Waterfall enrichment: what it is and how to bounce less

Waterfall enrichment means asking several contact data providers in turn until one finds a verified email: if the first can't match a person, the second tries, then the third, and you stop at the first success. It finds more emails than one source alone, but it only cuts bounces if you verify every result and never fill gaps by guessing.

Key points

• Waterfall enrichment queries providers in a fixed order and passes only the misses down the line.
• The second provider adds the most. Each one after that adds less while cost and setup keep rising.
• Catch-all emails can't be confirmed in advance, so label them risky and keep them apart.
• Vendor guidance puts a safe total bounce rate below 2 to 3%. Lower is always better.
• Record source, date and confidence on every field, because contact data decays.

What is waterfall enrichment?

Waterfall enrichment is a way of filling in contact data by querying several providers in a set order and stopping at the first verified match. Unify, a data vendor, defines it as "a data architecture where multiple B2B data providers are queried in a defined priority sequence."

Email enrichment is the step of adding a work email (and often other fields) to a record that only has a name and a company. A waterfall is simply email enrichment done with more than one source.

In practice it works like this:

1. You start with a list of people: name, company, maybe a LinkedIn profile.
2. You send everyone to your first provider.
3. Anyone the first provider can't match goes to the second.
4. Anyone still missing goes to the third, and so on.
5. Every email that comes back is verified before you use it.

Order matters. Put the source you trust most, or the one that costs least per match, at the top. Later providers only see the people earlier ones missed, so you don't pay twice for the same contact.

Is waterfall enrichment worth it?

Usually yes, for two or three providers, because the second source recovers a large share of the misses. Past three or four, the extra matches rarely justify the extra cost and complexity.

There is little neutral research here, so treat any numbers as estimates. As of October 2026, Unify's own analysis says a single provider typically returns matches for 55 to 70% of a contact list. It says "a second recovers 15-25% of misses" and "a third adds another 8-12%." Beyond four providers, it says gains "drop to 3-5% per additional source." The same page cites independent benchmarks with lower single-source rates, around 50 to 60%. These are a vendor's figures, so test on your own list.

Email finder accuracy is how often a found email is real and deliverable, not just how often a tool returns something. A source with a high hit rate and a high bounce rate costs you more than it gives.

Data decay is the rate at which contact records go out of date as people change jobs and mailboxes close. Unify's page, citing IndustrySelect, says "B2B contact data decays at roughly 22.5% per year."

Stale addresses can also hurt you twice. Spamhaus explains that some spam traps "were once-valid email addresses that ISPs have turned off." Mail to them "is rejected with a hard bounce for a period of time, often 12 months or more," and Spamhaus says hitting this kind of trap "is a significant red flag." An old list gives you bounces first, then traps.

How do you handle catch-all emails?

Treat catch-all emails as risky: verify what you can, keep them in a separate group, and send to them sparingly or not at all. Never mix them in with confirmed addresses.

A catch-all (or accept-all) domain is one whose mail server says yes to any address at that domain, then may bounce or discard the message later. That means a verification check can't tell you whether jane.smith@example.com is a real person.

The SMTP standard expects this. RFC 5321 says there may be cases where "an address appears to be valid but cannot reasonably be verified in real time," and that "In these situations, reply code 252 SHOULD be returned." A 252 is neither yes nor no.

Here is a simple way to handle them:

1. Label them. Give each result a status: verified, risky (catch-all or 252), or invalid.
2. Keep them apart. Put risky emails in their own segment so their bounces don't hide inside your main numbers.
3. Prefer a confirmed address. If a later provider in the waterfall returns a verified email for the same person, use that instead.
4. Send small. If you do email catch-alls, send to a small batch first and watch bounces closely.
5. Use another channel. For high-value people on catch-all domains, LinkedIn or a warm introduction is often safer.

Why guessing makes catch-alls worse

It is tempting to fill gaps by guessing from name patterns, such as jane.smith@, jsmith@ or jane@. On a catch-all domain every guess looks valid until it bounces after sending. Companies also don't all use one pattern, and every failed guess counts against your domain.

There is a data protection angle too. The UK regulator, the ICO, says organisations must "take all reasonable steps to erase or rectify inaccurate data without delay." A made-up address is not accurate personal data. A miss is better than a guess.

What bounce rate is safe?

Aim to keep your total bounce rate below 2 to 3%, and pause any campaign that goes over 5%. These are vendor guidelines, not rules published by Google or Microsoft, so treat them as a ceiling, not a target.

Email bounce rate is the share of sent emails that come back undelivered. Hard bounces (the address doesn't exist) are the ones that hurt most.

As of October 2026, Unify's page says "industry guidance is to keep total bounce rate below 2-3%," and that "If bounce rates are above 5%, your enrichment either lacks email verification or is returning stale data."

Google doesn't publish a bounce threshold, but its email sender guidelines are clear about what to do: "If messages start bouncing or start being deferred, reduce the sending volume until the SMTP error rate decreases." They also suggest you "Automatically unsubscribe recipients who have multiple bounced messages." For the full set of rules from Gmail and others, see our guide to email sender requirements in 2026.

These bands follow Unify's guidance above. Your own mailbox providers may react sooner.

Waterfall tools vs all-in-one tools: which fits you?

Build your own waterfall in a workflow tool if you want full control over providers and order. Use an all-in-one engine if you want predictable cost and less setup. Neither is better for everyone.

A workflow tool is a flexible platform where you connect data providers and design the enrichment steps yourself. Clay is the best-known example. As of October 2026, Clay's pricing page says data credits "are used when you purchase data from Clay's marketplace," while actions cover platform work such as running enrichments. It also says "If you bring your own API keys for third-party data, you skip Data Credit costs entirely," and "If an enrichment returns no result, you're not charged Data Credits or Actions." That last point suits waterfalls well, since many lookups miss.

An all-in-one engine handles enrichment as one step inside a bigger system, with the provider order and verification decided for you.

Pick a workflow tool if you have someone who enjoys building and tuning data flows, need niche providers, or want to bring your own API keys. Our piece on what a GTM engineer does covers that role. Pick an all-in-one engine if you want enrichment handled so you can focus on selling. For a direct look at both, see Clay vs Outbind.

What should you record for each field?

Store three things on every enriched field: the source that found it, the date it was found and last verified, and a confidence status (verified, risky or invalid). For example:

email: jane.smith@example.com | source: provider B | verified: 2026-10-14 | status: valid

This matches the ICO's guidance on accuracy, which says you should "accurately record the source of the information" and adds: "If you use the information for a purpose that relies on it remaining current, you should keep it up to date."

How do you set up a simple waterfall?

1. Pick two or three sources and put the most accurate first. Test each on a few hundred contacts you already know, and compare hit rate and bounce rate.
2. Pass only the misses down. Don't re-enrich people an earlier source already found.
3. Verify every result. Recheck even "verified" emails close to sending.
4. Never guess. No pattern fills.
5. Tag every field with source, date and confidence.
6. Track bounces by source. Move a provider down the order, or drop it, if its emails bounce more than the rest.
7. Re-verify before each campaign.

Where Outbind fits

Outbind builds enrichment into its client engine, so you don't design the waterfall yourself. It uses a verified pool first and other sources only for gaps, and records source, date and confidence on every field. It never guesses emails from name patterns. Plans start at $49 a month with a 7-day free trial, and sending never uses credits. See pricing for details.

FAQ

What is waterfall enrichment in simple terms?

It is asking several contact data providers in turn for the same person's email. If the first can't find it, the second tries, and you stop at the first verified match.

How many providers should a waterfall have?

Two or three is enough for most teams. Vendor figures from Unify show the second provider adds the most, and gains drop to a few percent per source beyond four.

Are catch-all emails safe to send to?

They are risky, because the server accepts any address and may bounce the message later. Keep them in a separate group, send small batches, and prefer a verified address or another channel.

What is a good email bounce rate?

Vendor guidance from Unify says to keep total bounces below 2 to 3% and to pause above 5%. Google publishes no exact number but tells senders to cut volume when bounces rise.

Does waterfall enrichment improve email finder accuracy?

It improves coverage, meaning more people get an email. Accuracy only improves if you verify every result, label risky ones and track bounces by source.

How often should you re-verify contact data?

Re-verify before each campaign, and recheck anything that has sat unused for a few months. Contact data decays as people change jobs.

Sources

• Unify: Waterfall enrichment for B2B contact data
• Google Workspace Admin Help: Email sender guidelines
• Spamhaus: Spamtraps, fix the problem not the symptom
• RFC Editor: RFC 5321, Simple Mail Transfer Protocol
• ICO: Principle (d): Accuracy
• Clay: Pricing