What Is Clay and Why Does It Matter?
Clay is a data enrichment and workflow automation platform purpose-built for go-to-market teams. At its core, Clay is a spreadsheet-like interface where each column can be powered by an API call, an AI prompt, a formula, or a data enrichment provider. This design lets GTM teams build sophisticated data workflows without writing code, though it also supports custom HTTP requests and JavaScript formulas for advanced users.
Clay matters because it solved a problem that previously required custom engineering. Before Clay, building a multi-provider enrichment waterfall required writing Python scripts, managing API keys, handling rate limits, normalizing data formats, and building error handling. This took weeks of engineering time and ongoing maintenance. Clay reduces this to hours. A GTM engineer or technically proficient RevOps professional can build a production-quality enrichment waterfall, connect it to a CRM, and start generating results in a single afternoon.
As of early 2026, Clay integrates with over 100 data providers and tools. It processes over 50 million enrichments per month across its customer base, which includes companies from seed-stage startups to Fortune 500 enterprises. The platform has become the de facto standard for GTM data operations, and proficiency in Clay is now listed as a requirement in over 70% of GTM engineering job postings.
Getting Started with Clay
To begin using Clay, create an account at clay.com. The free tier includes 100 credits per month, which is enough to explore the platform's capabilities but not enough for production use. The Pro plan at $149/month includes 10,000 credits, which supports moderate enrichment volume. The Explorer plan at $349/month includes 50,000 credits, and the Enterprise tier offers custom pricing with higher volumes and dedicated support.
Start by creating a new table and importing data. You can import from CSV, Google Sheets, HubSpot, Salesforce, or a webhook. For your first exploration, import a small CSV of 50-100 target companies with at minimum the company name and domain columns. This gives you a manageable dataset to experiment with before running larger enrichment jobs.
Connect your data provider API keys in Clay's Settings > Integrations page. At minimum, connect Apollo (free tier available with 10,000 credits/month), which gives you access to contact discovery and email enrichment. If you have accounts with other providers like People Data Labs, Clearbit, or ZoomInfo, connect those as well. Each integration requires an API key from the respective provider.
Core Concepts
Tables and Rows
Clay tables work like spreadsheets with superpowers. Each row represents a record (a company, a person, a deal, an event) and each column represents a data point. What makes Clay different is that columns can be dynamic: powered by API calls that execute for each row. You can think of each cell as the output of a function call, where the inputs are other columns in that row and the function is an API call, AI prompt, or formula.
Enrichment Columns
Enrichment columns are Clay's core feature. When you add an enrichment column, you select a data provider and a specific action (for example, 'Apollo - Find Person by Domain and Title'). You map input fields from your existing columns (company domain, title keyword) and Clay executes the enrichment for each row, populating the column with the provider's response. Enrichment columns can be chained: the output of one enrichment becomes the input to the next, enabling complex multi-step workflows.
Formulas and Conditionals
Clay supports formulas similar to Excel but with additional capabilities. Use formulas to clean data (trim whitespace, standardize formats), implement conditional logic (if email is empty, mark for next provider), perform calculations (score accounts based on multiple criteria), and combine data from multiple columns into a canonical output. Formulas use a JavaScript-like syntax and can reference any column in the current row.
AI Columns
AI columns let you run OpenAI or Claude prompts for each row in your table. This is extraordinarily powerful for tasks like researching a company and generating a personalized first line for an email, categorizing companies by vertical based on their website description, summarizing 10-K filings or press releases, scoring the relevance of a lead based on multiple data points, and generating custom fields that no data provider offers. AI columns consume Clay credits based on the model used and the prompt length. GPT-4o is the most common model for production use, offering a good balance of quality and cost.
Building Enrichment Waterfalls Step by Step
Planning Your Waterfall
Before building anything, define what you need. At minimum, most outbound enrichment workflows require: the contact's full name, current title, verified work email, LinkedIn URL, company name, company domain, industry, employee count, and optionally a direct dial phone number. Determine which data provider you will use for each layer, considering cost, coverage for your ICP, and the specific data points each provider returns.
Layer 1: Company Enrichment
Start by enriching company-level data. If you only have company names, use a Clearbit or Apollo company lookup to get the domain, industry, employee count, funding stage, and technology stack. Create a formula column that confirms the company matches your ICP criteria (e.g., employee count between 50 and 5,000, industry is SaaS or fintech, headquartered in the US). Mark non-qualifying companies with a 'Disqualified' flag and exclude them from further enrichment to save credits.
Layer 2: Contact Discovery
For qualifying companies, use Apollo's People Search to find contacts matching your target persona. Specify the title keyword (e.g., 'VP Sales,' 'Head of Revenue,' 'CRO'), seniority levels, and departments. Apollo returns up to 5 matching contacts per search. Create a formula column that selects the best match based on title relevance and seniority. If Apollo returns no results, add a conditional People Data Labs search as a fallback.
Layer 3: Email Discovery and Verification
With a specific person identified, find their verified work email. If Apollo did not return a verified email in the contact discovery step, run a dedicated email finder using Apollo's email finder endpoint, then PDL, then Lusha as fallbacks. Each subsequent provider only runs for records where previous providers failed. After the waterfall, run every email through an independent verification service like ZeroBounce or NeverBounce. Create a final 'clean_email' column that contains only verified emails.
Layer 4: Additional Enrichment
For records with verified emails, add context that supports personalization. Use an AI column to visit the company's website and write a one-sentence summary of what they do. Use another AI column to analyze the person's LinkedIn profile (if available) and identify a relevant talking point. Add technographic data from BuiltWith to identify which tools the company uses that are relevant to your product. This additional context transforms a contact list into a personalization-ready dataset.
API Integrations: Webhooks and HTTP Requests
Inbound Webhooks
Clay tables can be triggered by incoming webhooks, turning them into real-time processing engines. When you enable webhook input on a Clay table, you get a unique URL that accepts POST requests. Any system that can send an HTTP request can trigger a Clay workflow. Common use cases include: processing inbound leads the moment they submit a form (connect your website form to Clay's webhook, and the lead is enriched, scored, and routed in real time), triggering enrichment when a CRM record is updated (use HubSpot or Salesforce workflow rules to fire webhooks to Clay), and processing signal events from monitoring tools.
HTTP Request Columns
For data sources that Clay does not have a native integration with, HTTP request columns let you call any REST API directly. You configure the URL, method (GET, POST), headers (including authentication), and request body. The response is parsed and stored in the column. This opens up virtually unlimited data sources. You can query Google Maps for company location data, call the SEC's EDGAR API for public company filings, hit a custom internal API for proprietary data, or connect to niche data providers that Clay has not partnered with. HTTP request columns make Clay extensible to any use case.
Clay for Outbound: Connecting to Sending Platforms
The output of your Clay enrichment workflow typically flows to an email sequencing platform. Clay has native integrations with Instantly, Smartlead, Lemlist, and several other sending tools. The integration works by mapping Clay columns to sequence fields: email address, first name, company name, personalized first line, and any custom variables your email templates reference.
For Instantly, connect your API key in Clay's integrations page, then add an 'Add to Instantly Campaign' action column. Map the required fields and select the target campaign. When you run the table, each qualifying record is automatically added to your Instantly campaign with all personalization data included. The same process works for Smartlead and other supported platforms.
For tools without native Clay integrations, use a webhook output column. Configure a POST request to the tool's API with the enriched data as the payload. This works with any platform that has an API, including Salesloft, Outreach, HubSpot sequences, and custom-built tools.
Clay for Inbound: Real-Time Lead Processing
One of Clay's most valuable applications is real-time inbound lead enrichment. The typical setup works like this: a prospect submits a demo request form on your website. The form submission triggers a webhook to a Clay table. Clay immediately enriches the lead with company data, contact data, technographics, and intent signals. An AI column scores the lead based on your ICP criteria. A routing formula determines which AE should receive the lead. An output webhook pushes the fully enriched, scored, routed lead to your CRM.
The entire process takes 15-30 seconds from form submission to CRM entry. Compare this to the typical manual process where an SDR reviews the form submission hours later, manually looks up the company on LinkedIn, tries to determine if it is a good fit, and eventually routes it to an AE. Speed-to-lead studies consistently show that responding within 5 minutes generates 4-8x more conversions than responding within an hour.
Advanced Patterns
Lead Scoring
Build a lead scoring model directly in Clay using formula columns and AI. Create a scoring rubric: assign points for ICP fit (employee count in range: +10, right industry: +10, uses relevant technology: +15), engagement signals (visited pricing page: +20, downloaded whitepaper: +5), and data quality (verified email: +5, phone available: +5, LinkedIn URL found: +3). Sum the points in a formula column and create threshold categories: A leads (60+ points), B leads (40-59), C leads (20-39), D leads (below 20). Route A and B leads to AEs immediately, C leads to nurture sequences, and D leads to disqualification.
Territory Routing
Use formula columns to implement territory-based routing. If your sales team is organized by geography, use the enriched company headquarters data to assign regions. If organized by company size, use employee count or revenue tiers. If organized by vertical, use the industry classification from enrichment. The formula output is a rep name or team identifier that gets passed to your CRM for automated assignment.
Deduplication
Before pushing records to any downstream system, deduplicate within Clay. Use the company domain as the primary dedup key for company records and email address for contact records. Clay's built-in dedup functionality can identify and merge duplicate rows, or you can use formula columns to flag duplicates for manual review. For CRM integration, always check for existing records before creating new ones. Use an enrichment column that queries your CRM (via HubSpot or Salesforce integration) to check if the contact already exists, and skip creation if they do.
Clay vs Alternatives
Clay vs Apollo
Apollo is both a data provider and an outbound execution platform. Clay is a data orchestration layer that uses Apollo (and many other providers) as inputs. They are complementary, not competitive, for most use cases. Apollo is better if your needs are simple: find contacts at target companies and email them. Clay is better when you need multi-provider enrichment, complex logic, AI personalization, or integration with systems beyond Apollo's built-in capabilities. Most sophisticated GTM teams use both: Apollo as a data provider within Clay, and sometimes Apollo's sequencing for simpler campaigns.
Clay vs ZoomInfo
ZoomInfo is a data provider with some workflow features. Clay is a workflow platform that connects to many data providers including ZoomInfo. ZoomInfo's data is excellent for enterprise accounts, but it is expensive and locked within ZoomInfo's ecosystem. Clay lets you use ZoomInfo where it excels (enterprise firmographics) and supplement it with cheaper providers for segments where ZoomInfo's premium pricing is not justified. Teams replacing ZoomInfo with a Clay-orchestrated waterfall typically save 40-60% on data costs while improving coverage.
Clay vs Custom Code
Some engineering teams build custom enrichment pipelines in Python or Node.js. This approach offers maximum flexibility but requires ongoing maintenance, error handling, monitoring, and infrastructure management. Clay eliminates this overhead for 90% of use cases. The exception is extremely high-volume (millions of records per month) or highly custom workflows that require logic beyond what Clay's interface supports. Even then, teams often use Clay for prototyping and move to custom code only for specific high-scale production workflows.
Pricing Considerations
Clay uses a credit-based pricing model. Each enrichment, AI call, or API request consumes credits. The cost per credit varies by plan but ranges from approximately $0.005 to $0.015. A typical enrichment waterfall that queries two providers and runs one AI column per record costs roughly 5-8 credits per record, or $0.04-0.12 depending on your plan.
For planning purposes: if you process 5,000 outbound records per month with a three-provider waterfall and AI personalization, expect to use 30,000-50,000 credits per month. This maps to the Explorer plan ($349/month) or a custom Enterprise plan. Note that credits consumed by data providers may also require separate subscriptions to those providers for API access. Apollo's free tier includes 10,000 monthly export credits, which covers many use cases. PDL, Clearbit, and others charge separately.
The ROI calculation is straightforward. Compare the cost of Clay plus data providers against the cost of the manual processes or custom engineering they replace. A single GTM engineer using Clay can produce output that previously required a team of SDRs and a part-time developer. The tool cost is a rounding error compared to the labor savings.
Production Best Practices
Monitor Credit Usage
Clay credits can be consumed quickly if you are not careful. Set up usage alerts at 50% and 80% of your monthly credit allotment. Review your table usage weekly to identify columns consuming more credits than expected. Disable or delete test tables that are still actively running. Use conditional execution aggressively to avoid running enrichments on records that do not need them.
Version Control Your Tables
Clay does not have built-in version control, so create your own system. Before making significant changes to a production table, duplicate it as a backup. Name tables with clear conventions: 'Outbound Enrichment v3 - Production' vs 'Outbound Enrichment v4 - Testing.' Document the logic of each table in a shared wiki or Notion page, including provider ordering, fallthrough conditions, scoring logic, and output mappings.
Test Before Running at Scale
Always test enrichment workflows on a small sample (10-50 records) before running them on your full list. Check that each provider is returning expected results, conditional logic is working correctly, AI columns are producing quality output, output mappings to your CRM or sequencing tool are correct, and deduplication is working. Catching errors on 50 records costs a few dollars. Catching them on 50,000 records costs hundreds of dollars and creates a data cleanup headache.
Build for Reuse
Create template tables for your most common workflows: standard outbound enrichment, inbound lead processing, job change monitoring, and hiring signal detection. When you need to run a new campaign, duplicate the template and adjust the inputs rather than building from scratch. This reduces setup time from hours to minutes and ensures consistency across campaigns.
Data Hygiene
Archive completed tables rather than leaving them active. Clay tables with active enrichment columns continue to consume credits when new rows are added. Set clear data retention policies: how long do you keep enriched data in Clay before archiving or deleting? For GDPR and privacy compliance, ensure you have a process to delete personal data from Clay when requested and that your data retention practices align with your privacy policy.
Clay is the most important tool in the modern GTM technology stack. Mastering it is not optional for anyone serious about GTM engineering, outbound automation, or data-driven revenue operations. The platform's learning curve is moderate, taking most users 2-3 weeks to become productive and 2-3 months to become truly proficient. But the payoff in pipeline efficiency, data quality, and time savings is enormous. Start with a simple enrichment workflow, prove the value, and expand from there.