Signal Data Onboarding Guide

Welcome to Autobound Signal Data! This guide covers technical setup and working with our team.

Onboarding Timeline

Week	Milestone
Week 1	Kickoff call, credentials delivered, Slack Connect setup
Week 2	First data pull, schema review, pipeline built
Week 3-4	Signals matched to your records, first users enabled
Ongoing	Bi-weekly syncs for first 2 months

Onboarding Checklist

Team Setup

Task
☐ Join Slack Connect channel with Autobound
☐ Schedule kickoff call (30 min)
☐ Identify project team (see roles)

Technical Setup

Task	Resources
☐ Receive and activate GCP credentials	Setup instructions
☐ Verify bucket access	Delivery docs
☐ Download first data file	Bucket structure
☐ Parse and validate schema	Schema docs
☐ Match signals to your records	Entity resolution

Go-Live

Task	Resources
☐ Set up automated refresh pipeline	Sample script
☐ Surface signals to end users	—
☐ Schedule 30-day review call	—

Who Should Be Involved

Role	Responsibility
Data Engineer	Build ETL pipeline, manage GCS access
RevOps / Sales Ops	Map signals to CRM, define routing
Project Sponsor	Unblock resources, define success

Kickoff attendees: Data Engineer + RevOps + Sponsor recommended.

Working with Autobound

Slack Connect

Your shared channel is for:

Quick questions (response within 4 hours)
Data quality issues
Schema clarifications

Success Sessions

30-minute check-ins, 2-3x per month for the first 1-2 months:

Review pipeline progress
Troubleshoot integration issues
Optimize match rates
Share best practices

After go-live, we move to monthly or as-needed.

Core Milestones

Week 1: Access & Kickoff

Kickoff call completed
Credentials received and tested
Slack Connect active

Week 2: First Data Pull

Downloaded data from at least one bucket
Validated schema fields
Initial pipeline created

Week 4: Review & Optimize

Match rate calculated
Signal taxonomy defined for your use case
30-day review scheduled

Technical Setup Details

Authentication

# Activate service account
gcloud auth activate-service-account --key-file=~/credentials/autobound-key.json

# Verify access
gcloud storage ls --project=autobound-signal-delivery

Bucket Structure

Each signal type has timestamped folders:

gs://autobound-{signal-type}/
├── 2026-01-05T12-00-00Z/    ← Most recent
│   ├── output.jsonl
│   └── output.parquet

Always pull from the most recent folder. See Delivery docs for details.

Download Data

# List folders
gcloud storage ls gs://autobound-10k/

# Download latest
gcloud storage cp gs://autobound-10k/2026-01-05T12-00-00Z/output.parquet ./

Format	Best For
JSONL	Streaming, debugging
Parquet	Data warehouses, analytics

Schema Overview

{
  "signal_id": "uuid",
  "signal_type": "10k",
  "signal_subtype": "aiInvestment",
  "detected_at": "2026-01-05T12:00:00Z",
  "company": {
    "name": "Acme Corp",
    "domain": "acme.com"
  },
  "data": {
    "summary": "...",
    "relevance": 0.85
  }
}

Full Schema

Complete field reference

Matching Signals

Priority	Join Key
1st	`company.domain`
2nd	`company.linkedin_url`
3rd	`company.name` (fuzzy)

Entity Resolution

Best practices for matching

Automated Pull Script

#!/bin/bash
BUCKET="gs://autobound-10k"
LATEST=$(gcloud storage ls $BUCKET | tail -1)
gcloud storage cp "${LATEST}output.parquet" /data/signals/

Key Documentation

Resource	Link
Schema	/docs/schema
Entity Resolution	/docs/resolution
Delivery & Buckets	/docs/delivery
Signal Catalog	/docs/signal-catalog
Sample Files	/docs/files

Questions?

Slack Connect — Quick questions
[email protected] — Technical issues
Your account manager — Custom requirements