Waterfall Enrichment: What It Is, the Match-Rate Math and How to Run One
English
Waterfall enrichment queries several data providers in sequence and stops at the first verified hit. Learn the match-rate math, the cost per found record, the tools and how to run one.
On this page
- What is waterfall enrichment?
- Why a single provider caps out
- The match-rate math, with a worked example
- Cost per found record versus cost per attempt
- The verification step, and why it matters most for mobiles
- Where waterfalls are used
- 3 ways to run a waterfall
- How Overloop's phone number reveal uses a waterfall
- Questions to ask any waterfall vendor
- A note on GDPR and lawful basis
- Frequently asked questions
- Bottom line
Waterfall enrichment asks several data providers for the same missing field, one after the other, and stops at the first verified hit. This guide explains why that beats a single database, how to read the match-rate math, what a found record really costs, and the 3 ways to run a waterfall for emails, mobile numbers and company data.
Waterfall enrichment is a lookup method that queries several data providers in a fixed order for one missing field, such as a work email or a mobile number. The search stops at the first result that passes verification, and in most tools you pay only for that hit. Because every provider covers a different slice of the market, chaining 3 or 4 of them finds records that any single database misses.
The trade: more coverage per list, slightly more latency per record, and a pricing model you must read carefully. The 2 questions that decide whether a waterfall is worth it are "how many points of coverage does each extra provider add on my records" and "am I charged on attempt or on hit."
What is waterfall enrichment?
A waterfall is a sequence. You submit a prospect with whatever you already know, typically a name plus a company domain, or a LinkedIn profile URL. The system sends that input to provider A. If A returns a value that passes a verification check, the search ends and the value is written to the record. If A returns nothing, or returns something that fails verification, the same input goes to provider B, then C, and so on down the list.
Three properties define the model and separate it from simply buying 3 subscriptions:
- Sequential, not parallel. Providers are queried in order, and the order drives cost and speed.
- Stop at first verified hit. The waterfall takes the first answer that clears the verification gate, so provider quality early in the chain matters more.
- Pay on result. Most dedicated waterfall tools charge credits when a value is found and verified, not per attempt. Some charge on every provider call, which changes your cost per found record by a wide margin.
Why a single provider caps out
Every contact database is built from a different mix of sources: crawled public profiles, partner data, user-contributed address books, carrier feeds, registry filings. Those sources are not evenly spread across countries or job titles. One vendor may be strong on US mid-market SaaS directors and weak on German industrial buyers. Another may hold plenty of mobile numbers in the UK and almost none in France, where public records rarely include them.
So a vendor's headline coverage figure, when one is published at all, is an average over that vendor's customer mix, and on your list it can land far higher or lower. Enrich with one provider and you inherit its blind spots in full; the only fix is a second subscription and a manual re-run of the misses. A waterfall automates that re-run, and it makes the marginal value of each provider measurable: after a few thousand records you know how many points provider B adds on top of A, which is the number that should decide whether B stays in the chain.
The match-rate math, with a worked example
The concept that trips up most buyers is the difference between a provider's standalone coverage and its incremental coverage. Provider B might find 50 percent of your list on its own. But if most of those records were already found by provider A, B only adds the part A missed. The waterfall pays for increments, so increments are what you should evaluate.
Illustrative example, not measured data. The percentages are chosen to show the mechanics.
| Step | Records reaching this step | Found at this step | Points added | Hit rate on records reaching it | Cumulative coverage |
|---|---|---|---|---|---|
| Provider A | 1,000 | 550 | 55 | 55 % | 55 % |
| Provider B | 450 | 150 | 15 | 33 % | 70 % |
| Provider C | 300 | 80 | 8 | 27 % | 78 % |
| Not found | 220 | 0 | 0 | n/a | 78 % found, 22 % empty |
Two things in this example hold for most real waterfalls. The hit rate falls at every step, because each provider only sees the records the previous ones could not resolve: small companies, recent job changers, countries with thin public data. And the increments shrink fast. A fourth or fifth provider often buys 2 or 3 points, which may not justify the extra latency and one more data processor in your GDPR records.
Cumulative coverage = sum of increments so far
Step hit rate = records found at this step / records reaching this step
Provider order is a lever. If A and B have similar standalone coverage but A costs half as much per hit, put A first and B is only paid for the residue. If speed matters more than cost, put the highest hit-rate provider first. Some tools let you set a different order per country or per field, which is where the model gets useful.
Cost per found record versus cost per attempt
Vendors quote a price per credit. That number means nothing until you know how many credits one field consumes and whether credits are deducted on every provider call or only on a verified hit. Run the same 1,000 records through both billing models.
Illustrative example. Unit prices are invented to show the effect of the billing rule.
| Billing model | Billable events | Unit price (example) | Total spend | Records found | Cost per found record |
|---|---|---|---|---|---|
| Charged per attempt | 1,000 + 450 + 300 = 1,750 provider calls | $0.10 per call | $175 | 780 | $0.22 |
| Charged per verified hit | 780 hits | $0.15 per hit | $117 | 780 | $0.15 |
The per-hit price is 50 percent higher per unit and still cheaper per found record, because the 970 empty calls cost nothing. The gap widens as the list gets harder. When you compare vendors, convert everything to cost per found, verified record on a sample of your own data; nothing else is comparable.
Also read the multiplier before the price: a phone number usually costs 5 to 30 times the credits of an email, so a plan advertised as "1,000 credits" may be 1,000 emails or 33 mobile numbers.
The verification step, and why it matters most for mobiles
"Stop at first hit" only works if the hit is checked before the search stops. Without a verification gate, the first provider to return any string wins, including a stale number or a pattern-guessed email. Good waterfalls insert a check between each provider and the record.
For emails the check is the familiar one: syntax, MX records, then an SMTP-level test of whether the mailbox exists, with catch-all domains flagged as uncertain rather than valid. It is the same logic as a standalone email verification tool.
For mobile numbers the check matters more, because a wrong number costs a rep a call rather than a bounce. The useful checks are a carrier lookup (is this a mobile line, and in which country) and a line-status check (is the line active, or disconnected or ported). Providers that skip these show higher raw match rates and lower connect rates. When a vendor quotes a phone match rate, ask what verification a number passed to count as a match.
Where waterfalls are used
The method applies to any field that several providers hold with partial overlap. In B2B sales that is 3 fields in practice.
The original use case. Inputs are name plus domain or a LinkedIn URL. Increments per provider are moderate because email coverage is already high on English-speaking markets, so the waterfall earns its keep on SMBs and non-English countries.
Where waterfalls add the most. Mobile coverage varies sharply by country and provider, single-vendor match rates are lower than for email, and the per-record price is higher, so paying only on verified hits changes the economics. See the phone number finder page for how Overloop handles this.
Headcount, revenue band, industry, technology stack, funding. Here the waterfall often merges rather than stops: headcount from the first provider that has it, industry from the next. Same fallback logic, applied field by field.
3 ways to run a waterfall
1. Buy a dedicated waterfall tool
Clay, FullEnrich, BetterContact and Pipecorn sell the waterfall itself: you bring a list, they hold the provider contracts and the ordering logic. Entry prices below are as listed on each vendor pricing page on September 15, 2026. Several use sliders or calculators and displayed currency can vary by location, so confirm before buying.
Vendor-listed entry points, checked September 15, 2026. Not a ranking.
| Tool | Free tier | Entry paid plan (as listed) | Phone versus email credit cost | Billing rule |
|---|---|---|---|---|
| Clay | Free plan, up to 200 rows per table, multi-provider waterfalls included | Launch, starts at $167 per month, or $54 per month billed yearly, from 15,000 actions per month | Two meters: actions for platform work, data credits for data bought through Clay's 150+ providers; per-field cost depends on the provider chosen | Not stated on the pricing page; confirm per provider |
| FullEnrich | 50 credits, no card | Pro plan with 1,000 credits per month; the monthly price is set through a credit selector on the page, so read the displayed amount for your volume | 10 credits per phone, 1 credit per email | Charged only when data is found and verified |
| BetterContact | 50 credits | Starter $15 per month with 200 credits; Pro price computed by an on-page calculator, not published as a fixed figure; Enterprise from $799 per month | 10 credits per phone, 1 credit per email | Charged only when valid |
| Pipecorn | 150 credits | Build, $83 per month billed yearly, 3,000 to 6,000 credits per month; Grow $233 and Scale $583 per month billed yearly | 30 credits per phone, 3 credits per email | Never billed on bounces; credits roll over |
The fit is a revenue operations team that enriches lists outside its outreach tool, then pushes them into a CRM or sequencer. The cost is a separate contract, a separate credit balance to watch, and one more hop for the data.
2. Build your own with automation and provider APIs
You can assemble a waterfall in Zapier, Make or n8n: a trigger on a new row, a call to provider A's API, a filter on "empty result," a call to provider B, and so on, ending with a verification API and a write to the CRM. Teams do this when they already hold provider contracts or want full control over order and logging.
The trade-offs: each provider API is billed on its own terms, usually per call, so you lose the pay-on-hit advantage; you maintain the error handling when an API changes; every provider is a separate data processor to document; and latency is the sum of the chain plus the automation platform's polling interval. It works for a few hundred records a month with an engineer nearby, and poorly beyond that.
3. Use a platform with the waterfall built in
The third option is an outreach or prospecting platform that runs the waterfall behind one button and writes the result to the prospect record you are about to sequence. You give up picking providers and setting the order. You gain one credit balance, one data processor agreement, and no export-import step between enrichment and the first email. When enrichment volume tracks outreach volume, this is usually the cheaper path in total.
How Overloop's phone number reveal uses a waterfall
Overloop added phone number reveal in September 2026. When a user clicks "Find phone" on a prospect, or selects several prospects and runs the reveal in bulk, Overloop queries several phone data providers in sequence and stops at the first match. A revealed number costs 5 credits. Prospects with no match are not billed. The number lands on the prospect record next to the email, so the same contact can go into an email and LinkedIn sequence with a call step, and revealed numbers export to CSV for dialers or CRMs.
For scale: 1 credit sources a prospect from the advertised 450M+ contact database, 1 credit finds a valid email, 5 credits reveal a phone. The Starter plan is $69 per user per month with 250 credits and 1 email account, Growth is $99 per user per month with 500 credits and 3 email accounts, and Enterprise is custom with 1,000 credits, all with a 7-day free trial. A Growth user who spent every credit on phones would get 100 revealed numbers a month, the kind of arithmetic worth doing for any credit-based tool.
Overloop does not publish a phone match rate, for the reason explained above: it depends on your list's countries and personas, and one average would mislead as many buyers as it informs. Run the trial on a real slice of your list and count the hits. Details on the phone number finder page and the B2B contact database overview.
Questions to ask any waterfall vendor
Most of the price comparison happens in these answers, not on the pricing page.
- Charged on attempt or on hit? And what counts as a hit: any returned value, or one that passed verification?
- What is the credit multiplier per field? Convert the plan into "mobiles per month" before comparing.
- Can I re-enrich contacts already in my CRM? Some tools only enrich records sourced inside the tool. Ask whether re-running the same contact is billed again.
- Which countries are covered, and does provider order change by country? Ask about mobile coverage for your top 3 markets specifically.
- What verification runs before a value is accepted? Phones: carrier lookup, line status, or none. Emails: SMTP check and catch-all handling.
- What is the GDPR basis and where is data stored? Sub-processors, data residency, and how deletion requests reach the providers in the chain.
- How do results leave the tool? CSV export, CRM push, API, or only inside the vendor's interface.
- Do my finds become the vendor's inventory? Some tools feed found values back into their shared database.
A note on GDPR and lawful basis
Frequently asked questions
What is waterfall enrichment in simple terms?
It is a lookup that tries several data providers in order for one missing field and stops at the first result that passes verification. If the first provider has no work email or mobile number for a prospect, the second is asked, then the third. In most tools you pay only when a value is found.
How much does waterfall enrichment improve match rates?
It depends entirely on how little the providers overlap on your list. In the worked example on this page, a first provider at 55 percent grows to 78 percent after 2 more providers, but those figures are illustrative. Measure the increment each provider adds on a sample of your own records; that number, not a vendor average, tells you whether the extra provider is worth keeping.
Is waterfall enrichment cheaper than a single data provider?
Per record found, usually yes, when the tool charges only on verified hits, because empty lookups cost nothing. Per attempt, it can be more expensive. Convert every quote into cost per found, verified record on the same sample before comparing.
Does waterfall enrichment work for phone numbers?
Yes, and mobile numbers are where the model pays off most, because single-provider mobile coverage varies widely by country and the per-record price is higher. Insist on carrier or line-status verification before a number counts as a match. Overloop's phone number reveal runs several phone providers in sequence and charges 5 credits per revealed number.
Can I build a waterfall myself with Zapier, Make or n8n?
You can chain provider APIs with a filter on empty results and a verification step at the end. It gives full control over order and logging, but each API is billed on its own terms, you maintain the integrations, and every provider becomes a data processor you must document. It suits low volumes with an engineer available.
Is using a waterfall enrichment tool GDPR compliant?
The tool can be compliant as a processor, but the lawful basis for enriching and contacting a prospect is your responsibility as the controller. Review the vendor's sub-processor list and data processing agreement, document your legitimate interest assessment, and make sure deletion requests reach every provider in the chain. This page is not legal advice.
Bottom line
Waterfall enrichment is a simple idea with an outsized effect on mobile numbers. Evaluate it with 2 numbers, the coverage each provider adds on your own list and the cost per found, verified record, and ignore vendor averages. Buy a dedicated tool if enrichment is a separate function in your team, build one if volumes are small and control matters, and prefer a platform with the waterfall built in if the records go straight into a sequence anyway. Whichever route, write down who saw the data, because a waterfall by design shows it to more than one party.
Find the mobile, then run the sequence
Overloop reveals phone numbers with a built-in provider waterfall, 5 credits per found number, then puts the prospect straight into an email and LinkedIn sequence. 7-day free trial.
Try Overloop free
Nicolas Finet
CEO at Sortlist and Overloop. Writes about B2B prospecting, outbound systems and the operational tradeoffs behind sales software.