Clay

## How does Clay find work email addresses using only a person's name and company?

## Overview Clay finds work email addresses from a person's name and company by utilizing a multi-layered, sequential 'waterfall' enrichment process. This system is designed to maximize the probability of finding a valid email by querying numerous integrated data providers in a specific order, rather than relying on a single source. The primary inputs for this process are a person's full name and the company's domain. While a LinkedIn Profile URL is not required, providing one significantly increases the accuracy of the search. ## Key Features Clay integrates with a comprehensive list of email enrichment providers, including Prospeo, DropContact, Datagma, Hunter, PeopleDataLabs, Nimbler, Apollo, Lusha, Snov.io, and Findymail, among others. The system is also designed for cost-efficiency, as data credits are typically refunded for failed attempts by a specific provider in the chain. In addition to its third-party database integrations, Clay employs its own AI-powered web scraper, 'Claygent.' This agent acts as another layer in the enrichment process, capable of visiting public web pages—such as corporate websites, social media profiles, news articles, or conference speaker lists—to search for contact information that may not be present in structured databases. This is particularly useful for finding emails of individuals who are less represented in traditional B2B data sets. ## Technical Specifications Email validation is an integral part of the workflow to ensure the deliverability of the emails found. By default, Clay uses ZeroBounce for validation, but users have the flexibility to substitute other providers like Findymail. The validation process typically includes multiple checks: a syntax check for correct formatting, a domain check to verify the mail server is active, an SMTP check to confirm the specific mailbox exists, and a reputation check to flag potential spam traps or disposable addresses. ## How It Works The waterfall mechanism is central to Clay's effectiveness. It involves a sequence of searches across a wide array of third-party data providers. If the first provider in the sequence fails to return a valid email, the system automatically proceeds to the next, and so on. This continues until an email is found or all providers in the sequence have been queried. This method substantially increases coverage compared to using any single tool. ## Use Cases For example, users can direct Claygent to find an email for a specific person on a company's 'About Us' page. ## Limitations and Requirements Users should be aware of certain limitations. AI-based scraping can sometimes extract irrelevant emails, and approximately 40% of websites may not have a publicly visible email address, underscoring the importance of the multi-provider waterfall as a backup. No single method guarantees 100% accuracy, and secondary verification is often recommended to protect sender reputation. ## Comparison to Alternatives Different validation tools offer varying levels of stringency; for instance, Findymail is described as more 'conservative,' which may result in fewer bounces but could miss some valid emails, whereas ZeroBounce is considered more 'aggressive.' The system also has mechanisms to handle edge cases. By default, Clay considers emails from 'catch-all' domains as valid. However, for users prioritizing deliverability, a toggle can be enabled to only accept emails marked as 'Safe to Send,' effectively filtering out catch-all addresses. The validation process also helps identify and categorize role-based addresses like 'info@' or 'support@'. ## Summary In conclusion, Clay's method for finding work emails is a robust, multi-pronged approach that combines a waterfall enrichment sequence across more than a dozen data providers with the web-scraping capabilities of its AI agent, Claygent. This is coupled with an integrated, multi-layered validation process to ensure data quality and deliverability. While this system significantly increases coverage and success rates, its accuracy is ultimately dependent on the availability of information across its network of public and private data sources. The platform's flexibility allows users to customize the process, such as by choosing their preferred validation provider or even reducing costs by connecting their own OpenAI API keys for AI-driven searches.