How it works
A pipeline, not a pile of exports
Four stages that feed each other — designed so a growth team can run the whole loop daily without a data engineer in the loop.
Extract
Pull audiences from any supported channel.
- Create extraction tasks from profiles, hashtags, posts, or search
- Track every job with live status and full history
- Results land in the lead store, deduped by platform identity
Enrich
Expand records through the platform data service.
- Resolve and enrich with built-in response caching
- Idempotency keys protect against double-spending on retries
- Enrichment history shows exactly what changed, and when
Monitor
Watch targets continuously instead of re-scraping by hand.
- Scheduled sweeps with per-group health and last-check state
- Events stream every meaningful change on tracked targets
- Bulk actions and matrix views keep large watchlists manageable
Automate
Turn the pipeline into workflows that run themselves.
- Visual builder with triggers, nodes, and conditions
- Webhooks, schedules, and workflow templates included
- Durable runs with retries — failed nodes can be resumed individually
The credit model
Usage-based pricing with honest accounting
Credits are the single currency across extraction, enrichment, monitoring, and AI features — governed by a ledger designed so you never pay twice for the same result.
Reserve before you spend
Data operations reserve credits up front — so a job can never silently overspend your balance.
Refunded when nothing is spent
If an upstream spend never confirms, the reservation is released back. Only real spend sticks.
A ledger, not a counter
Every purchase, grant, usage, and refund is a row in a transaction ledger you can audit.