Most Clay users set up one enrichment provider, run their list, get 60–65% email coverage, and assume that's what Clay delivers. It's not. That's what a single provider delivers.
Clay waterfall enrichment is the approach Clay was designed around: a sequenced, stop-on-success lookup that queries one provider at a time, charges credits only when a result is found, and keeps falling through the sequence until either a verified email appears or the stack runs out of options. The mechanic is straightforward, but the documentation stops short of answering the questions that actually matter for practitioners: which providers to stack, in what order, at what cost per contact, and what the output needs to look like before it feeds your sequencing tools.
This post covers the exact provider sequence DemandLab runs at volume for clients, the real credit cost math with a worked example for 1,000 contacts, the Claygent decision logic that most guides skip entirely, and the output schema that connects a finished enrichment table to Instantly and Dripify without manual cleanup.
What Clay Waterfall Enrichment Is (and Why Single-Provider Lookup Fails)
Waterfall enrichment is a sequenced lookup strategy for building email contact data. Clay queries your first data provider; if it returns a verified email, the sequence stops and credits are charged at that provider's rate. If Provider 1 returns nothing, Clay moves to Provider 2 at no cost, and continues down the sequence until a match is found or the stack is exhausted.
This stop-on-success mechanic is what separates waterfall enrichment from running all providers simultaneously. You pay only for the provider that delivers a result, not for the entire stack.
The reason single-provider setups fail is coverage, not price. Hunter.io carries approximately 60–65% of US B2B contacts in its database [CITATION NEEDED: Hunter coverage rate — Source: Hunter.io transparency page or third-party enrichment benchmark]. Apollo covers a partially overlapping set at a comparable rate. Neither provider fills the other's gaps because they draw from similar upstream sources. Using either alone means 35–40% of your list has no email attached when the enrichment run finishes.
Coverage ceiling by approach:
| Approach | Expected Email Coverage | Avg. Cost per Found Email | Best For |
|---|---|---|---|
| Single provider (Hunter only) | 60–65% | Lowest | Small lists, tight budgets |
| Single provider (Apollo only) | 58–63% | Low-medium | HubSpot-integrated teams |
| Two-provider waterfall | 75–80% | Medium | Mid-volume outbound |
| Four-provider waterfall | 85–92% | Medium-high | High-volume sequencing programs |
The drop from 90%+ to 62% coverage isn't a data quality problem. It's a missing sequence. A properly configured four-provider waterfall running the same list gets coverage that a single provider cannot approach.
For teams building a signal-based GTM system with Clay, enrichment coverage is not a nice-to-have. A 35% gap in email data means 35% of your targeted accounts never enter a sequence, regardless of how good the sequencing is.
The DemandLab Provider Stack: Exact Sequence and Why This Order
DemandLab runs the following provider sequence for client enrichment tables. The order is deliberate: cheapest and fastest providers go first to minimize credit spend; highest-cost providers go last as fallbacks only.
The stack:
| Tier | Provider | Database Coverage | Credit Cost Level | Best For |
|---|---|---|---|---|
| 1 | Hunter.io | ~65% US B2B | Lowest | Volume outbound, verified format matching |
| 1 (alt) | Prospeo | ~68% US B2B | Low | Senior/executive contacts, LinkedIn-sourced data |
| 2 | Apollo.io | ~72% US B2B | Low-medium | SMB audiences, US-focused campaigns |
| 3 | Tomba | ~30% (personal email) | Medium | Founders, early-stage, no corporate email |
| 4 | Claygent | ~80% of remaining unknowns | Highest | Executives, international contacts, recent role changes |
Tier 1 choice: Hunter vs. Prospeo
Hunter.io is the default choice when you are running high-volume lists and credit cost is the primary constraint. It uses pattern-matching (first.last@company.com) combined with SMTP verification and covers the broadest set of standard US business email formats. Default to Hunter for volume lists; default to Prospeo when your ICP is Director-level and above.
Prospeo is the better choice when your list skews toward VP-level and above. Prospeo sources its data from LinkedIn company data and achieves higher accuracy for senior roles at the cost of slightly higher per-lookup pricing.
Tier 2: Apollo
Apollo's strength is volume. Its database covers 265M+ contacts and performs particularly well on SMB companies and US-centric audiences. The caveat: in our experience running Apollo at volume for clients, data freshness varies. A found email may be 12–24 months old for contacts who have recently changed roles. Position Apollo at Tier 2 rather than Tier 1 because its per-lookup credit cost is slightly higher than Hunter or Prospeo, and Hunter should resolve the majority of straightforward business email lookups first.
Tier 3: Tomba
Tomba indexes personal email addresses better than business email providers. Use it when the target contact works at a very small company, operates as a founder or consultant, or has a non-standard email format that Hunter and Apollo's pattern-matching cannot resolve. Personal email deliverability is lower than corporate email, so flag Tomba-sourced emails in your output schema and apply a higher qualification threshold before sequencing.
Tier 4: Claygent
Claygent is not a database query. It is a web research agent. It browses LinkedIn profiles, personal sites, and company directories in real time to find contact information that no static database has indexed. For this reason, it is 5–10x more expensive per lookup than Tier 1 providers and should only run on contacts that all prior tiers have failed to enrich.
The Credit Math: What 1,000 Contacts Actually Costs
The credit cost conversation is entirely absent from the SERP on Clay waterfall enrichment. Every ranking result either ignores cost entirely or publishes numbers that favor the provider writing the article. Here is what 1,000 B2B SaaS contacts (US-based, VP/Director/Manager level) actually costs through the DemandLab stack.
Worked Example: 1,000 Contacts
The following credit costs are approximate based on standard Clay plan pricing at mid-2026 rates. Your actual cost per lookup will vary by plan tier and provider contract.
Assumptions:
- List: 1,000 US B2B SaaS contacts, VP-Director-Manager level
- Waterfall sequence: Prospeo (Tier 1) → Apollo (Tier 2) → Tomba (Tier 3) → Claygent (Tier 4)
- Credits cost at approximate market rates as of mid-2026
| Provider Tier | Contacts Resolved | Cost Per Found Email | Subtotal |
|---|---|---|---|
| Tier 1 (Prospeo) | 680 (68%) | ~$0.010 | $6.80 |
| Tier 2 (Apollo) | 130 (13%) | ~$0.015 | $1.95 |
| Tier 3 (Tomba) | 80 (8%) | ~$0.020 | $1.60 |
| Tier 4 (Claygent) | 55 (5.5%) | ~$0.075 | $4.13 |
| No match found | 55 (5.5%) | No charge | $0.00 |
| Total | 945 emails (94.5%) | ~$0.015 avg | ~$14.48 |
Total enrichment cost for 1,000 contacts: approximately $14–18, depending on your specific credit plan and provider pricing tier.
Key benchmark: At this rate, enriching 10,000 contacts per month runs approximately $140–180 — less than most teams spend on a single tool subscription that delivers half the coverage.
Where Credits Get Wasted
Three mistakes drain credits without improving coverage:
Wrong provider order. Putting Claygent at Tier 2 instead of Tier 4 means your most expensive provider runs on 70–80% of your list before cheaper providers have attempted a lookup. Credit spend goes up; coverage stays the same.
Stop-on-success disabled. If the "stop if found" toggle is not enabled in Clay, every contact runs through the full sequence regardless of whether an email was found at Tier 1. A contact that Hunter resolves immediately still triggers Apollo, Tomba, and Claygent lookups. This multiplies your credit spend by 3–4x with zero benefit.
Re-enriching already-enriched contacts. Every time you run the Clay table without a pre-filter, contacts with verified emails run through the enrichment stack again. Add a filter condition at the table level: only run enrichment where "Email Status" is blank or null.
For teams who want to understand how intent data can drive prioritization — deciding which contacts to enrich first based on buying signals — the cost-per-enriched-contact math changes significantly when you run the waterfall only on accounts already showing intent.
Claygent vs. API Providers: When to Use Each
Most articles treat Claygent as another enrichment provider. It is not. API providers (Hunter, Apollo, Prospeo, Tomba) query static databases that were built by scraping, purchasing, or aggregating contact records at some point in the past. Claygent is a web research agent that browses the web in real time and extracts contact information from wherever it can be found.
This distinction matters operationally.
Use Claygent when:
- The contact is a C-suite executive or VP at a company with fewer than 50 employees. Database coverage drops sharply at small companies.
- The contact is based in Europe, APAC, or LATAM. US-centric providers have poor coverage outside North America.
- The contact changed roles in the last 6 months. Static databases return the old business email at the prior company.
- The contact has an unusual email format that Hunter's pattern-matching cannot resolve.
Do not use Claygent when:
- You have a large volume list of mid-market US B2B contacts. API providers resolve the majority at a fraction of the cost.
- Budget is constrained. Claygent at scale is expensive. Position it at the end and set a credit cap on how many contacts fall through to it each run.
- The contact already has a verified email in your CRM. Add a pre-filter to skip enrichment entirely.
The DemandLab waterfall sequence: each provider tier attempts lookup only after all prior tiers return no result.
How to Set Up Clay Waterfall Enrichment: Step-by-Step
This walkthrough assumes you have a Clay account with at least one provider connected. If you are starting from scratch, connect Hunter.io or Prospeo first — they are the lowest-cost Tier 1 options and the most widely used.
Step 1: Build Your Input Table
Start with a Clay table containing four minimum fields: First Name, Last Name, Company, and LinkedIn URL. LinkedIn URL is the most reliable input for all providers — prioritize sourcing it before running enrichment. Tables built on company domain alone have lower resolution rates on providers that rely on name-plus-domain matching.
Add an "Email Status" column using a formula: it will track verified, not found, and pending states as the waterfall runs.
Step 2: Connect Providers in Sequence
In Clay, open the enrichment column menu and select your Tier 1 provider (Hunter.io or Prospeo). Map the input fields: First Name, Last Name, and Company to the provider's expected input schema.
Enable "Stop if found." This toggle is buried in the column settings but is the most critical configuration in the entire setup. If it is turned off, Clay queries every provider for every contact regardless of whether an earlier tier already found an email. Turn it on for every enrichment column in the waterfall.
Add Provider 2 (Apollo) as the next enrichment column with the same input mapping. Set the trigger condition: Apollo only runs if Tier 1 returned null. Add Tomba as Tier 3 with the same conditional trigger. Add Claygent last as Tier 4.
Step 3: Add Email Verification
After the waterfall sequence resolves, add an email verification step using Hunter Verify. Only verify contacts where an email was found in the waterfall — verification consumes credits on every lookup, so running it against the full list wastes budget on blank fields.
Set the "ready_to_sequence" boolean column to TRUE when two conditions are met: an email was found by the waterfall AND verification passed.
Step 4: Add the Pre-Filter
Apply a table-level filter: only run enrichment on contacts where Email Status is blank or null. This prevents re-querying contacts enriched in previous runs, which is the single most common source of unnecessary credit spend on repeat enrichment cycles.
Step 5: Run a Test Batch
Before running your full list, test with 50 contacts. Review: which tier resolved the most contacts? What percentage fell through to Tier 3 and 4? A healthy Tier 1 resolution rate for a standard US B2B list is 65–70%. If Tier 1 resolves fewer than 50%, verify the input mapping before running the full list. The most common cause: sending full name as a single combined string instead of separate first name and last name fields. That mapping error consistently drops Tier 1 resolution rates by 15–25 percentage points.
The Output Schema That Feeds Instantly and Dripify
The enrichment run is only useful if its output connects cleanly to your sequencing tools. Most teams export a CSV and clean it manually — that step eliminates itself when the Clay table is built with the right output schema from the start.
Required output fields:
| Field Name | What It Contains | Maps To |
|---|---|---|
| verified_email | Found and verified email address | Instantly: email; Dripify: email |
| first_name | Contact first name | Both: personalization tokens |
| last_name | Contact last name | Both: personalization tokens |
| company_name | Current company name | Both: personalization, filtering |
| job_title | Current job title | Both: personalization, segmentation |
| linkedin_url | LinkedIn profile URL | Dripify: connection request target |
| ready_to_sequence | Boolean: TRUE = email found + verified | Export filter — only TRUE contacts advance to sequences |
| enrichment_tier | Which provider found the email (1–4) | QA signal: Tier 4 contacts may warrant a higher qualification threshold |
Connecting to Instantly
Export the Clay table filtered to ready_to_sequence = TRUE as a CSV. Map verified_email to Instantly's email field, first_name and last_name to contact name fields. Use enrichment_tier as a custom variable tag inside Instantly for internal routing — you can use it to direct Tier 4 contacts into a different, lower-volume sequence with higher personalization if budget allows.
Connecting to Dripify
Dripify requires a LinkedIn URL to execute connection-request sequences. Filter your export: push only contacts where linkedin_url is present AND ready_to_sequence is TRUE. Map the linkedin_url field to Dripify's Profile URL input. Contacts without a LinkedIn URL should route to Instantly only, not Dripify, to avoid sequence enrollment errors.
For a complete picture of routing enriched contacts into your outbound GTM system, including how enrichment tables connect to intent-triggered sequences, the signal-based GTM Clay guide covers the full system architecture.
Frequently Asked Questions
Q: What is waterfall enrichment in Clay?
Waterfall enrichment is a sequenced provider lookup strategy where Clay tries one data provider at a time, stops when a verified result is found, and charges credits only at the provider that delivers. This stop-on-success logic is what separates waterfall setups from querying all providers simultaneously. A properly configured waterfall of four providers achieves 85–92% email coverage at a fraction of the cost of running all providers on every contact.
Q: Which Clay enrichment providers give the best email coverage?
The provider sequence that consistently delivers 85–92% email coverage for US B2B SaaS contacts starts with Hunter.io or Prospeo (Tier 1), falls through to Apollo (Tier 2) for broader database coverage, then to Tomba (Tier 3) for personal email fallback, and uses Claygent (Tier 4) as a final web-research fallback for contacts not found in any database. The order is critical — starting with cheaper providers keeps aggregate cost per found email low.
Q: How much does Clay waterfall enrichment cost per 1,000 contacts?
At DemandLab's provider stack, enriching 1,000 B2B contacts runs approximately $14–18 and achieves 94–95% email coverage. The per-contact average works out to roughly $0.015 when most contacts resolve at Tier 1 and 2. Contacts that fall to Claygent cost approximately $0.075 each. Keeping Claygent as a last-resort provider — rather than Tier 2 or 3 — is what keeps the aggregate cost in the $14–18 range rather than $40–60.
Q: When should I use Claygent vs. an API-based enrichment provider?
Use Claygent when a contact cannot be found through standard API providers. Claygent performs real-time web research and finds emails from LinkedIn profiles, personal sites, and company directories that no static database indexes. It is the right tool for executives at small companies, international contacts, and people who recently changed roles. It costs 5–10x more than API providers per lookup, so position it at the end of the waterfall and let cheaper providers resolve the majority of your list first.
Q: What output fields should a Clay enrichment table include to feed Instantly and Dripify?
At minimum: verified_email, first_name, last_name, company_name, job_title, linkedin_url, a ready_to_sequence boolean (TRUE only when email is found and verified), and an enrichment_tier field that records which provider found the email. The ready_to_sequence field is the export gate — filter your output to only advance contacts where this is TRUE. Contacts lacking a linkedin_url should route to Instantly only, not Dripify.
Q: What email coverage rate should I expect from waterfall enrichment?
A four-provider waterfall achieves 85–92% email coverage for standard US B2B SaaS audiences. Coverage drops to 70–80% for European audiences due to GDPR-related gaps in provider databases. If you are running below 75% coverage with a four-provider stack, check three things: provider order (Claygent should be last), whether "stop if found" is enabled on every enrichment column, and whether your input mapping sends first name and last name as separate fields rather than a combined full-name string.
Q: How do Clay credits work in a waterfall setup?
In a waterfall, Clay only charges credits at the provider where a result is found. If Provider 1 returns a verified email, you pay Provider 1's per-lookup rate and the sequence stops. If Provider 1 returns nothing, no credits are charged and the sequence moves to Provider 2. This continues until either a result is found or all providers in the sequence have been tried. Contacts with no match found across all tiers consume credits only from tiers where a lookup was attempted and nothing returned.
Sources
- HubSpot, State of Sales Report 2025 — B2B sales productivity and data enrichment benchmarks
- [CITATION NEEDED: Hunter.io coverage rate — Source: Hunter.io transparency page or third-party enrichment benchmark]
- [CITATION NEEDED: Apollo database size and coverage — Source: Apollo.io product documentation or third-party review]
The practitioners who run the best outbound programs are not the ones with the biggest contact lists. They are the ones who get emails for 90%+ of their targeted accounts and pay a rational amount per result. Clay waterfall enrichment is the mechanism that closes that gap.
If you want to see where your current GTM data infrastructure stacks up, take the GTM Maturity Assessment — it is free, takes five minutes, and shows exactly which parts of your pipeline infrastructure are holding back your outbound performance.