Enrichment provider

How enrichment providers fit a pipeline

An enrichment provider sits behind an API or a UI that maps your key to their graph. You send what you know. They return what they know. The contract is the input key, the field set, the latency, and the bill.

RichAPI frames data enrichment as filling empty fields on a person or company record from outside sources when you need them 1. Profile enrichment and Company enrichment each cost 1 credit per call and bill on any 2xx response, including an empty payload 1. That billing rule is part of what provider means in practice: you are paying for the attempt shape the vendor chose.

Waterfall enrichment is how teams use several providers without stopping at the first miss. Clay describes waterfall enrichment as searching sequentially across multiple tools until a valid match appears 2. At RichAPI, Email finder, Email verifier, and Phone finder run as waterfalls, and a call where no source returns data costs zero credits 2. Other endpoints still bill on 2xx even when empty 2.

Coverage is not universal. The same provider can look strong on US SaaS titles and weak on regional manufacturing contacts. That is why RichAPI's waterfall cost calculator treats hit rate as a property of the user's own list, not a vendor headline 3.

Providers also age. Fields returned today become data decay tomorrow. Store provider name and retrieval date next to every write so a later refresh knows what to trust 1.

A provider bake-off should freeze a sample, define blanks, and score match rate and fill rate under one rubric. Without that, you are comparing marketing slides.

Security and notice obligations travel with the provider choice. If personal data is enriched from other sources, plan how you will inform people under the notice rules summarized on the RichAPI enrichment page 1.

Provider contracts also hide rate limits and geographic coverage. A source that is excellent on US SaaS may return empty bodies for APAC manufacturing domains. Your bake-off sample must mirror the segments you will actually buy, or the winner will fail in week two.

Operational fit includes how receipts look. You need a provider id, a billed flag, and enough request identity to reconcile a credit dispute without opening a support roulette ticket 2. If the vendor cannot show that, score them lower even when raw fill looks fine.

People sometimes confuse search providers with enrichment providers. Search creates new rows. Enrichment fills rows you already chose to keep 1. Mixing those budgets is how teams overpay for contacts they will never email.

Contract length should not freeze a bad miss policy. If a provider bills every empty lookup, keep them off the first waterfall step even when annual pricing looks friendly 2. Put the honest source first. Use the expensive one only on residual blanks after you measure lift.

Share the bake-off packet with finance. Include sample size, blank definition, credit rule, and usable fill after verification. Providers that refuse to run on your frozen file are telling you the coverage claim will not survive daylight.

How to evaluate an enrichment provider

Start with the input key. A provider that only accepts a LinkedIn URL cannot enrich a CSV of emails until another step creates that URL 1. Key mismatch kills more bake-offs than field maps do.

Next read the miss policy. A waterfall provider that bills zero on empty results lets you add a backup step cheaply 2. A 2xx-billed provider charges even when the body is thin 1. Those rules change the math more than a one-point coverage gap.

Then measure on your file. The waterfall cost calculator states that hit rate belongs to your list and cannot be supplied by a vendor 3. Run the same frozen sample of several hundred rows through each candidate 3.

Watch freshness and overwrite behavior. A provider that replaces a good CRM title with a stale one raises fill rate and lowers trust. Track net-new fills separately from overwrites.

Finally, check ops fit for rate limits, receipts, MCP or REST access, and whether credits never expire on the plan you buy 3. A provider that is hard to audit will lose to a slightly narrower one you can prove.

Keep firmographics and person fields on separate scorecards when the provider is really two products behind one logo.

Ask for a delete and retention story. Enrichment providers hold derivatives of your keys. Know how long logs live and whether you can purge a customer who exercises deletion rights.

Re-run the bake-off when your ICP shifts. A provider that won last year on mid-market SaaS may lose this year on enterprise manufacturing. Treat provider choice as a living scorecard next to match rate.

Example of picking providers for title and email on one list

RevOps needs current titles and work emails for a large inbound lead cohort. Provider A takes LinkedIn profile URLs and returns title and company for 1 credit per call on RichAPI-style enrichment pricing 1. Provider B is used only inside an email waterfall where misses bill zero 2.

They freeze a few hundred rows. Provider A's title match rate is scored on rows that already have profile URLs. Provider B's email fill is scored only after verification. The waterfall cost estimate uses volume times hit rate times credits per hit on the miss-free path 3.

The runbook names each provider, the field it owns, and the credit rule. That is how an enrichment provider becomes a line item, not a vague data partner.

Three months later the same scorecard shows Provider A's titles drifting. Ops cuts that provider's share of the waterfall and raises the fresher source without rewriting the whole stack. That is provider management as a portfolio, not a one-time RFP.

The scorecard from that run becomes the template for the next cohort, so arguments about coverage start from the same denominator every time.

Enrichment provider vs waterfall enrichment

An enrichment provider is one source. Waterfall enrichment is an ordered chain of providers for a single value.