Schema-Driven Data Validation

Schema-Enforced Lead Data Validation at Scale

We built a system that ingests messy lead data from diverse sources, standardizes it to buyer-defined schemas, and validates every field against typed rules—managed through an admin UI.

Year
2024
Practice
Lead & legal platforms

Overview

Lead-generation platforms ingest data from many suppliers at once, and every supplier sends it differently—different field names, different formats, different gaps. LeadCRM addresses this by enforcing a buyer-defined schema at the moment a lead is ingested, so data that does not conform is caught before it reaches a campaign rather than failing somewhere downstream.

The Challenge

Lead data arrives from partner APIs, CSV uploads, and web forms, each with its own structure and quality issues. Field names vary across providers ("phone," "phoneNumber," "mobile"), formats differ, and missing or invalid values pass through undetected, causing downstream failures in campaign delivery. Vertical-specific requirements—an insurance lead needing particular policy fields, for example—often go unmet because nothing enforces them at the door.

The Solution

We built schema enforcement directly into LeadCRM's ingestion path, on the existing React, Laravel, and MySQL stack. For each campaign, the buyer defines the required fields and their types using five attribute types: string, email, integer, boolean, and date. On every call to the ingestion endpoint, each incoming lead is validated against that campaign's schema; numeric values are normalized to a consistent string form, and key-value filters are evaluated for an exact match. A lead that fails validation or a filter is rejected at ingestion with a recorded reason (for example, "Filter Rejected: state is invalid") instead of silently continuing. An admin UI lets non-technical users define a campaign's schema and inspect why specific leads were rejected.

Outcome (client-reported)

The client reported materially faster lead processing, fewer rejections caused by formatting rather than genuine quality problems, and lower operational cost from removing manual data cleaning. By enforcing data standards at the point of ingestion, buyers were able to safely take on more supplier sources without a proportional increase in data-handling overhead.

Brief us

If you have a system you'd rather not replace — start there.

We respond in writing, within two business days. No discovery call gates the first conversation — brief us and we'll tell you whether we're a fit, or refer you on.

or write to hello@intuio.io