All five tiers operational

Verified work emails, found by an AI agent.

An AI agent descends a five-tier waterfall — free signals first, paid vendors only when nothing cheaper resolves. The AI judges weak signals, ranks LinkedIn sources, and learns every domain pattern so the next call is cheaper.

Free tier · 200 enrichments/moNo credit cardSOC 2 in progress
request.py
POST
from waterfallreach import Client

client = Client(api_key="wr_live_…")

result = client.enrich(
    linkedin_url="linkedin.com/in/jane-doe",
    # or: name, company, domain, email, phone
    tier_ceiling=3,          # skip Tier 4
    risk_ceiling="T2",       # compliance cap
)
print(result.email, result.verified)
response.json
200 · 142ms
// exited at tier 0 — free signals only
{
  "email": "jane.doe@acme.com",
  "verified": true,
  "title":    "VP Engineering",
  "company":  "Acme Inc",
  "domain":   "acme.com",
  "source":   "tier_0_permutation",
  "risk_class": "T0",
  "tier_exited": 0,
  "cache_hit":   true
}
Why it matters

Three outcomes that compound.

Each tier of the waterfall optimizes a different axis. Together they make the system cheaper, more accurate, and less risky as you use it.

Cost

Skip the vendor lock-in

Replace three single-vendor enrichment tools with one waterfall. Cost-per-verified-contact drops because cheaper tiers run first.

Quality

Bypass gateway false-positives

MX fingerprinting detects Proofpoint, Mimecast, Barracuda, and Cisco. Catch-all probes walk the MX priority chain.

Compliance

T0–T3 source-risk class

Every email arrives with a legal-risk class. Customers cap the ceiling. GDPR posture without code changes.

The cascade

Five tiers. Independently bypassable.

Set a tier ceiling. Cap the compliance class. Early exit on the first verified match. The cascade descends only as far as it needs to.

T0

Free signals

POST /v1/enrich?ceiling=0

20+ public-records modules. An LLM fuses them into a ranked candidate.

T1

Self-hosted SMTP

POST /v1/verify-smtp

MX fingerprinting detects gateways. Catch-all probe walks the chain.

T2

LinkedIn router

POST /v1/enrich/linkedin

Multi-source router. An LLM ranker picks the best candidate.

T3

Paid finders

POST /v1/enrich?ceiling=3

Premium vendors. Only fired when cheaper tiers cannot resolve.

Surfaces

Three ways to call it.

One core. Three integration surfaces — pick whatever fits your stack.

HTTP API

FastAPI service

Docker container. Health checks, structured logs, idempotency keys, SHA-256 row fingerprints.

docker pull ghcr.io/.../waterfallreach
Python

pip install

Drop-in client for batch pipelines. Async-friendly. Streaming results over large lists.

pip install waterfallreach
n8n

Workflow templates

Pre-built nodes for orchestration. Bulk enrich, verify, and reconciliation flows.

n8n import wr-bulk-enrich

Pilot access is open.

Bring a list of LinkedIn URLs, domains, or names. We'll run the cascade and show you exactly which tier each row exits on — and what it would cost.