Clay Data Enrichment Guide: Fields, Integrations & Waterfall Setup (2026)

Clay data enrichment is the process of taking a thin record, usually a name and a company or a LinkedIn URL, and filling in everything you actually need to run outreach: a verified work email, a mobile number, an accurate job title, firmographics, and technographics. The difference with Clay is that it does this from a single spreadsheet-style table while querying many data providers at once, instead of locking you into one vendor's database.

This guide is written for RevOps practitioners, founders running their own outbound, and GTM engineers who are building enrichment into a repeatable system. If you have ever exported a list, fed it through one tool, and watched half the emails come back blank, this is the workflow that fixes that. We will cover the fields you can enrich, the integrations that matter, how waterfall enrichment actually works, a step-by-step setup, a worked credit-cost example, and the mistakes that quietly drain credits.

What Clay actually enriches

People talk about contact enrichment as if it were one thing, but in practice you are enriching several distinct categories of data, each with its own providers and reliability profile. Knowing which bucket a field falls into tells you how to source it and how much to trust it.

Contact-level fields

These attach to a person: work email, personal email, direct dial, mobile phone, current job title, seniority, department, and LinkedIn URL. Email and phone are the fields most worth running through a waterfall, because no single provider has good global coverage and match rates swing hard by region and seniority.

Company-level fields

Firmographics describe the account: employee count, revenue band, industry, headquarters location, funding stage, and domain. These tend to be more stable and cheaper than contact data, and a single strong provider often covers most of your needs.

Signal and technographic fields

This is the data that makes outreach relevant: the tools a company runs, hiring activity, recent funding, and intent signals. Technographics and intent are less complete than firmographics, so treat them as prioritization signals rather than hard filters.

Provider names shift over time and Clay adds new ones regularly, so treat the table as a starting map rather than a fixed list. The point is the pattern: each field type has several viable sources, and that is exactly why a waterfall beats any single vendor.

The integrations that matter

Clay ships with a large catalog of native integrations, and the number is less useful than knowing which categories you need. In practice you are assembling four groups, and the real skill is chaining them so the output of one becomes the clean input of the next.

1. Sources that get records into Clay

This is where rows originate. The most common Clay integrations for sourcing are LinkedIn Sales Navigator (import a saved search directly into a table), your CRM (pull an existing HubSpot or Salesforce view to re-enrich stale records), a Google Sheet sync, a raw CSV upload, and the Clay-native company and people search. A useful chaining pattern is to source companies first, then use a "find people" step to expand each account into its decision-makers before any contact enrichment runs, so you are only paying to enrich people who sit inside an account you already qualified.

2. Enrichment providers that fill in fields

These are the data vendors that answer a specific question about a row. For work email, concrete examples include Prospeo, Datagma, FindyMail, Hunter, and Apollo. For phone, Datagma, Nimbler, and ZoomInfo. For firmographics and technographics, Clearbit, BuiltWith, and HG Insights. The chaining principle here is that an enrichment column can take its input from a previous enrichment column, not just from the source. For example, Clearbit returns a clean company domain, that domain feeds the email-finder waterfall, and the verified email then feeds a deliverability check. Each provider hands the next one a better input than the raw list had.

3. AI and research steps

Clay's built-in AI (Claygent) and your own prompts sit between enrichment and destination. Typical uses are scraping a prospect's homepage to extract a one-line value proposition, summarizing a recent funding announcement into a personalization hook, or classifying a company into an ICP tier from its website copy. These steps are best chained after firmographic enrichment, because the AI works far better when it already has a verified domain and industry to reason about rather than guessing from a company name.

4. Destinations that receive the finished record

Once a row is enriched, verified, and gated, it is pushed out. Common destinations are HubSpot and Salesforce for CRM sync, Smartlead and Instantly for cold-email sequencing, Slack for real-time alerts on high-intent accounts, and a generic webhook for anything custom. The chaining rule for destinations is to make them conditional: only rows that passed the ICP gate and the email-verification step should fire the HubSpot or Smartlead action, so junk never leaves the table.

You do not need to wire all four at once. A clean first build is one source, a contact waterfall, an ICP filter, and one destination. Everything else is an upgrade you add when the basic loop is producing clean data.

How waterfall enrichment works

A waterfall enrichment is a single ordered sequence of providers that all answer the same question, such as "what is this person's work email." Clay runs them top to bottom. It tries provider one; if that returns a valid result, it stops and bills you for that one lookup. If provider one comes back empty or fails a validation rule, Clay falls through to provider two, then provider three, and so on until a usable value appears or the list runs out.

Two properties make this powerful. Coverage compounds, because a provider that is strong in one region or seniority band covers the gaps of another. Cost stays controlled, because most waterfalls only charge on a successful return, so you are not paying five vendors to all guess at the same row. The result is a match rate no single provider can hit, at a blended cost that is usually lower than running each provider in parallel.

The trade-off is ordering. The whole model depends on putting the cheapest, most reliable provider first and the expensive long-shot last. Get the order wrong and you pay premium rates on records a cheaper source would have covered. For a deeper provider-by-provider breakdown, see our comparison of waterfall enrichment in Clay versus ZoomInfo versus Apollo.

A worked credit-cost example

Numbers make the ordering argument concrete. Imagine a list of 1,000 contacts and a three-step email waterfall. Clay bills in credits, and credit cost varies by provider, but the relative shape holds regardless of the exact rate card. Suppose Provider A costs 1 credit per successful match and hits 55 percent, Provider B costs 2 credits and recovers 20 percent of the remainder, and Provider C costs 4 credits and recovers half of what is left.

Run the waterfall in the right order. Provider A matches 550 rows at 1 credit each, which is 550 credits. The 450 misses fall to Provider B, which matches 20 percent, so 90 rows at 2 credits, which is 180 credits. The remaining 360 rows fall to Provider C, which matches 50 percent, so 180 rows at 4 credits, which is 720 credits. Total: 820 matched contacts (an 82 percent match rate) for 1,450 credits, or roughly 1.77 credits per valid email.

Now invert the order and run the expensive Provider C first. It would attempt all 1,000 rows, match around 410 of them at 4 credits each, which is 1,640 credits on its own, and you have spent more on a single step than the entire correctly-ordered waterfall cost to reach a higher match rate. That gap, paying premium rates on rows a 1-credit provider would have covered, is exactly the mistake good ordering prevents. The same logic is why you gate on ICP before the waterfall: enriching 1,000 rows when only 600 fit your ICP wastes credits on 400 records you would never contact.

How to set up a waterfall in Clay: step by step

Step 1: Get a clean input

Bring your list into a Clay table from Sales Navigator, your CRM, or a CSV. Before enriching anything, dedupe on a stable key such as LinkedIn URL or email domain plus full name. Enriching duplicates is the fastest way to waste credits, and it is entirely avoidable at this stage.

Step 2: Normalize the core fields

Make sure you have a clean first name, last name, and company domain. Most email providers need name plus domain to perform well, so add a quick step to derive the domain from a company name or website if it is missing. Quality in equals quality out.

Step 3: Add the first provider as a column

Add an enrichment column for your strongest, most cost-effective email provider. Map its inputs to your normalized fields. Run it on a small test batch of 25 to 50 rows first so you can read the real match rate before committing credits to the whole list.

Step 4: Stack providers into a waterfall

Use Clay's waterfall feature, or chain columns with conditional run logic, so the second provider only fires when the first returns blank. Add a third provider the same way. Order them cheapest-reliable to most-expensive. A common B2B pattern is three email providers in sequence.

Step 5: Verify before you trust

Send the winning email through a verification step. A syntactically valid address is not a deliverable one, and pushing unverified emails into a sequence is how sender reputation gets damaged. Keep only results that pass.

Step 6: Gate, then push

Add a filter or conditional rule so only ICP-matching, verified records flow to your destination. Then map the clean fields into HubSpot, Salesforce, or your sequencer. The gate is what keeps junk out of your CRM and your credit burn predictable.

Common mistakes that waste credits and trust

The first and most expensive mistake is enriching before filtering. If a row does not match your ICP, it should never reach a paid provider. Put your firmographic and ICP filters upstream of the contact waterfall so you only buy data on accounts you would actually contact.

The second is bad waterfall ordering. Teams often place a premium provider first out of habit, which means you pay top rates on records a cheaper source would have matched. Reorder by blended cost-per-valid-result, not by brand reputation.

The third is skipping verification. Match rate and deliverability are different metrics. Google's sender guidelines penalize senders whose lists generate high bounce rates, so a 90 percent match rate with 30 percent invalid addresses will hurt your domain more than a smaller, clean list ever would.

The fourth is not monitoring per-provider performance. Match rates drift as vendor databases change. Review hit rates monthly and reorder or swap providers when one slips. Treat the waterfall as a living config, not a set-and-forget block. If you are still choosing tools, our roundup of top lead enrichment tools for sales outreach is a useful companion, and if you are building the broader system, see how this fits into a modern GTM engineering stack.

Putting it together

Clay's value is not any one provider, it is the orchestration. You decide which fields you need, route each one to a ranked set of sources, verify the output, and gate it before it touches your systems. In our experience at DevCommX, a three-provider email waterfall with proper filtering commonly moves a B2B list from roughly half coverage to the 80 to 90 percent range, while keeping spend tied to results rather than guesses. Build the simple loop first, measure each provider, and add complexity only where the data tells you to.

Build This With DevCommX

DevCommX builds autonomous, signal-based AI SDR systems for B2B teams and you own the infrastructure, not just a managed campaign. Clients typically go from setup to 40+ qualified demos within 6 weeks, because the system triggers on real buying signals instead of static lists. Book a GTM strategy call to map this to your pipeline.

FAQ

What is Clay data enrichment?

It is the process of taking a thin record, such as a name and company, and filling in missing attributes like work email, phone, job title, firmographics, and technographics by querying multiple data providers from inside Clay's spreadsheet interface.

How does waterfall enrichment work in Clay?

It queries data providers in a ranked sequence. Clay tries the first provider and only moves to the next when a result is missing or fails a quality check. You are typically charged only when a provider returns a usable value, which raises coverage while controlling cost.

Which Clay integrations should I start with?

For most B2B teams, start with two or three email providers in a waterfall, one phone provider, and a firmographic source. Add technographic and intent providers later, once your core contact data is reliable.

Does Clay replace ZoomInfo or Apollo?

No. Clay sits on top of many providers, including ones comparable to ZoomInfo and Apollo, and routes each lookup to whichever source has the data. It is an orchestration layer rather than a single database.

How do I avoid wasting Clay credits?

Gate expensive enrichments behind conditional run logic, dedupe before enriching, order waterfalls cheapest-reliable first, and only enrich records that match your ICP filters. Most waste comes from enriching rows that should never have entered the table.

What is a good email match rate to expect?

In our experience at DevCommX, a single provider often lands in the 40 to 65 percent range on a typical B2B list. A well-ordered three-provider waterfall commonly reaches the 80 to 90 percent range, depending on geography, seniority, and list quality.

References

• Clay University and official documentation
• Gartner research on data quality and analytics
• ZeroBounce resources on email verification best practices