How Waterfall Enrichment Works

What is waterfall enrichment?

Waterfall enrichment is one lookup walked across several data providers in sequence, stopping at the first provider that returns a verified answer. Cleanlist walks a pool of 25+ providers per lookup and charges once for the answer, however many providers had to run to find it. This page explains the method. The product that runs it, with the provider pool, the coverage by field and the price per result, is Cleanlist's waterfall data enrichment.

The name describes the shape of the run. A record enters at the top, the first provider is asked for it, and if that provider holds nothing the record falls to the second, then the third, and so on until either a provider returns a value that passes verification or the pool is exhausted. Nothing about the sequence is visible to the person who asked for the contact. They upload a list and get a list back.

The reason the pattern exists is coverage. No B2B data provider holds every person in every market, and the gaps are not random: a provider that is strong on North American technology companies is often thin on European manufacturing, and a provider that is strong on senior titles is often thin below director level. Asking one provider is asking one opinion. Asking several in order, and stopping at the first that answers, is what turns several partial datasets into one usable one.

Waterfall enrichment is a routing strategy rather than a dataset. It describes how a lookup is executed, not who owns the underlying record, which is why the same term covers a team wiring four provider APIs together in a script and a platform that ships the sequence already assembled.

The same idea travels under several names, and they mean the same thing in practice: cascading enrichment, multi-provider enrichment, multi-source enrichment, provider fallback and, inside engineering teams, simply a fallback chain. If a vendor says it queries providers in priority order and bills you for the result rather than the attempt, it is describing a waterfall whatever it calls it.

How does a waterfall enrichment run actually execute?

Each field runs its own waterfall. Cleanlist normalises the input, orders the pool in cost order, calls providers one at a time, verifies each candidate answer before accepting it, stops at the first that passes, and bills once for the field.

Email and phone are separate runs on the same person, because the providers that are good at one are frequently not the providers that are good at the other. A record can therefore come back with a verified work email found on the second provider asked and a direct dial found on the ninth, or with an email and no phone at all.

Cost order matters more than it sounds. Running the cheapest provider that could plausibly hold the record first, and only escalating when it comes back empty, is what keeps a per-result price viable. A sequence that always calls the most expensive provider first would find the same answers and cost several times as much to operate.

Verification sits inside the loop rather than after it. A provider returning a string that looks like an email is not a hit. The address is checked against syntax, then the domain's mail exchange records, then the mailbox itself, and only an address that survives that is accepted and billed.

The four moving parts, named:

Input normalisation. Names, domains and company strings are cleaned to a common shape before any provider is called, so the same person is not missed by one provider over a formatting difference.

Provider order. Cost order, set by Cleanlist rather than by the buyer, so a lookup escalates only when the cheaper source came back empty.

Stop condition. The first verified answer. Providers below the hit are never called, which is why the price is the same whether the answer came from the first provider or the last.

Billing. Once per accepted field. 1 credit for a verified work email, 10 for a direct dial, 11 for both on the same contact.

Why does the order of the providers matter?

Order decides what a run costs and how long it takes, and it decides coverage only when the chain is allowed to stop early. A waterfall permitted to run to the end of the pool finds the same answer whichever provider happens to hold it. What changes is how many paid calls were made before that provider was asked.

The arithmetic is easiest to see with two sources. Put the one that answers six rows in ten first, and the run pays the second source on four rows in ten. Reverse them, and the run pays the first source on six rows in ten instead. Identical answers, different bill, and the gap compounds across a pool of 25+ and a list of several thousand.

Order decides latency for the same reason. Each escalation adds one provider's response time to that row, so the rows that needed the most work are the slowest rows in the job. That is the argument for running a bulk job asynchronously rather than holding a request open: a synchronous design times out on exactly the records that were about to succeed.

The case where order does change coverage is a capped chain. Teams building their own sequence usually cap it, at three or four sources, to keep a per-record budget predictable. Everything below the cap is never asked, so a record that only the sixth source holds is simply lost, and the order has quietly become a coverage decision rather than a cost one. Cleanlist does not cap the pool. The sequence runs until something verifies or the pool is exhausted, and because a miss costs nothing, escalation is billed to us rather than to the person who asked for the contact.

There is one more effect, and it only appears when you try to measure a running waterfall. Order decides which source gets credited for a field. If the first one answers most rows, the ones below it are rarely exercised, so a coverage audit run against a live waterfall measures the order and not the pool.

How do you choose the provider order in a waterfall?

Order by price ascending within a coverage band rather than by price alone, keep a separate order for each field, and measure on your own list rather than on a vendor's. The cheapest source is the right first ask only if it plausibly holds the record, because a cheap source with no coverage of your market turns every row into a wasted call and a slower run.

The rules that survive contact with a real list:

Segment before price. Split the pool by where each source is genuinely strong: region, seniority band, company size, and whether the field is an email address or a phone number. Order by price inside each band, not across the whole pool.

One order per field. Work email and direct dial are different markets. A single global order tuned for email will spend heavily to find phone numbers, which is the field where the money actually goes at 10 credits against 1.

Rank on verified survival, not on fill rate. A source that returns something for eight rows in ten and survives a mailbox check on half of them is worse than one that returns five and survives all five. Raw fill rate hides that difference, and the difference is your bounce rate.

Put the free steps above the first paid call. Normalisation and search itself cost nothing on Cleanlist, and deduplicating against records you already hold costs nothing anywhere, so anything resolvable without a provider call belongs at the top of the sequence.

Do not tune on fifty rows. A gap of a few points needs hundreds of records to separate from noise. Two hundred to five hundred rows drawn from one ICP segment is the smallest sample worth acting on.

Re-measure quarterly. Coverage and prices move, so an order tuned once at setup decays. Cleanlist maintains the sequence centrally for that reason, which is why it is not a customer-facing setting.

What a Cleanlist customer controls is the field rather than the sequence: email only at 1 credit, phone only at 10, both at 11. That is the decision that changes a bill. The order underneath it is maintenance work, and it is the part of a self-built waterfall that nobody wants to own in year two.

Why does waterfall enrichment beat a single data provider?

Because the gaps in one provider are not the gaps in the next. On the Cleanlist 500-Lead Enrichment Benchmark, 2026, a 25+ provider waterfall returned 98% verified work email and 85% direct dial across 500 stratified B2B leads. That is Cleanlist's own result on its own sample, and it tests no other vendor.

The phone column is the one to read. Direct dials are the harder of the two fields to find, because fewer sources hold them and the ones that do hold different people, so phone is the field where running several providers in sequence changes most what a calling team can do with a list.

Underneath the headline number, four things separate a sequence from a single source, and only the first is the one buyers ask about.

Coverage. Provider gaps are not random. They follow how the data was collected, so a source strong on North American technology companies is routinely thin on European manufacturing, and a source strong on senior titles is thin below director level. One provider is one collection method. A sequence is several.

Freshness. A licensed single source answers from a record written at its last refresh. A waterfall resolves the record at the moment you ask and verifies the address before returning it, which is the only way to catch a mailbox that closed last week.

Cost shape. Single-source contracts are usually flat: a seat or an annual license, priced the same whether a rep ran four hundred lookups or four. A waterfall is priced per returned record, so the invoice tracks usage and an unused seat costs nothing.

Failure mode. When a single source misses, the row is empty and stays empty until the next refresh, and nothing tells you whether the person is hard to find or the source is thin in that market. When a waterfall misses, every source in the pool missed, which is a far stronger statement about the list.

Coverage claims that do not name their denominator or their input list are not comparable to each other and should not be treated as if they were. The figures above name both: 500 stratified B2B leads, run once through the waterfall. No other vendor was run on that list, so no other vendor's figure on this page can be read against them.

There is a second effect that a coverage number does not show. A record that only one provider in the pool holds is a record your competitors calling the same market probably do not have, because they are asking one provider and it is not that one.

The honest limit belongs on our own page. A waterfall inherits the weaknesses of the sources inside it. If several of them license from the same upstream collection, adding them adds cost and not coverage, which is why a provider count on a homepage is a weak signal and why the only number that settles the question is a run against your own list.

What hit rate should you expect from waterfall enrichment?

Expect a range, and expect it to be a property of your list as much as of the pool. On the Cleanlist 500-Lead Enrichment Benchmark, 2026, a 25+ provider waterfall returned 98% verified work email and 85% direct dial across 500 stratified B2B leads. That is Cleanlist's own result on its own sample, and it tests no other vendor. Published claims elsewhere in this category are lower and much wider, and the spread between them is the useful finding.

Google's AI Overview for waterfall enrichment, collected on 2026-09-14, states that single databases usually find 40% to 60% of contact details and that a waterfall setup pushes match rates to 80% or higher. A page cited inside that same answer opens by saying traditional providers deliver about 30% mobile and email coverage. Three numbers, three different denominators, and not one of them says which list it was measured on.

Five things move your own rate more than the choice of vendor does:

Seniority. Director and above is the best covered band in almost every source. Individual contributors below that line are where pools diverge.

Region. North American contact data is the deepest market. European mobile numbers are the field where the difference between one source and a sequence is largest, and the field where a coverage claim measured in the United States tells you least.

Company size. Very small companies have thin public records, so a list aimed at firms under twenty people returns a lower rate than the same job titles at firms of five hundred.

Field. A verified work email is the easier ask. A direct dial is where a pool earns its cost, and it is the column to read first when two vendors post similar headline numbers.

Input quality. A row carrying a LinkedIn URL or a company domain resolves far more often than a row carrying a person's name and a company string typed by hand. Enrichment cannot repair an ambiguous input, and a surprising share of disappointing hit rates are an input problem wearing a vendor's name.

Getting your own number is cheap. Take three hundred rows from the segment you actually sell into, run them once, and divide the rows that came back with a verified work email by three hundred. Because a miss costs nothing, that test bills you only for what it found.

What does waterfall enrichment cost per record?

Cleanlist charges per accepted field rather than per provider call: 1 credit for a verified work email, 10 for a direct dial, 11 for both, 0.5 for validating an address you already hold, and 0 when the whole pool comes back empty. People Search and Company Search cost nothing at all.

Because the price is per result, the arithmetic is the same every month. Starter is $79 for 1,500 credits, which is 1,500 verified work emails or about 136 fully enriched contacts at 11 credits each. Pro is $229 for 5,000 credits and Scale is $599 for 15,000. Annual billing takes 25% off every paid plan, and an extra seat is $20 a month.

The line worth reading twice is the last one. A lookup that walks all 25+ providers and finds nothing costs zero credits. That is the difference between paying for data and paying for attempts, and it is why a pool this deep is affordable to run: escalation is free to the buyer and expensive only to us.

Credits come out of one shared team wallet on every plan, including Free, so a rep who runs four hundred lookups and a rep who runs four spend from the same balance and nobody buys a second subscription to try it.

Which data providers are in the Cleanlist waterfall?

More than 25, and the pool includes Wiza, Prospeo, Findymail, LeadMagic, Datagma, Crustdata and Lusha alongside the rest. Cleanlist holds the contracts, the API keys and the rate limits, so a customer buys one plan rather than 25.

The commercial argument for a managed pool is not that any one of those providers is bad. It is that a team wiring them together itself signs 25 agreements, holds 25 sets of credentials, reconciles 25 invoices against one month of usage, and rebuilds a parser every time one of them changes a response shape. That work never finishes, and it is not the work a revenue team was hired to do.

The sequence is ours to set and to keep tuned. Provider coverage moves, prices move, and a provider that was the right first call for European mobiles last quarter may not be this one. Tuning the order centrally is the part of the product that keeps working after the integration is done.

How is waterfall enrichment different from licensing a contact database?

A licensed database sells you access to one company's stored records for a fixed annual fee. A waterfall spends nothing until a lookup is asked for, queries several sources at that moment, and verifies the answer before it is delivered.

The two models fail differently. A database's weakness is freshness: a record sits in it between refresh cycles, and B2B contact data decays at roughly 2.1% a month and 22.5% a year according to Cognism's published figures, with email addresses the fastest field to go. A waterfall's weakness is that it answers only the question you asked. You cannot browse it the way you can browse a database you licensed.

The cost shapes are different too. A licensed database charges the same whether a seat used it heavily or not at all, which is why per-seat data spend tends to drift away from usage over a year. A waterfall charges per returned record, so the invoice tracks what the team actually looked up.

Neither model is universally right. A team that wants to sit inside a UI and explore a market of 300 million stored contacts is buying a database, and should. A team that has a list, a CRM or a search result and wants the missing fields filled and verified is buying a waterfall.

Does the waterfall verify data, or only find it?

It verifies before it accepts. Every candidate email is checked against RFC 5322 syntax, then the domain's MX records, then the mailbox itself over an SMTP handshake, with nothing sent to the address, and an address that fails is not counted as a hit and not billed.

Risk is reported rather than hidden. A catch-all domain that accepts every address is returned flagged as accept_all instead of green, a role address such as info@ or sales@ is labelled as one, and a disposable domain is marked and dropped. Collapsing those three into a single green tick is the industry habit this product exists to argue against, because each of them needs a different decision from the person about to send.

Validating a list you already hold is the same machinery at 0.5 credits an address, without the enrichment. That is the cheaper first move for a team whose problem is bounce rate rather than coverage.

The verdicts, and what each one means for the person about to send:

Deliverable. The mail server confirmed the mailbox exists. Safe to send.

Catch-all. The domain accepts everything, so a yes proves nothing. Flagged, never guessed.

Invalid. The mail server refused the address. Never charged, never synced.

Disposable. A burner domain. Marked and dropped.

Role address. info@, sales@ and the rest. Labelled so you can route it rather than treat it as a person.

What happens when the waterfall finds nothing?

The row comes back marked as a miss and costs zero credits. No provider in the pool is billed to the customer for an empty answer, and no placeholder or pattern-guessed address is written in to make the run look complete.

That last part is the one to check when comparing tools. A guessed address in the shape firstname.lastname@company.com is trivial to generate and will pass a syntax check, so a vendor that fills gaps with guesses can report a coverage number close to 100%. It shows up later as bounces, and bounces are charged to your sending domain rather than to the vendor.

A miss is also information. On a list where a quarter of the rows come back empty, the useful next question is usually about the list rather than about the enrichment: the ICP may be aimed at companies too small to have public contact data, or at a market where the pool is genuinely thin.

What breaks in a waterfall enrichment setup?

The failures are quiet ones. A waterfall rarely throws an error. It returns fewer rows than it did last month, or it returns rows that look correct and bounce three weeks later, and both read as normal variation until somebody measures.

Nine failure modes, in roughly the order they show up:

A source goes empty for a region. Coverage shifts upstream, the source stops answering for a market it used to cover, and the only visible signal is a rep saying the data got worse. Nothing logs an error, because an empty answer is a valid answer.

A response shape changes. A provider renames a field, the parser stops reading it, and that source silently contributes nothing while still being called and still being billed. This is the most common reason a self-built sequence decays.

Rate limits turn into retries turn into duplicate charges. A throttled call retried without an idempotency key can bill the same lookup twice, and the duplicate is invisible inside a monthly total.

Verification is skipped or sold separately. A returned string that was never checked at the mailbox is a candidate rather than a result. Counting it as a hit inflates the fill rate and moves the cost onto your sending domain.

Gaps are filled with pattern guesses. The tell is the gap between the fill rate and the share that survives an independent check, and a pool that guesses shows a wide one.

Catch-all domains are reported as deliverable. The domain accepts every address, so acceptance proves nothing about the mailbox. Collapsing that into a green tick makes a decision on your behalf without telling you a decision was made.

Billing is per provider call rather than per accepted field. The price of a record then varies with how hard it was to find, which makes a budget unforecastable and gives the vendor a reason to stop early or to guess.

The order is never re-tuned. Prices and coverage move every quarter. A sequence set at implementation and left alone drifts into paying more for less, slowly enough that nobody attributes it to the sequence.

Two chains run on the same row. A CRM sync that re-enriches a contact the list already enriched pays twice for one answer. Deduplicate before the run rather than after the invoice.

Four of those nine are settled by a vendor's design rather than by your operations: whether verification sits inside the loop or is sold separately, whether billing counts provider calls or accepted fields, whether a miss costs anything, and who maintains the order. Cleanlist settles all four the same way on every plan, including Free. The failure nobody can remove is a thin market. If the pool genuinely does not cover the people on your list, the run comes back empty and correct, and no amount of sequencing changes that.

Does waterfall enrichment work for company data as well as contacts?

Yes, and the firmographic side behaves differently enough to be worth separating. Company fields are attributes of an organisation rather than of a person, they decay far more slowly than an email address, and on Cleanlist an enriched record comes back with up to 180 firmographic properties while Company Search itself costs 0 credits.

The waterfall still matters here, because company coverage is uneven in the same way contact coverage is. A provider with good headcount and industry data on North American software companies is often weaker on privately held European manufacturers, so the same sequence-and-stop pattern applies to a field like employee count.

What separates the two in practice is the refresh cadence. A person changes jobs and their work email dies the same day, which is why contact fields need reverification. A company's founded year never changes, its industry rarely does, and its headcount moves on a quarterly rather than a daily scale. A team that reverifies contacts monthly and firmographics quarterly is usually spending in the right proportion.

The boundary to know before you buy: the stored company record is the descriptive set, name, domain, headcount, industry, headquarters location and founded year among the rest. Signals about a company's current situation, funding events, hiring activity, tech stack or recent news, are produced by an AI research column at the moment you ask for them rather than read out of a stored field, and they are priced separately.

Can you build a waterfall enrichment sequence yourself?

Yes, and teams with an engineer to spare do. The build is a weekend and the maintenance is permanent, which is the part that decides it.

The moving parts are the same for everybody: a commercial agreement and a key per provider, a normalisation layer so a name and a domain reach every API in the shape it expects, per-provider rate limiting and retry, a parser per response format that has to be repaired whenever a provider ships a change, a verification stage so a returned string is not trusted on sight, deduplication when two providers answer with different values for the same field, and a reconciliation job that matches 25 invoices against one month of internal usage.

The honest test is whether somebody owns it as part of their job. A sequence nobody owns drifts: a provider quietly goes empty for a region, the coverage number falls, and the first sign is a rep complaining that the data got worse.

There is a real case for building. If enrichment is inside your own product rather than inside your go-to-market, if you have negotiated volume rates a reseller cannot match, or if your legal position requires direct contracts with each data source, the build is the right answer and the maintenance is a cost you were always going to carry.

How large a list can you run, and does it run synchronously?

A list holds 2,500 leads on any paid plan and 100 on the Free plan, and a bulk run is asynchronous. Enrichment jobs return a workflow_id that you poll until the results are ready, rather than holding a request open.

Rows enter a list from a CSV upload, a CRM import, People Search and Company Search, or the Chrome extension, so a list does not have to start life in a spreadsheet. CSV upload is a paid-plan feature, and CRM import arrives with the two-way sync on Pro and Scale.

Asynchronous is the right shape for a waterfall specifically. A single lookup that escalates through several providers takes as long as the providers it had to ask, so a synchronous API would spend most of a large run holding connections open and timing out on exactly the records that needed the most work.

How do you run waterfall enrichment on a list you already have?

Six steps, and the third is the one teams skip. Get the rows into a list, choose the field before you run, test on a sample, read three numbers off the sample, run the rest asynchronously, then re-verify on a cadence.

1. Get the rows in. A Cleanlist list fills from a CSV upload, a CRM import, People Search or Company Search, or the Chrome extension. Search costs 0 credits, CSV upload is a paid-plan feature, and CRM import arrives with the two-way sync on Pro and Scale. A list holds 2,500 leads on any paid plan and 100 on the Free plan.

2. Choose the field. Email only is 1 credit, phone only is 10, both are 11. On a 2,000-row list that is the difference between 2,000 credits and 22,000, an elevenfold swing in what the job costs. It is the decision that decides the invoice, and taking it by default is how a month's allowance disappears in an afternoon.

3. Sample first. Two to three hundred rows from one segment, run once. A miss costs nothing, so the sample bills only for what it found, and it answers the question no published coverage figure can: whether this pool covers your market.

4. Read three numbers, not one. Rows that returned a verified work email, rows that returned a direct dial, and credits actually spent. A single fill rate hides which of the two fields you are paying for.

5. Run the rest asynchronously. A bulk job returns a workflow_id you poll rather than an open request, because a sequential lookup takes as long as the providers it had to ask and the hardest rows are the slowest ones.

6. Re-verify on a cadence. Contact fields decay at roughly 2.1% a month and 22.5% a year on Cognism's published figures, so a list enriched in January and mailed in April is a different list. Validating an address you already hold is 0.5 credits, half the price of enriching it again. Firmographics move on a quarterly scale rather than a daily one and do not need the same cadence.

The finished rows leave through a two-way sync with HubSpot, Salesforce or Pipedrive, a one-way push to Outreach, Salesloft or Lemlist, a CSV export on every plan, or the REST API and the MCP server. Pushing a lead to a CRM is 0.2 credits.

Where does waterfall-enriched data land?

Into a Cleanlist lead list first, then out. Two-way sync with HubSpot, Salesforce and Pipedrive on Pro and Scale, one-way out to Outreach, Salesloft and Lemlist on the same two plans, CSV export on every plan, plus the REST API and the MCP server.

Pushing a finished lead to a CRM costs 0.2 credits. The public REST API is a Pro and Scale feature and authenticates with clapi_ Bearer keys and OAuth scopes, and every request returns a signed cost quote before any credits are spent, so an integration can decline a job it thinks is too expensive before it runs. The MCP server is available from Starter upward, which is what lets an assistant such as Claude run enrichment directly. Both are held back during the 14-day trial.

The direction of each integration is worth stating plainly, because it is the thing buyers assume rather than check. HubSpot, Salesforce and Pipedrive read and write. The three sequencers receive rows and send nothing back.

How do you evaluate a waterfall enrichment vendor?

Run your own list through it and count, because every published coverage number was measured on somebody else's input. Take 200 to 500 real records from your own ICP, run them through each vendor unchanged, and compare four things: how many rows came back with a work email, how many with a direct dial, what each run cost, and how many of the returned emails survive an independent verification pass.

That last check is the one people skip and the one that separates vendors. Export the returned addresses, verify them somewhere other than the tool that produced them, and see how many hold up. A vendor that pattern-guesses gaps will post a high fill rate and a low survival rate, and the difference is bounces you pay for later.

Five questions worth asking before the test, because the answers change how you read the result:

Are misses billed? If a lookup that returns nothing still costs money, a deep pool becomes expensive rather than free to escalate, and the vendor has an incentive to guess.

Is billing per provider call or per accepted field? Per call means the price of a record varies with how hard it was to find, which makes a budget impossible to forecast.

Is verification inside the run or sold separately? An address that was never checked at the mailbox is a candidate, not a result.

Are catch-all and role addresses reported honestly? A vendor that returns a catch-all as green has moved a decision onto you without telling you.

Who tunes the provider order, and how often? A pool that was tuned at signup and never again decays quietly as provider coverage shifts.

One caveat about our own position. Cleanlist does not hold SOC 2 Type II or ISO 27001 certification, and if your security review requires either of them, that is a real reason to buy elsewhere and worth finding out on day one rather than in procurement.

How do you measure whether a waterfall is working?

Four numbers, measured on the same batch, on a fixed cadence: fill rate, verified survival rate, cost per verified record, and bounce rate thirty days later. Each one alone can be moved without improving anything, which is why they are only useful together.

Fill rate. Rows that came back with a value, divided by rows submitted. The denominator is rows submitted, including every row that came back empty. A fill rate quoted against rows where data was found is always 100% and always meaningless.

Verified survival rate. Of the addresses returned, the share that still passes a mailbox check run somewhere other than the tool that produced them. This is the number that separates a pool which finds data from a pool which generates it, and it is the check most buyers skip.

Cost per verified record. Total spend divided by the count of records you can actually use, never by the count of lookups you ran. On Cleanlist a verified work email is 1 credit, roughly 5 cents at the Starter monthly rate, and a lookup that found nothing adds nothing to the numerator, so the two definitions converge. On a tool that bills per attempt they do not.

Bounce rate at thirty days. The only measure that reads the whole chain, because it catches the addresses that passed every check at run time and died between the run and the send. Contact data decays at roughly 2.1% a month on Cognism's published figures, so part of any bounce rate is arithmetic rather than a defect in the data you bought.

Two method rules make those four numbers comparable month to month.

Measure per field, never per row. A row counted as enriched because it picked up a job title tells you nothing about the email column, and a row-level rate is the easiest figure in this category to flatter.

Hold the input fixed. Re-running last quarter's test against a different list measures the list. Keep one stratified sample, re-run it, and change one variable at a time. That is the discipline behind the 500-lead benchmark described earlier on this page.

If a vendor publishes a rate without naming its denominator, ask for the denominator before comparing that rate with anything.

When is waterfall enrichment the wrong tool?

When the field you need is not a contact field. A waterfall finds and verifies who somebody is and how to reach them. It does not tell you that they are in a buying cycle, what software they run, or what happened on their last earnings call.

Three cases where a different purchase is the right one. If your motion depends on intent signals, a platform built around them, such as ZoomInfo or Cognism, is what you are actually shopping for. If you want to compose your own multi-step research pipelines and have somebody who will own them as a job, Clay is built for that. If you need to browse and slice a large stored database as a research surface rather than fill gaps in a list you already have, license a database.

There is also a boundary inside Cleanlist worth knowing before you buy. The stored fields are the contact and company record: work email, direct dial, LinkedIn URL, title, seniority, department, company name, domain, headcount, industry, headquarters location and founded year. Anything past that set, funding events, hiring signals, tech stack or recent news, is produced by an AI research column at the moment you ask for it rather than read out of a stored field, and it is priced separately.

The shortest way to describe the boundary: Cleanlist is an orchestration layer over other people's data rather than a data vendor holding a database of its own.

What does waterfall enrichment mean in Apollo, Clay or ZoomInfo?

All three run one, and the difference is where the provider pool lives and how the meter counts. Apollo builds the waterfall into its own platform and bills it from the same credit pool as everything else: 1 credit for an email and 8 for a phone number, on a Basic plan at $65 a month or $49 a month billed annually, with 30,000 credits per seat granted upfront for the year and a free plan at 75 credits per seat per month. Prices here come from the Cleanlist B2B data pricing index, fetched 2026-09-01.

Clay is an orchestration builder rather than a fixed pool. Launch is $185 a month, or $167 billed annually, and it runs two meters, Actions and Data Credits, selected and priced independently. The $167 headline is $54 of actions plus $113 of data credits. What a verified email costs is set per marketplace provider and published inside the app rather than on the pricing page, which is why Clay cannot be reduced to a single cost per 1,000 verified emails at all.

ZoomInfo runs its waterfall in parallel inside GTM Studio and publishes no purchasable price. Its pricing page returned HTTP 403 to two user agents and a press-and-hold bot challenge in headless Chromium when we tried to read it on 2026-09-01.

Cleanlist: best for teams paying per seat for data they use unevenly. It runs waterfall enrichment across 25+ providers and charges 1 credit per verified email and 10 per direct dial, with search costing nothing, starting at $79 a month.

Is a waterfall sequential or parallel, and does the order change the price?

Sequential and parallel are the two shapes, and the choice decides who absorbs the cost of the providers that did not answer. A sequential waterfall asks one provider at a time and never pays for the providers below the hit. A parallel waterfall asks several at once, returns faster because nothing waits in line, and pays for every call it made whether or not it needed them. That second cost does not disappear. Somebody absorbs it, either the vendor inside its margin or the buyer inside the credit price.

Cleanlist runs sequential, per field, in cost order set centrally rather than by the buyer. That is why a lookup which escalates to the ninth provider costs the same 1 credit as one that stopped at the first: the escalation is billed to us and not to the person who asked for the contact.

The honest version of the trade is that parallel is the faster shape and it has a real argument behind it. ZoomInfo makes that argument publicly, describing its GTM Studio waterfall as querying 25+ vendors simultaneously and returning a best-match chosen by confidence rather than the first match found (pipeline.zoominfo.com/operations/waterfall-enrichment, read 2026-09-07). Best-match is a genuine advantage on the fields where providers disagree. It is worth less on a verified work email, where an address either survives an SMTP check or does not.

What happens when two providers return different values for the same person?

Most of the time nothing conflicts, because within one field the run stops at the first verified answer and the providers below it are never asked. Conflicts appear only on the fields where more than one provider answered in the same pass, and there Cleanlist resolves them with per-field provider weighting and recency rather than with one global ranking of sources.

Three separate things get confused under the word deduplication, so they are worth separating.

Normalising the input. This happens before any provider is called. Names, domains and company strings are cleaned to a common shape, so the same person is not counted twice and not missed by one provider over a formatting difference.

Deduplicating across providers. The stop condition already does this. Email and phone are separate runs on the same row, so a verified work email routinely comes from one provider and the direct dial from another, and neither one had to be reconciled against a competing answer.

Reconciling a real disagreement. Title, headcount and industry are the fields where several sources hold a value at once. There the weighting is per field and recency counts, so a job title from a source that refreshes titles often can win on the same record where a different source won the email address.

A vendor that cannot say which of those three it does is describing marketing rather than mechanism, and it is a fair question to ask in a trial.

Why does escalating through 25+ providers not cost 25 times as much?

Because the buyer pays per accepted field and Cleanlist pays per provider call. Escalation is a cost we carry, and that is the only reason a pool this deep is affordable to point at a whole list.

The arithmetic is checkable. Starter is $79 a month for 1,500 credits and a verified work email is 1 credit, so 1,000 verified work emails cost $52.67 at monthly billing and $39.33 at annual billing, which is $59 a month. A direct dial is 10 credits, so 1,000 direct dials cost $393.33 at the annual rate. All four figures are computed in the Cleanlist row of our own published pricing index, b2b-data-pricing-index-2026-09.csv, fetched 2026-09-01.

The part a vendor page usually leaves out is that the rate is not the cheapest in the category. On that same index Cleanlist is 15th of the 22 vendors publishing a comparable monthly rate, with 14 cheaper per 1,000 verified emails, and 15th of 23 on annual billing. Clearout is $6.50, Skrapp and Smartlead are $19.50, Dropcontact is $19.75, Apollo is $26.00 and Snov.io is $39.00. Above us, FullEnrich is $55.00 and BetterContact is $75.00.

What the higher rate buys is the part a per-1,000 figure cannot show: one contract instead of 25, verification inside the run rather than sold as a second product, and a lookup that finds nothing costing zero.

Is waterfall enrichment GDPR compliant?

Compliance is a property of how the data is used rather than of the method, and the lawful basis for B2B contact data in the EEA and the UK is normally legitimate interests under Article 6(1)(f) rather than consent. That basis is not automatic. It requires a documented Legitimate Interest Assessment weighing your interest against the individual's rights, and an objection route that actually works, because an Article 21 objection to direct marketing is close to absolute.

The waterfall-specific point, which the rest of this category does not make: a waterfall multiplies the number of upstream sources sitting behind one record. Every provider in the pool is another collection with its own notice and its own lawful basis, so the subprocessor list is the first document a data protection review should ask for, and the answer should be a document you can read today rather than a promise. Cleanlist discloses the providers in the waterfall as subprocessors in its privacy policy.

Cleanlist collects none of this data from the individuals it describes, operates no crawler and holds no database of its own. It licenses from commercial providers, each responsible for the notice and lawful basis on its own collection.

None of this is legal advice. Treat it as the questions to take to your counsel rather than the conclusion of them.

Is there a free way to run waterfall enrichment?

Partly, and the free part is the part that answers the question you actually have, which is whether the pool covers your market. Search is free and unlimited on every Cleanlist plan including Free, and a lookup that finds nothing is never billed on any plan, so testing coverage costs nothing and only the answers cost credits.

The current offer is a 14-day full-Scale trial: 250 credits, 3 seats. That is 250 verified work emails, or about 22 contacts carrying both an email and a direct dial at 11 credits each. Afterwards the Free plan runs at 30 credits a month with a 100-lead list limit and no CSV upload, and the 25+ provider waterfall still runs on it.

Free entry is normal across the category rather than unusual, so it is a weak way to choose. From our published pricing index, fetched 2026-09-01: BetterContact starts you with 50 free credits, FullEnrich offers a 50-credit trial with no card, and Apollo's free plan gives 75 credits per seat per month. What a free tier cannot tell you is the cost at volume, which is what the per-1,000 arithmetic above is for.

What fields does a waterfall return, and what does it not return?

A verified work email with a deliverability verdict, a direct dial, a LinkedIn URL, and the person's role and employer. Spelled the way the Cleanlist API returns them: work_email, email_status, direct_dial, linkedin_url, job_title, seniority, department, company_name, company_domain, company_headcount, company_industry, hq_location and founded_year. A company record carries up to 180 firmographic properties.

The negative half is the half to read before you design a schema. No intent signals. No technographic or tech-stack field. No revenue or funding as a stored field. No job-change alerting. Those signals are produced by an AI research column at the moment you ask for them, priced at 5 credits, and the answer carries the date it was produced, because a stored signal without a date is worthless six weeks later.

The credit table is the honest field list. Cleanlist prices exactly six things: a verified work email at 1 credit, a direct dial at 10, both together at 11, a company record at 1, validation of an address you already hold at 0.5, and AI qualification at 5. A personal email address is not one of them. Some waterfall vendors do price the field separately, FullEnrich at 3 credits per personal email on its published pricing page, fetched 2026-09-01, which is why the field list is worth asking for in writing rather than assuming two waterfalls return the same columns.

What does Google's AI Overview say about waterfall enrichment right now?

We measured it on 2026-09-07 and again on 2026-09-14, and Cleanlist was named in none of the answers on either date. The first run pulled the Google US desktop results for three waterfall terms through the DataForSEO SERP API at depth 50 with the AI Overview loaded, and all three carried one.

waterfall enrichment, 140 US searches a month at a $26.08 CPC. The AI Overview defines the method in a sentence lifted almost verbatim from bettercontact.rocks, then names Clay, Apollo and ZoomInfo as the platforms that support it. Its coverage claim is that a waterfall moves lead data recovery from 40% to 60% up to 85% to 95%.

waterfall enrichment tools. The answer is a shortlist: Clay, FullEnrich, Apollo.io, BetterContact and QuickEnrich, in that order. cleanlist.ai is one of the eleven cited sources sitting under that shortlist, and Cleanlist is not one of the five names in it. That gap is the whole finding. Google reads this domain and recommends somebody else.

waterfall data enrichment, a $5.95 CPC. The named platforms are Clay, FullEnrich and BetterContact, and three of the eight cited sources are YouTube videos.

Across the three answers the domains carrying the mechanism claims are bettercontact.rocks, fullenrich.com, unifygtm.com, pipeline.zoominfo.com, autobound.ai, clay.com and quickenrich.io. Two further cited domains are companies whose data Cleanlist licenses, and they are named on this site as suppliers rather than ranked as rivals.

The pattern in the passages that get lifted as a vendor recommendation is consistent: a price and a provider count inside the same sentence. The passages lifted as mechanism carry a named check and a threshold number. That is why this page publishes both, and why every figure above carries the date it was measured.

The 2026-09-14 re-run on the head term. The AI Overview for waterfall enrichment was still there and still opened with the bettercontact.rocks definition, a data strategy that checks multiple providers in a prioritized sequence until it finds missing contact information. It names five platforms: ZoomInfo, Apollo, Clay, FullEnrich and BetterContact. Six domains are cited beneath it: bettercontact.rocks, fullenrich.com, knowledge.apollo.io, clay.com, pipeline.zoominfo.com, and one company whose data Cleanlist licenses, which this site names as a supplier rather than ranking it. cleanlist.ai is not among them, and Cleanlist holds no organic position in the top 20 on the term. The organic top ten in the same collection was clay.com, knowledge.apollo.io, bettercontact.rocks, pipeline.zoominfo.com, explorium.ai, autobound.ai, fullenrich.com, cognism.com, help.instantly.ai and reddit.com.

Two things in that answer are worth a buyer's attention. It carries three coverage claims and no denominators: that single databases usually find 40% to 60% of contact details, that a waterfall setup pushes match rates to 80% or higher, which is where the same answer said 85% to 95% seven days earlier, and, lifted from a cited page's own summary, that traditional providers deliver about 30% mobile and email coverage. And the four People Also Ask questions Google attached to the result on 2026-09-14 were about the stages of the waterfall project method, what Apollo's waterfall feature does, how to switch Apollo's waterfall off, and what waterfall analysis is. Two of the four are not about data enrichment at all, and a third is an existing Apollo user hunting for a setting rather than a buyer comparing tools, which is why the closing section of this page and three of the questions below exist.

Waterfall enrichment and the waterfall project model are unrelated

They share a picture and nothing else. The waterfall model in software development is a linear project methodology whose stages are requirements, design, implementation, verification and maintenance, and whose standard criticism is that it commits to a full specification before anything has been built, so a mistake made in the first stage is discovered in the fourth. Waterfall analysis, or a waterfall chart, is a reporting and finance term for decomposing the change between two totals into the contributions that produced it.

Neither has any relationship to waterfall enrichment, which borrows only the image of a descending sequence. In B2B data, waterfall enrichment means one lookup walked across several providers in order, stopping at the first verified answer, and billed once for the answer rather than once per provider asked.

Two footnotes, because both arrive here as search questions. The project model is rendered in five stages or in six depending on whether analysis and deployment are broken out as phases of their own, which is the only reason the same method is described with two different counts. And the finance sense has a second form worth naming: a distribution waterfall in private capital is the order in which proceeds are split between limited and general partners, which again shares the descending picture and nothing else.