Signal Ark’s External Ingestion capability lets you push signals from any source directly into the Signal Ark pipeline — not just the sources Signal Ark monitors natively. If you operate a custom web scraper, license data from a third-party provider, or build internal tooling that detects buying intent, you can send that data to Signal Ark and have it scored, classified, and routed alongside natively sourced signals.Documentation Index
Fetch the complete documentation index at: https://docs.signalark.app/llms.txt
Use this file to discover all available pages before exploring further.
What you can ingest
Signal Ark provides two dedicated ingestion endpoints:Market signals
Send funding events, job postings, tech install detections, news mentions, web visits, and other market signals. Signal Ark deduplicates, classifies, and ICP-scores each signal on arrival.
Social engagement events
Send social interactions — likes, comments, reposts, and mentions — from LinkedIn, X (Twitter), Hacker News, or any platform you monitor. High-warmth, ICP-matched leads are automatically promoted to your sales pipeline.
Common use cases
- Data providers and enrichment partners: If you supply market intelligence data, ingest signals on behalf of your customers’ Signal Ark workspaces.
- Custom scrapers and monitors: Pipe the output of your own web monitors or keyword trackers into Signal Ark for AI classification and routing.
- Internal tooling: Connect internal systems — product analytics, support tickets, or usage telemetry — that generate signals relevant to your sales motion.
- Data warehouse exports: Replay historical signal data into Signal Ark to backfill account context or bootstrap ICP models.
How ingested signals are processed
When you POST signals to the ingestion API, Signal Ark runs the following pipeline automatically:- Schema validation — Each signal is validated against the required field schema. Malformed records are rejected with a descriptive error.
- Deduplication — Duplicate signals (same company, same source, same signal within a short time window) are collapsed and counted in the batch response stats.
- Account matching — Signal Ark attempts to match each signal to an existing account using
company_domain. Providing a well-formatted domain significantly improves match rates. - AI classification — Accepted signals enter the AI pipeline for categorization, ICP scoring, and “Why Now” generation.
- Play routing — Signals that match active Play Templates trigger the configured delivery actions (Slack alerts, CRM pushes, webhooks).
Ingested signals flow through the same pipeline as natively sourced signals. There is no secondary tier — your data receives the same AI enrichment and routing as everything else in Signal Ark.
Authentication
All ingestion API requests require a Signal Ark API key. See Authentication for setup instructions.Validate before you ingest
Use the Validate endpoint to test your signal payloads during development without writing data to Signal Ark. It runs full schema validation and account-matching simulation against the same rules as the live ingestion endpoint, without consuming any credits or triggering the AI pipeline.API reference
POST /v1/signals/ingest
Batch ingest market signals
POST /v1/social/ingest
Batch ingest social engagement events
POST /v1/signals/validate
Validate signal payloads without ingesting
POST /v1/signals/feedback
Submit feedback to improve AI scoring