Clay: Complete Guide to Data Enrichment and B2B Prospecting with AI
Machine-translated from Italian. Read the original
ClayWhat is Clay
Clay is a platform for automating sales prospecting and B2B data enrichment. It works by combining a relational spreadsheet interface with federated access to dozens of external data providers (including company databases, search engines, public registries, and professional sources) and AI models for text processing.
The core of the tool lies in its ability to orchestrate sequential search flows (waterfall enrichment). Instead of querying a single data provider, Clay allows you to query a series of databases in cascade until the sought data (e.g., verified business email, technology stack, or employee count) is found. It also integrates AI agents capable of visiting websites, summarizing key information, mapping job postings, or drafting hyper-personalized outreach drafts at scale.
Getting Started
Getting started with Clay requires a logical configuration of workflows, similar to structuring a simplified ETL pipeline for sales.
- Workspace Creation: Sign up at clay.com and set up the company workspace.
- Initial List Import: Upload a CSV file with a list of company domains or target profiles, or connect a direct data source (e.g., CRM list or export from Sales Navigator/database).
- Waterfall Configuration: Create a data enrichment column. To find a contact's email, define the provider cascade by configuring fallback criteria in case the first provider does not return a valid address.
- Using Claygent (AI Agent): Add an AI prompt-based column to analyze the target company's website (e.g., "Identify if the company uses Kubernetes, citing the line from the job description or documentation").
- Flow Verification and Testing: Run the process on 10-20 rows to verify the correctness of extracted data before applying automation to larger lists, consuming credits.
Concrete Use Cases
1. Technology Stack Detection for B2B Vendors
A cloud consulting firm needs to identify SMBs using specific hypervisors or cloud providers.
- Action: Clay scans company domains, analyzes DNS records, meta tags, and open job positions on dedicated portals via AI prompts.
- Expected Result: A table with a boolean column (e.g., "Uses AWS/Azure") and corresponding documentary evidence link, ready to segment the campaign.
2. Automated B2B Email Research and Validation in Cascade
A sales team must contact IT Directors of 500 Italian manufacturing companies but only has first name, last name, and company domain.
- Action: Execution of a waterfall logic that queries multiple verified contact providers and validates SMTP server reachability before saving the record.
- Expected Result: Email address validity rate above 85-90% without having to subscribe separately to three or four different tools.
3. Account-Based Marketing with Hiring Signals
An IT-specialized recruiting agency intends to monitor companies opening selections for critical roles (DevOps, Data Engineer).
- Action: Through agent-assisted scraping and data enrichment, Clay periodically monitors the "Careers" pages of a target list.
- Expected Result: Automatic alert and row populated with posting date, open position, and associated Talent Acquisition Lead contact.
4. Normalization and Cleaning of Messy CRM Data
An internal database contains duplicate records, non-normalized company names, and uncategorized industry sectors.
- Action: Clay receives the payload from the CRM via webhook, uses AI to extract the correct legal form, normalizes the industrial sector according to standard taxonomies (e.g., NACE/ATECO or LinkedIn industry), and returns it to the CRM.
- Expected Result: Cleaned and synchronized data without manual data cleaning interventions.
Pricing and Plans
| Plan | Price | Period | Notes |
|---|---|---|---|
| Free | 0 | - | 500 actions, 100 data credits per month |
| Launch | 167 $ | month | 15,000 actions, 3,000 data credits |
| Growth | 446 $ | month | 40,000 actions, 6,000 data credits |
| Enterprise | See official site | annual | On request |
Source: official pricing page, verified by Penzai on 11/13/2026. Prices may change: always check before purchasing.
Pros and Cons
Strengths
- Cascade enrichment flexibility: Allows maximizing data coverage rates by querying multiple sources without having to sign separate enterprise contracts with each data vendor.
- Integrated agents for contextual scraping: AI can read unstructured web pages, bypassing the rigidity of traditional scrapers based on XPath/CSS selectors that break when layouts change.
- Direct integration between data and logic: The table environment makes every data transformation step transparent and inspectable per row.
Limits and Critical Issues
- Rapid credit consumption: With cascade logic or complex AI agent calls, monthly credits and actions can run out quickly if preventive stop filters are not set.
- Complexity for non-technical users: Although it uses a table interface, building advanced logic, formulas, and HTTP calls requires a mindset close to data engineering.
- GDPR considerations for the European market: Mass enrichment of personal contacts (including nominal work accounts) requires careful assessment of legitimate interest, proper notice to data subjects, and GDPR compliance.
When NOT to Use It
- If the target market is purely B2C: Clay is designed for company data and professional profiles; it is not suitable for tracking or profiling private consumers.
- For small contact lists (fewer than 50-100 contacts per month): The platform cost and learning curve do not justify the expense compared to manual research.
Alternatives
Several tools exist on the market dedicated to prospecting, data enrichment, and sales automation:
- Salesforce Agentforce by Salesforce: Salesforce AI agents for sales, customer service, and marketing in the CRM.
- Docusign IAM by Docusign: Electronic signature and agreement management with AI contract analysis.
- Zendesk AI by Zendesk: Customer service software with AI agents that respond to tickets.
- Apollo.io by Apollo: Sales prospecting platform with contact data and AI.
- Intercom Fin by Intercom: Customer support with Fin, the AI agent that resolves requests.
Integrations and API
Clay is designed to sit upstream or midway between data sources and enterprise operational systems:
- CRM and Sales Engagement: Bidirectional integration with major market CRMs and email/outreach sending tools, allowing reading of leads and automatic updating of enriched records.
- Webhooks and Custom HTTP Calls: Native functionality to make GET/POST requests to external endpoints; this enables querying proprietary APIs or internal services before proceeding with enrichment.
- No-code Automation Tools: Support for platforms like Make or Zapier to extend data routing logic.
For the complete and constantly updated list of individual CRM connectors and natively integrated providers, it is recommended to verify on the official website.
GitHub, Open Source, and Skills
Clay is a proprietary closed-source SaaS platform; no official repositories of the core software are available.
- nicbarker/clay: High performance UI layout library in C. (18,434 stars, C, Zlib license)
- pissang/claygl: A WebGL graphic library for building scalable Web3D applications (2,865 stars, JavaScript, BSD-2-Clause license)
- HaD0Yun/CozyClay: Open source previs software in the browser: block a scene, pose characters, author camera moves and cuts, then take the same shots to an AI video model. (2,016 stars, JavaScript, AGPL-3.0 license)
- CLAY-3D/OpenCLAY: CLAY: A Controllable Large-scale Generative Model for Creating High-quality 3D Assets (985 stars)
- pissang/clay-viewer: 3D model viewer with high quality rendering and glTF2.0/GLB export (796 stars, JavaScript, BSD-3-Clause license)
GitHub data as of 11/20/2026. "Official" only if the repository belongs to the vendor's organization.
To operate effectively on Clay, recommended skills develop along this progression:
- Basic Level: Tabular data management and cleaning (CSV, Excel), understanding email deliverability metrics, writing basic prompts for information extraction from texts.
- Intermediate Level: Use of conditional logic (IF/THEN), regular expressions (Regex) for filtering strings and domains, configuration of cascade workflows (waterfall enrichment), and credit/action budget management.
- Advanced Level: Connection via Webhook and custom REST APIs, pipeline ETL flow modeling, automation of triggers to CRM via complex JSON payloads, regulatory compliance audit on data processing (GDPR).
Tips and Insights
- Set rigid exit conditions (stop conditions): Before configuring an enrichment call or a costly AI scan in credits, insert control columns that verify the presence of minimum prerequisites (e.g., if the "Domain" column is empty or generic like gmail.com, skip execution).
- Isolate tests in a limited view: Do not apply new formulas or enrichment columns on tables with thousands of rows all at once. Always work with filtered views of 5-10 records until output quality validation is confirmed.
- DNS and MX Verification: Use domain validity and MX record verification columns first before spending credits on nominal email search providers; if the domain does not accept mail, any email search will be in vain.
Questions for the Community
- How do you manage the balance between enrichment rates with external providers and GDPR compliance notices for EU-based contacts in your organizations?
- What criteria or formulas do you use to prevent unexpected credit exhaustion when setting up complex waterfalls?
- Have you integrated Clay via webhook with proprietary internal APIs or databases? With what architecture and error handling pattern?