Lead Enrichment

Overview

The most common lead enrichment failure is sending every record to every provider — paying 3x the necessary cost while introducing conflicting data that requires manual reconciliation. The non-obvious rule of enrichment is the waterfall pattern: sequence providers from cheapest to most expensive, and from broadest coverage to most specialized, with each provider only firing when the previous one returned empty or low-confidence results. This approach maximizes coverage while minimizing cost per enriched record.

This skill provides a complete waterfall enrichment methodology that chains multiple data providers in a cost-optimized sequence. It specifies separate enrichment strategies for company data (firmographics, technographics, funding, revenue) and person data (contact details, job history, social profiles), with domain-based matching as the primary join key. Confidence thresholds determine when a record is "good enough" versus when it needs the next provider in the waterfall.

The deliverables include a Waterfall Configuration (provider sequence with trigger conditions), a Company Enrichment Table schema, a Person Enrichment Table schema, Confidence Threshold Rules for data quality gating, and a Normalization Protocol for reconciling conflicting data across providers. The methodology draws from DAMA-DMBOK data-quality dimensions and Ziellab's 3-waterfall architecture for production-grade enrichment pipelines.

When to Use

• User says "enrich these leads" or "append data to this list" → activate this skill
• User asks "how do I set up an enrichment waterfall" or "which enrichment providers should I use" → use this skill
• User mentions "Clay enrichment," "data enrichment," "contact enrichment," "company enrichment" → this skill applies
• User has a lead list from lead-finding and needs to fill in missing data
• User is building a Clay table and needs to configure enrichment columns
• User wants to normalize data from multiple enrichment sources