconfirmed. Related features don’t establish support.

Still unconfirmed for Make or n8n

  • Rerank retrieved search resultsn8n: Unconfirmed · Make: UnconfirmedMake: The reviewed knowledge guide does not establish a native reranking control.n8n: The vector-retrieval overview does not establish the exact reranking action; verify the chosen node.
  • Ground answers with a knowledge graphn8n: Unconfirmed · Make: UnconfirmedMake: The reviewed knowledge-file guide does not establish a native graph-grounding control.n8n: The vector-store overview does not establish the graph-grounding claim in the check.
  • Filter retrieved documents by metadatan8n: Unconfirmed · Make: UnconfirmedMake: The reviewed knowledge guide does not establish arbitrary metadata filters.n8n: The retrieval overview does not establish every store’s metadata-filter options; check the selected node.
  • Combine keyword and meaning-based searchn8n: Unconfirmed · Make: UnconfirmedMake: Native hybrid search is unconfirmed in the reviewed knowledge guide.n8n: A generic vector-store overview does not establish a native combined search operation.
  • Inspect AI costs by modeln8n: Unconfirmed · Make: UnconfirmedMake: Agent token totals do not establish a monetary cost breakdown by model.n8n: Execution and token outputs do not establish the check’s exact model-cost analytics claim.
  • Inspect worker statusn8n: Unconfirmed · Make: UnconfirmedMake: This exact operation remains unconfirmed. Related features don’t establish support.n8n: The exact worker-status view remains unconfirmed in the reviewed queue guide.
  • Label execution traces for attributionn8n: Unconfirmed · Make: UnconfirmedMake: This exact operation remains unconfirmed. Related features don’t establish support.n8n: This exact operation remains unconfirmed. Related features don’t establish support.
  • Monitor changes in AI behaviorn8n: Unconfirmed · Make: UnconfirmedMake: This exact operation remains unconfirmed. Related features don’t establish support.n8n: This exact operation remains unconfirmed. Related features don’t establish support.
  • Map identity-provider groups to project rolesn8n: Unconfirmed · Make: UnconfirmedMake: Equivalent automatic group-to-team-role mapping is unconfirmed in the reviewed SSO guide.n8n: The SAML guide alone does not establish this exact mapping operation.
  • Enforce organization-wide sharing rulesn8n: Unconfirmed · Make: UnconfirmedMake: The custom-role guide documents permissions; equivalent organization-wide sharing policy is unconfirmed.n8n: Project roles establish access boundaries; the exact broader sharing policy remains unconfirmed.
  • Require company SSO for hosted chatn8n: Unconfirmed · Make: UnconfirmedMake: A native public chat protected by company SSO remains unconfirmed.n8n: Chat Trigger has n8n user authentication; the exact company-SSO deployment needs checking.
  • Require company SSO for a hosted formn8n: Unconfirmed · Make: UnconfirmedMake: Equivalent native hosted form and SSO gate remain unconfirmed.n8n: Form Trigger documentation does not establish the exact company-SSO claim in the check.
  • Revoke a connected app’s authorizationn8n: Unconfirmed · Make: UnconfirmedMake: Credential-request documentation discusses connection authorization; revocation in the external provider still needs checking.n8n: Credential removal and provider grant revocation are different operations; provider-side behavior remains unconfirmed here.
  • Authenticate access to an agentn8n: Unconfirmed · Make: UnconfirmedMake: This exact operation remains unconfirmed. Related features don’t establish support.n8n: This exact operation remains unconfirmed. Related features don’t establish support.
  • Inject encrypted credential valuesn8n: Unconfirmed · Make: UnconfirmedMake: This exact operation remains unconfirmed. Related features don’t establish support.n8n: This exact operation remains unconfirmed. Related features don’t establish support.
  • Set a higher execution priorityn8n: Unconfirmed · Make: UnconfirmedMake: Earlier Make pricing listed priority execution on Pro. Current access remains unconfirmed.n8n: Equivalent per-workflow priority control is unconfirmed in the reviewed queue documentation.
  • Keep the engine running without internet accessn8n: Unconfirmed · Make: UnconfirmedMake: An on-prem agent requires cloud connectivity; isolated OEM operation remains unverified.n8n: A VPN or self-hosted install does not establish fully disconnected licensing and operation here.
  • Set policies across multiple automationsn8n: Unconfirmed · Make: UnconfirmedMake: The reviewed platform guides do not establish this exact operation. Verify the chosen component and access.n8n: The reviewed platform guides do not establish this exact operation. Verify the chosen component and access.
  • Deploy with a Helm chartn8n: Unconfirmed · Make: UnconfirmedMake: This exact operation remains unconfirmed. Related features don’t establish support.n8n: This exact operation remains unconfirmed. Related features don’t establish support.
See the dated evidence ↗

133 comparison checks confirmed the capability in Make and the named competitor. These comparison checks recorded 2 confirmed differences and 89 entries with an unconfirmed answer. The same capability can appear in different dated comparisons.

Costs & catches

The plan price is the starting point.

These Make prices were checked on 25 September 2026. Confirm the current plans before paying. The prices cover Free and selected Core allowances. They don’t cover every plan, extra usage, taxes or your final bill.

Check current pricing ↗
Free · 1,000 credits$0per month
Core · 10,000 credits$10.59per month
Core · 20,000 credits$18.82per month
Core · 40,000 credits$34.12per month
Core · 80,000 credits$64.71per month
Core · 150,000 credits$116.47per month

Make’s first paid plan over time.

The dates below keep each price check. Only USD prices per month, paid monthly, are comparable.

  1. $10.59Core · 10,000 credits / month
  2. Not comparablePer month, paid yearly · currency unverified
The later check used different terms.

The later check showed a per-month price paid yearly with no confirmed currency, so it can’t show whether the earlier USD monthly price changed.

See Make’s pricing terms

What changed

Make’s earlier answers are not directly matched.

Different published labels are not treated as equivalent capability checks.

  1. No matching earlier capability answers

    Not compared

    The published labels don’t establish a like-for-like earlier check. They don’t establish that the capability answers stayed the same.

Skipped orders still used credits. A September order-list replay used 80 Core credits without saving new rows. Check your own workload’s credits before choosing a tier.

The research behind the review

Know what the answer rests on.

Where do these findings come from?

This Make review draws on Make vs Zapier, Make vs n8n. Individual check dates are shown only where the source supplies them. Undated answers say the check date is unavailable.

What was demonstrated, and what was documented?

The clip shows the request workflow and Make’s run history. The comparison confirms the saved results. A later automatic retry saved one request twice after a lost reply. Other capabilities come from documentation, not this clip.

What would still need checking for your workflow?

Your app permissions, failure handling, repeated saves and workload costs still need checking. Some older capability checks have no recorded date. This fictional job doesn’t establish general reliability.

How does this publication earn money?

This review contains affiliate links. We may earn a commission when you buy, at no extra cost to you. No vendor saw this page before publication.

Your next step

Try the job your workflow must do.

Start with your essential app action and a small input. Check saved results, repeated saves and credits before relying on the workflow.

Try Make

Make · Request queue · 2 October 2026

Make’s request workflow and run history. The comparison confirms six saved requests and two other outcomes. A successful run alone doesn’t confirm the results were saved or that retries avoided duplicates.