FunnelFlux Pro API DocumentsFunnelFlux Pro API Documents
  • Changelog
  • Status
  • Dashboard
  • Documentation
  • Authentication
  • Domains
  • Assets
  • Reporting
Getting Started
    General InformationAPI Best PracticesAuthenticationGenerating Asset IDsCreating Funnels
Feature Information
    Data MappingLogic ScriptsMiddleware NodesMiddleware TemplatesWebhook Payloads
Other
    Changelog
Other

Changelog

16/11/2026

  • Added customer API keys to the Auth API under a new API Keys tag: POST /api-keys (apiKeyCreate) creates a key and returns its plaintext secret once (APIKeyIssued, Cache-Control: no-store), GET /api-keys (apiKeyList) lists the account's active keys oldest first (APIKeyList), and DELETE /api-keys/{id} (apiKeyRevoke) revokes one permanently. FunnelFlux API keys carry the ffk_ prefix and are sent as Authorization: Bearer <secret>; only the SHA-256 hash is stored, so every read after creation carries masked instead of the secret. See the Authentication guide.
  • [Breaking-ish] API key management accepts a web-app session credential only. An API key or a legacy Auth0 token is refused with 403 session_required (new SessionRequired response) before anything is read or written; a session whose Clerk user carries no FunnelFlux account identifier answers account_not_linked instead. POST /api-keys also refuses a blank or over-255-character name with 400 invalid_name, the 51st active key with 409 api_key_limit_reached, and answers 503 with issuing_login_unavailable, issuing_login_lookup_failed or api_key_unavailable (APIKeyError). DELETE /api-keys/{id} answers 404 not_found for an unknown or already revoked key and 409 ambiguous_api_key when the id matches more than one row.
  • POST /auth/reset_password operation id is now authResetPassword (was authResetPasswordRetired). Behaviour is unchanged: Auth0 provider only, 410 endpoint_retired under Clerk, 503 account_changes_frozen while the sign-in provider cutover freeze is on.
  • [Breaking-ish] POST /trafficsource/data-mapping/clear/ (clearTrafficSourceDataMapping) accepts an omitted or empty field again: that clears every manual name of the one traffic source across campaign, field1 and field2. This restores the clear-all behavior the 13/11/2026 entry said was removed; send field to clear a single field. DataMappingClearRequest.field is optional again, and its enum now also allows ''.
  • [Breaking-ish] DataMappingClearRequest now sets additionalProperties: false: the body may carry only idTrafficSource and field, each spelled exactly so and at most once. Unknown or differently-cased keys, and a null field, are refused with 400 invalid_request and nothing is cleared — omit field to clear all fields.
  • Added PUT /funnel/move/ (moveFunnelsToFunnelGroup) to move funnels into a funnel group. idFunnelGroup is the canonical target; idCampaign remains a temporary alias and is ignored when idFunnelGroup is set; with neither, the request answers 400 "Missing idFunnelGroup". PUT /campaign/funnel/move/ (moveFunnels) is now documented as deprecated in favour of the new path and accepts the same input.
  • Outbound webhook payloads now carry schema_version "1.1" (was "1.0"). Version 1.1 is the version that always includes the top-level referring_page_id and referring_page_name fields; no other field changes. The webhook JSON Schema accepts "1.0" and "1.1" and requires the two fields only for "1.1". If your receiver checks schema_version for an exact "1.0", accept "1.1" as well. Middleware payloads keep their own version.
  • The webhook hit object no longer documents time_on_page; webhooks never sent it.
  • ConversionsUpdateParams.postbackCalls no longer declares default: none. Omitting postbackCalls does not stop delivery on conversion or custom-event uploads; only none or disablePostbacks: true does. onlyOnce and all behave the same as omitting it.

14/11/2026

  • Integration ad-account and conversion-goal reads now document provider authentication failures in IntegrationUnauthorized: the 401 response can contain a provider error message without error_description. Gateway authentication failures retain the existing Unauthorized bodies.

13/11/2026

  • [Breaking] POST /trafficsource/data-mapping/clear/ now requires field (campaign, field1, or field2) and clears exactly that field. Omitting or sending an empty field returns 400 invalid_request; the former clear-all-fields behavior is removed.
  • Data-mapping upload and clear operations moved from the Traffic Sources tag to the new Data Mapping tag in the Assets API.
  • Added the Data Mapping guide, covering field declarations, uploads, clears, validation, and report name resolution.

12/11/2026

  • Requests authenticated with a customer API key or a legacy Auth0 API token now share one per-account, per-minute cap (default 60). When it is exhausted, the API gateway answers HTTP 429 with a RateLimitError body (code API_RATE_LIMITED) and a Retry-After header, and the operation is not performed. Web-app session traffic is not limited. Every operation in the Assets, Reporting, Billing and Auth specs now documents the 429 response; batch reads (byIds, list endpoints) instead of looping single-item reads.
  • Auth API operations now document a shared 401 body (access_token_required, no_access, or invalid_token for an invalid or expired token or API key) and a 403 account_not_linked response for a valid token or API key whose Clerk user has no FunnelFlux account. No account is resolved or created in that case.
  • POST /auth/register (authRegister) is now documented. It works under the Auth0 provider only and can answer 409 registration_in_progress or 503 account_changes_frozen / registration_unavailable, with Retry-After. Under the Clerk provider it answers 410 endpoint_retired with replacement: checkout; new accounts are created only through checkout.
  • [Breaking] Under the Clerk sign-in provider, password login (/auth/login), the MFA OTP verification endpoint (/auth/mfa/otp), and the MFA status, enrollment and authentication-method delete endpoints answer 410 with EndpointRetiredError (error endpoint_retired, replacement sign_in) and Cache-Control: no-store, before any credential or body is read. They keep working under the Auth0 provider.
  • Added PUT /trafficsource/data-mapping/upload/ (uploadTrafficSourceDataMapping, tag Traffic Sources): merges ID-to-name rows into a traffic source's manual names for campaign, field1 or field2. Request DataMappingUploadRequest; response DataMappingUploadResponse with added and updated counts. At most 10,000 rows and 4 MiB per request. The upload is all-or-nothing: if any row is refused, nothing is stored and rowErrors lists every refused row with its DataMappingRowError.reason. Errors use DataMappingError codes (invalid_request, no_rows, too_many_rows, invalid_rows, not_found, dictionary_sync_enabled, field_not_declared, body_too_large).
  • Added POST /trafficsource/data-mapping/clear/ (clearTrafficSourceDataMapping, tag Traffic Sources): removes a traffic source's manual names for one field, or all three fields when field is omitted or empty, and returns the removed count. Names synced from an integration are never touched, and the field does not need to be declared.
  • TrafficSource has a new dataMapping object (TrafficSourceDataMapping: campaign, field1Slot, field2Slot, suffixIds) that declares which fields carry manual name mappings. Declaring any field while integration.features.dictionarySync.enabled is on is rejected with 400 dictionary_sync_conflict.
  • [Breaking-ish] TrafficSource.conversionTrackingMode (classic or integrations) is now a top-level field, and the mode is no longer read from meta. When the field is omitted on save, the stored mode is classic. Conversions route through integrations only when the mode is integrations, a connection is set, and integration.features.capi.enabled is true. Saving one setting never changes the other. A save with integrations removes the stored classic postbacks and conversionTrackingSettings, other values are rejected with 400, and unlinking the integration sets classic.
  • Page now has a read-only owner field, filled by loading endpoints and ignored on save.
  • [Breaking-ish] PUT /reporting/update/conversions/ now documents re-upload rules for the same hit and transaction ID. A re-upload within 5 minutes is ignored as a duplicate. Between 5 minutes and 3 hours, a revenue change replaces the revenue without calling postback URLs or webhooks again unless forcePostbacks is true. After 3 hours, the upload replaces the conversion and is delivered again. Negative revenue deletes the conversion at any time. disablePostbacks: true or postbackCalls: none stops all outbound delivery and overrides forcePostbacks. postbackCalls values onlyOnce and all are accepted but have no effect, so onlyOnce no longer prevents a re-upload from being delivered again.
  • PUT /reporting/update/customevents/ now documents that custom events are not deduplicated. Each accepted upload goes to postback URLs, webhooks, Meta and TikTok again, while Google Ads and Microsoft Ads receive each hit and event number only once. Set disablePostbacks: true or postbackCalls: none to record events without sending them. Negative revenue is rejected.
  • [Breaking-ish] PUT /reporting/update/cost/, POST /reporting/update/reset/ and DELETE /reporting/update/reset/ now require nonzero timeStart and timeEnd, with timeEnd strictly after timeStart, timeStart on or after 2020-01-01T00:00:00Z, and timeEnd no later than 12 calendar months after timeStart. Other requests are rejected with 400 before anything is counted, changed or deleted. For reset requests, a missing restriction is reported before time-window errors.
  • Cost update and reset requests now reject a restrictToTrackingFields key whose value list is empty or contains only blank values (400). Omit the key to leave tracking fields unrestricted.
  • DrilldownReportParams.restrictToMetrics: when it and calculatedMetrics are both empty, the default metric set excludes CustomEvent1-10, AverageTimeOnPage and BounceRate. Unsupported names are ignored, and a request naming only unsupported metrics returns rows with no metrics. Names match regardless of case, surrounding whitespace, underscores, spaces, hyphens and colons. bounceRate and averageTimeOnPage can now each be requested alone, and a request naming one never receives the other.

11/11/2026

  • [Breaking] FunnelFlux no longer adds origin_page_id or origin_page_name to action_data in webhook and middleware payloads. Read the new top-level referring_page_id and referring_page_name fields instead. The other action_data origin keys (origin_node_id, origin_node_name, origin_node_type) are unchanged, and action parameters submitted by the visitor's request still pass through as customer data even when their names match the removed keys.
  • Every webhook payload (node_execution, on_action, conversion_triggered, custom_event_triggered) and every middleware POST payload now carries top-level referring_page_id and referring_page_name strings. They name the most recent page in the current journey before the current step, or before the converting page hit for conversion and custom-event webhooks. Both are always sent from the release that contains runtime change 01a85a4df, and both are empty strings when no earlier page exists. When the page-name lookup fails, referring_page_id is still set and referring_page_name is an empty string. They are JSON payload fields, not URL tokens. The webhook payload JSON Schema declares both as optional strings, so stored 1.0 payloads from earlier deliveries stay valid; schema_version stays "1.0".

03/11/2026

  • [Breaking] IntegrationCapabilities.capi.forced (returned by GET /integrations/capabilities) is renamed to capi.available, matching costSync.available and dictionarySync.available. It is true when the provider offers conversion delivery; a traffic source's own choice stays integration.features.capi.enabled. forced is no longer returned, so clients that read it must switch to available.
  • Ledger and Reporting now document a single authentication scheme, Authorization: Bearer TOKEN. The key query parameter and X-API-Key header schemes are no longer listed.
  • Traffic source meta.conversionTrackingMode (classic or integrations) works with integration setup to choose the route: Classic URL, HTML pixel and custom-event postbacks fire unless the stored mode is integrations and the linked connection plus CAPI switch are on. Stored Classic never sends provider conversions; provider delivery for stored Integrations requires full setup; a legacy or unset mode retains classic delivery and sends CAPI when the integration is fully set up. Saving in Integrations mode removes the stored classic postback, customEventPostback and conversionTrackingSettings; saving in Classic mode, or without a mode, keeps them.
  • TrafficSourceIntegration.platform is documented as read-only: it is derived from the linked connection and any client-sent value is ignored.
  • Saved short links: archived links keep routing traffic and keep their slug reserved; unarchive returns them to the active list. Delete remains the only operation that stops routing.
  • Data reset (POST and DELETE /v1/reporting/update/reset/) now requires at least one restriction besides timeStart/timeEnd and rejects blank restrictToFunnelIDs entries with 400. restrictToTrackingFields keys are documented.
  • Cost updates document a 400 response for rejected segment restrictions; Known Proxy and Hosting Provider connection types require includeProxyHostingTraffic=true. Entrances excluded by includeFilteredTraffic or includeProxyHostingTraffic keep their existing cost. A restrictToDeviceType of Unknown also matches entrances with no device data.
  • Connection Type filters: "" and the Connection data not available label match rows with no connection data only; "0" matches id 0 only.
  • Integration queue rows now include transactionId and destinationEvent.
  • Funnel nodes document the armor node type and nodeArmorRefParams, with the ArmorConfig schema.

22/10/2026

  • Reporting filters for Connection: Connection Type, Device: Device Type and Filtered Status now accept the display label (case-insensitive) or the numeric id string returned by drilldown rows. Unknown values and wildcards return 400. For Device Type, the empty id "" and the Unknown label match id 0 and rows with no device data; "0" matches id 0 only. For Connection Type, "" matches id 0, and only the Connection data not available label matches rows with no connection data.
  • Reporting endpoints no longer document a 429 Rate limit exceeded response. The reporting API no longer applies its own rate limit; account-level rate limits still apply.

12/10/2026

  • [Breaking-ish] IntegrationCapabilities (returned by GET /integrations/capabilities) no longer includes disabledReason. enabled no longer defaults to true: it is now always derived from the provider's configured capability set and is false when that set is empty. Clients should stop treating a missing enabled as enabled and stop reading disabledReason.
  • GET /integrations/{id}/ad-accounts now documents adAccounts as a required response field. Successful discovery with no visible accounts returns adAccounts: [] rather than omitting the field.

08/10/2026

  • Added POST /reporting/logs/integrations/queue (tag Logs): read-only, paginated view of the authenticated owner's delayed integration queue rows. Request body IntegrationQueueLogParams; response IntegrationQueueLogReport with IntegrationQueueLogReportRow rows and IntegrationQueueLogMetadata. status=completed means the delayed-work row closed, not that the conversion was delivered; an expired held row is status=failed with lastError exactly expired. Semantic and filter validation failures return 422 with code 606.
  • POST /reporting/logs/integrations description now documents the row outcome values (queued, delivered, skipped, failed, unknown) and the matching outcome filter. unknown matches a stored empty outcome, and no outcome is excluded by default.

03/10/2026

  • IntegrationCapabilities.capi.configFields entries (IntegrationCAPIConfigFieldRequirement) now carry a required scope field (traffic_source_link) and no longer expose the previous alternatives array. [Breaking-ish] — callers reading alternatives must switch to scope; scope is required.
  • IntegrationCapabilities.costSync.dailyReconcile now exposes scheduleMode (local_delay or fixed_utc_hour) and finalizationUtcHour, alongside the existing finalizationDelayMinutes, so clients can render the reconciliation schedule.
  • Added optional per-link CAPI credentials on TrafficSourceIntegration: datasetId (Meta dataset or pixel ID) and eventSourceId (TikTok Events API event_source_id — not the Marketing API numeric pixel ID); both are required when CAPI is enabled for that provider. A per-link testEventCode is also accepted for provider test events.
  • TrafficSourceIntegrationEventMapping destination type now also accepts conversion_goal, in addition to conversion_action and event_name.

09/09/2026

  • Middleware config timeoutMs now accepts 1 through 60000. The effective cap is plan-dependent (Middleware basic 10s, Middleware Pro 30s synchronous); values above 30000 take effect only with loadingPage.enabled (detached execution, up to 60s).
  • Added MiddlewareConfig.loadingPage (MiddlewareLoadingPage / MiddlewareLoadingTheme / MiddlewareLoadingTextStep) for the Middleware Pro interstitial shown while a detached request runs.
  • Added MiddlewareConfig.responseData (MiddlewareResponseData / MiddlewareResponseVar) for Middleware Pro response-variable extraction, optional redirect query append, and JS delivery.
  • Added MiddlewareConfig.restrictSentData (MiddlewareRestrictSentData / MiddlewareSentDataFilter) to allow-list outbound tracking_fields and visitor-submitted data-buffer keys on all plans.
  • Added authenticated POST /reporting/logs/integrations for paged integration delivery logs (IntegrationLogParams / IntegrationLogReport).
  • IntegrationCapabilities.capi now includes configFields (IntegrationCAPIConfigFieldRequirement) describing required CAPI configuration keys, labels, and alternatives.

21/08/2026

  • Added Saved Short Links under the Assets API (Short Links tag): GET /shortlink/list/, GET /shortlink/find/byId/, GET /shortlink/find/bySlug/, POST /shortlink/save/, PUT /shortlink/save/, DELETE /shortlink/delete/, POST /shortlink/duplicate/, PUT /shortlink/archive/, and PUT /shortlink/unarchive/. The target is a structured funnel entrance, not a generic URL.
  • DELETE /shortlink/delete/ hard-deletes the row and frees its slug; this is not reversible. Use PUT /shortlink/archive/ for the recoverable op — archived links keep their slug reserved and are not routable at Edge.
  • POST /shortlink/duplicate/ requires both a name and a slug (the server cannot derive a user-visible slug) and always creates an active copy. Slug collisions return duplicate_slug, or duplicate_slug_archived when the holder is archived.
  • GET /integrations/{id}/ad-accounts/{adAccountId}/conversion-goals now accepts optional loginCustomerId for manager/login customer context from ad-account discovery.

18/08/2026

  • [Breaking] Reporting removed Insight: Time on Page from the drilldown grouping attributes. Time on page is now stored at millisecond resolution, so per-value grouping is no longer offered; use Insight: Time on Page Segments for duration bands.
  • [Breaking-ish] restrictToMetrics on POST /reporting/drilldown now returns the default metric set when empty, rather than all metrics. averageTimeOnPage and the new bounceRate are excluded from that default set.
  • Added bounceRate to drilldown rows: the share of measured page views, as a percentage from 0 to 100, that ended in under 3000 ms. The denominator is beacon-measured page views only, so the metric describes JS-tracked traffic.
  • averageTimeOnPage and bounceRate are opt-in and are returned only when named explicitly in restrictToMetrics. Both are omitted — never 0 — when nothing was measured, so render an absent value as "no data" rather than zero.
  • IntegrationAdAccount now returns timezone, businessId, and businessName.
  • POST /reporting/drilldown now documents a 400 response for an invalid grouping attribute, i.e. one not returned by GET /reporting/attributes/.
  • restrictToMetrics on POST /reporting/logs/incoming-postbacks is now a documented enum of the valid row-field names.
  • The 200 response on the reporting cost-update endpoint is clarified: cost updates are accepted for asynchronous processing, matching segments are queued, zero-entrance segments complete as no-ops, and warehouse application happens after the response.
  • Tag responses now include tagType and tagValue.
  • Documented the outgoing-connection rules enforced when saving a funnel: webhook and visitorTag nodes must have exactly one outgoing connection to an active node; externalUrl nodes must have none; root, rotator, condition, logic, actionProxy, and middleware nodes must have at least one; lander, offer, landerGroup, offerGroup, and label nodes are unrestricted.

09/08/2026

  • Added Auth API integration lifecycle and synchronization endpoints under /integrations, including OAuth, reauthorization, ad-account, conversion-goal, and sync-status operations.
  • [Breaking] Billing subscription and usage routes now use /subscriptions/me and /usage/events instead of /subscription and /usage; added addon trial start, cancel, convert, and eligibility endpoints.
  • Added TrafficSource.integration configuration schemas for server-side delivery, dictionary sync, cost sync, and event mapping.
  • Reporting now documents averageTimeOnPage, explicit restrictToMetrics values, and optional timezone fields for conversion and custom-event uploads.

16/07/2026

  • Added authenticated, owner-scoped POST /reporting/logs/incoming-postbacks for paged raw incoming-postback logs. restrictToMetrics selectively returns requested row fields (case-insensitive); conversion rows return eventNumber: null.

15/06/2026

  • [Breaking-ish] Middleware destination timeoutMs now accepts 1 through 10000, clamps to 10 seconds, and still defaults to 5 seconds.
  • Middleware Config list status now explicitly allows an empty query value; empty or not-deleted excludes deleted configs.
  • [Breaking] Middleware routes now use idRoute string route keys plus priority for evaluation order. The default route is idRoute: "default" with priority: 0; non-default routes use priorities 1 through 255.
  • [Breaking] Middleware funnel connections now use connectionMiddlewareParams.onRouteId with string route keys such as approved, review, or default.
  • [Breaking-ish] Middleware reporting logs now return matchedRouteID as the string route key and no longer expose matchedRouteLabel. Middleware summary rows include averageLatencyMs.
  • Added hasBasicAuthPassword and hasHmacSecret to MiddlewareDestination so clients can detect and preserve stored middleware destination secrets. REST read responses now return ******** placeholders instead of real basicAuth.password or hmacSecret values when those secrets are set.
  • Added nodeMiddlewareRefParams.destinationOverrides for middleware funnel nodes, supporting node-level headers, basicAuth.username, basicAuth.password, hmacSecret, hasBasicAuthPassword, and hasHmacSecret without overriding destination URL, method, or type.
  • Documented execute_timeout_ms in outbound Middleware Node POST payloads and added Middleware Function Templates with copy-ready JavaScript, TypeScript, Python, Go, Cloudflare Workers, and Google Cloud Functions examples.
  • Removed MiddlewareRoute.responseMode from the Middleware Config contract. Middleware routes now always continue through the selected funnel connection; no navigate, return_success, or return_reject response modes are exposed in the MVP API.
  • Tightened middleware condition validation docs: body_json.<path> accepts dot segments only, status_code values must be integer HTTP codes from 100 to 599, and body_text supports text operators.
  • Updated MiddlewareDestination.url documentation to clarify that middleware destinations must be HTTPS outside development environments.
  • Documented that FunnelFlux token placeholders are plaintext-only for middleware destination URLs and are not expanded inside custom header values.
  • Added global page group category endpoints on both /page/group/* and /pagegroup/* paths using categoryType=landerGroup or categoryType=offerGroup; only global page groups are category-managed.
  • Added reporting attributes Element: Global Lander Group Category and Element: Global Offer Group Category.

12/06/2026

  • Added Middleware Config endpoints under the Assets API for listing, finding, creating, updating, deleting, duplicating, archiving, unarchiving, and moving middleware configurations by category.
  • Added POST /reporting/logs/middleware/summary to return middleware configuration assets merged with middleware execution-log aggregates.
  • Added a Middleware Nodes guide covering synchronous routing behavior, request payloads, delivery headers, route evaluation, timeouts, fallback behavior, and destination restrictions.

09/06/2026

  • Updated POST /auth/login to document the MFA-required response shape with mfa_required, mfa_token, factor, and mfa_requirements; internal MFA follow-up endpoints remain hidden from the public API docs.
  • Corrected Auth API token and user response field types: expires_at and updated_at are numeric timestamps, permissions is an array of strings, and email_verified is a boolean.

08/06/2026

  • Added Logic Script asset endpoints under the Assets API for listing, finding, creating, updating, deleting, duplicating, archiving, unarchiving, and moving Logic Scripts by category.
  • Added Logic Script helper endpoints: GET /logicscripts/language for the backend-owned language catalog and POST /logicscripts/validate for validating script code, static route keys, and dynamic route detection.
  • Added funnel logic-node routing support through connectionLogicParams.onRouteKeys[]; connect a default route for runtime fallback when validation, compilation, execution, or dynamic route matching cannot produce a connected route.
  • Updated funnel delete documentation to clarify that Logic Scripts are global assets and are not deleted by funnel cascade.
  • Added a Logic Scripts guide for building route-key scripts, including a copyable AI prompt for generating valid snippets.

01/06/2026

  • Added Webhook Config endpoints under the Assets API for listing, finding, creating, updating, deleting, duplicating, archiving, unarchiving, and moving webhook configurations by category.
  • Added POST /reporting/logs/webhooks/summary to return webhook configuration assets merged with webhook delivery-log aggregates.
  • Added funnel schema support for webhook nodes and action-triggered webhook dispatch via nodeWebhookRefParams, connectionActionParams.webhookConfigIDs, and connectionActionProxyParams.webhookConfigIDs.
  • Added Webhook Payloads, including the public webhook_outbound_payload_v1 delivery headers, event types, examples, and downloadable JSON Schema.

23/05/2026

  • Bulk delete and archive endpoints for offer sources, categories, campaigns (funnel groups V1), and funnel groups V2 now accept a ParentMutationRequest body with a cascadeAction of cascade, move, or orphan, plus an optional targetParentID (required when cascadeAction is move). Omitting cascade options preserves existing behaviour. Legacy single-ID query parameters (idOfferSource, idCategory, etc.) remain accepted for backwards compatibility.
  • Funnel delete now documents its cascade scope: funnel-scoped (local) assets are soft-deleted alongside the funnel — page groups, conditions and logic scripts whose restrictToFunnelId points at it, plus AI node settings belonging to that funnel.
  • Added POST /reporting/update/customevents/ to submit custom event updates.
  • Visitor Tag endpoints (/visitor-tags/...) are now fully documented: find, list, list-by-status, create, update, duplicate, delete, archive/unarchive, and move-category.

08/05/2026

  • Conditions now use a route ID and priority key for determining what routes go where, and the route processing priority. This separates ordering from connection logic. Use the new route IDs and priorities. For IDs, use route_{id} and generate an ID with our normal ID generator for consistency. Route IDs can also be names of your choosing.
  • Consolidated category APIs to one canonical /category API rather than endpoints nested under other assets.
  • Many other changes not previously added to this changelog.

19/06/2024

  • [Breaking Changes] - The raw events API has been updated to a new approach, allowing more direct data based on our DB columns directly.
  • All metrics are now selected through the restrictToMetrics attribute
  • A maximum of 15 columns of data can be returned per request
  • You can now directly access tracking field data with this approach
  • Maximum paging length has been increased from 1,000 -> 10,000
  • Note that because this accesses raw data, it currently does not resolve names, just IDs. We will extend this later to add mappings for asset IDs

11/06/2024

  • Added conversionTime attributes to conversion uploads
  • Added disablePostback attributes to conversion uploads
  • Updated hit reporting spec, pending changes in the UI

03/04/2024

  • Added restrictToMetrics and returnUniqueVisitors information to reporting API.

15/07/2023

  • Updated API domain to api.funnelflux.pro

03/09/2022

  • Added ID generation article

22/06/2022

  • Initial release of the v1 API docs, with limitations and some omissions
Last modified on November 16, 2026
Webhook Payloads