What is waterfall enrichment in Clay? Waterfall enrichment in Clay is a column configuration that calls multiple data providers in sequence. If Provider 1 returns a result, the cascade stops. If Provider 1 returns nothing, the column automatically calls Provider 2, then Provider 3, until a result is found or all providers are exhausted. You pay only for successful calls.
No single data provider covers everyone. That is not a complaint about vendors. It is just arithmetic: email coverage rates across providers vary by industry, company size, and geography, and any one source leaves gaps. Waterfall enrichment in Clay solves this by calling a second provider only when the first one comes up empty, and a third when the second fails. You maximize coverage without paying for duplicates. Here is how to set it up correctly and why provider order is the decision most teams get wrong.
What is waterfall enrichment in Clay? Waterfall enrichment in Clay is a column configuration that calls multiple data providers in sequence. If Provider 1 returns a result, the cascade stops. If Provider 1 returns nothing, the column automatically calls Provider 2, then Provider 3, until a result is found or all providers are exhausted. You pay only for successful calls.
The average email coverage rate for a given B2B provider sits somewhere between 50 and 70 percent of a typical list, depending on the industry and seniority mix of contacts. The variance is higher for small companies, niche industries, and certain geographies. This is not a data quality problem in the traditional sense. It is a coverage problem.
A single provider returns nothing for 30 to 50 percent of your list. You have two bad options without a waterfall: suppress those contacts entirely (leaving pipeline on the table) or manually research them one by one (expensive and slow).
Waterfall enrichment in Clay handles this automatically. You define the provider sequence once, and Clay handles the fallback logic without human intervention.
See what-is-email-waterfall-enrichment for a conceptual primer if you want the foundational explanation before the Clay-specific workflow.
Rule 1: Put your highest-coverage provider first. The primary provider should be the one that covers the largest percentage of your typical ICP. This minimizes the number of rows that need to fall through to backup providers, which directly controls your API cost.
Rule 2: Put your most accurate provider first, not necessarily your cheapest. If you are prioritizing deliverability, accuracy beats coverage. A provider that returns fewer results but returns accurate ones is better in position one than a provider that covers 80 percent of your list but returns a lot of stale or invalid emails.
Rule 3: Position specialty providers where they match your list. Some providers have stronger coverage for specific industries, geographies, or company sizes. If you are enriching a list heavy in European mid-market companies, a provider with stronger European coverage should appear earlier in your waterfall for that list.
See b2b-phone-number-data-providers for a similar provider-order discussion for phone number enrichment, which has its own waterfall logic.
Step 1: Create a new column. In your Clay table, click the plus icon to add a column. Name it something clear: "Email - Waterfall" or "Work Email." This is a container column, not a single-provider call.
Step 2: Add the first provider. Inside the column configuration, add your primary provider. Set it to return an email address. Do not add a fallback yet.
Step 3: Add the conditional fallback. Clay lets you chain enrichment steps inside a single column. After the first provider, add a condition: "If result is empty, call..." and select your second provider. Repeat for Provider 3 if needed.
Step 4: Add a confidence flag. Add a separate column that outputs which provider returned the result. This is metadata that helps you analyze provider performance over time. If Provider 1 fills 55 percent of your rows and Provider 3 is filling 10 percent, you know Provider 2 is the gap-filler that matters.
Step 5: Add an email validation step. After the waterfall runs, add a syntax and deliverability validation column. Some providers return formatted addresses that fail at send time. Validate before the rows feed into a sequence. InboundLabs contacts come with 98% email deliverability on verified contacts, which is why starting from a verified seed list reduces how many rows need the full waterfall treatment.
The Cascade Coverage Stack is the visual model for how waterfall enrichment fills gaps across providers. Each provider fills what it can, gaps fall to the next provider, and unresolved rows get suppressed or flagged at the end.
The typical cascade result: Provider 1 fills 55 to 65 percent. Provider 2 adds 15 to 20 percent. Provider 3 adds 5 to 10 percent. Final coverage: 80 to 90 percent. The remaining 10 to 20 percent go to a suppress list or a manual review queue. Hypothetical coverage rates; your actual numbers will vary by list composition and provider selection.
Primary provider criteria:
Backup provider criteria:
Fallback provider criteria:
See waterfall-enrichment-tools for a comparison of which tools support native waterfall logic versus which require manual chaining.
Some rows will remain empty after the full waterfall runs. You have three options:
Suppress them. If the row has no email and no phone after three providers, it is hard to reach via automated outreach. Remove it from active sequences.
Flag for manual review. If the account is high-value enough to warrant human research, create a manual review queue. A rep can look up the contact directly or use a different research method.
Try a different contact at the same company. Sometimes the specific contact you are targeting has no public email, but a different person with the same buying authority at the same company does. A fallback column that tries to find an alternative contact title at the same domain is a useful fourth step for high-priority accounts.
See clay-table-tutorials for the full table build workflow, of which waterfall enrichment is one step.
Calling all providers regardless of the primary result. This is the most expensive mistake. If you configure every provider to run in parallel rather than in sequence, you pay for three provider calls on every row, including the ones Provider 1 already answered. Waterfall logic is sequential, not parallel.
Using the same data source in both positions. If Provider 1 and Provider 2 both source their data from the same underlying database, your waterfall is redundant. When Provider 1 has no record of a person, Provider 2 with the same source will also have no record. Diversify sourcing methods.
Skipping validation after the waterfall. A result from a backup provider is more likely to be stale or low-confidence than a result from your primary. Always validate after the cascade completes.
Running the waterfall on an unfiltered list. If 40 percent of your imported rows do not match your ICP, you are running three provider calls for contacts you will never use. Filter first on ICP criteria, then run enrichment. See how-to-build-an-icp-list-for-outbound-sales for the filtering step.
InboundLabs is the most efficient seed layer for a waterfall. When you start a Clay table with contacts from InboundLabs's database of 280M verified B2B contacts, you begin with a higher baseline coverage rate. The 98% email deliverability on verified contacts means many rows will not need the full waterfall at all.
Use InboundLabs to pull your ICP-matched list with verified emails and verified direct dials first. Then use Clay's waterfall enrichment to fill in additional data layers: job title enrichment, company news, tech stack, intent signals, AI personalization snippets. The waterfall in this model is filling data enrichment gaps, not just email gaps, which is where it adds the most value.
Monthly plans, no annual lock-in. Start free, no credit card needed.
See how InboundLabs finds verified contacts instantly. inboundlabs.app
Waterfall enrichment in Clay is not a single feature. It is a provider sequencing strategy that compounds coverage across sources while controlling API spend. The teams that get it right define provider order by coverage profile and sourcing methodology, validate after each cascade, and suppress unfilled rows rather than sending to them. The teams that get it wrong call all providers in parallel, pay three times as much, and wonder why their reply rates are low.
What is waterfall enrichment? Waterfall enrichment is a data enrichment strategy that calls multiple providers in sequence. Each provider only fires when the previous one returns no result. The cascade continues until a result is found or all providers are exhausted, then the row is either populated or flagged as unfillable.
Does Clay support native waterfall enrichment? Yes. Clay has built-in conditional logic that lets you chain multiple enrichment providers inside a single column. When Provider 1 returns empty, Clay automatically triggers Provider 2. This happens at the row level without any manual intervention. Check clay.com for current feature documentation. (Checked September 2026.)
How many providers should I stack in a waterfall? Three providers is the practical maximum for most use cases. Beyond three, you are filling such a small percentage of rows that the cost-per-result becomes very high. A three-provider stack typically achieves 85 to 92 percent coverage on a well-filtered ICP list.
Why does provider order matter? Because the first provider fires on every row, making it the most expensive position. You want your highest-coverage, best-value provider in position one. Position two and three fire progressively less often and can be more expensive per call since they are handling a smaller percentage of rows.
What happens to rows that no provider can fill? Best practice is to suppress them from active sequences and put them in a manual review queue. Sending to a contact with no verified email means relying on guessed formats, which harms sender reputation and reply rates. High-value accounts warrant manual research; the rest should be suppressed.
How is waterfall enrichment different from parallel enrichment? Waterfall enrichment is sequential: each provider only fires if the previous one fails. Parallel enrichment calls all providers simultaneously and picks the best result. Parallel is faster but significantly more expensive. Waterfall is the standard for cost-conscious outbound teams.
Can I use waterfall enrichment for phone numbers too? Yes. The same cascading logic applies to phone number enrichment. See b2b-phone-number-data-providers for which providers have the strongest phone coverage and how to order them for a phone waterfall.
LSI keywords: waterfall data enrichment, Clay enrichment waterfall, email waterfall, provider cascade, B2B email coverage, contact enrichment strategy, data enrichment cascade, multi-provider enrichment, Clay enrichment tutorial, email deliverability enrichment, outbound data workflow, API enrichment cost control
What is a Clay table? A Clay table is a spreadsheet-style workspace where each row represents an account or contact. Each column can call a data provider, run an AI prompt, or perform a calculation. The table enriches and scores records automatically, reducing per-contact research to a column definition written once.
What is a Wappalyzer alternative? A Wappalyzer alternative is any tool that identifies the technologies a company uses, with many also adding contact data, firmographics, or intent signals on top. Pure detection tools (like BuiltWith) scan websites for front-end technologies. Full sales intelligence platforms layer those signals onto verified B2B contacts so you can act on the data immediately.
BuiltWith is a server-side web crawler that identifies the technologies, analytics tools, hosting providers, and frameworks detected on any website. It maintains a database of over 50,000 technologies across millions of domains and provides historical adoption data showing when companies added or dropped specific tools. It is the most comprehensive technology coverage database available for B2B sales teams doing technographic segmentation.
Technographic data is information about what software, hardware, and technology infrastructure a company uses. It is collected by crawling public web properties, analyzing job postings for technology mentions, or tracking software usage through browser extensions. In B2B sales, it is used to identify prospects using complementary or competing tools, to qualify accounts by tech sophistication, and to personalize outreach with relevant tech-stack context.
No commitment. No credit card. Just 50 free verified contact lookups.