TL;DR
- People Data Labs is a candidate for matching an email to a person record and structured attributes; assess match confidence and field bundles against your qualification rules.
- Dropcontact offers asynchronous B2B contact enrichment. Verify which extra inputs improve the result and how your application handles a pending request.
- Tomba documents person and combined person-company enrichment from an email, including a distinct error object for no match or unsupported inputs.
- Apollo offers people enrichment and match-confidence fields. Its credit and optional phone-waterfall behaviour should be priced separately from a basic person lookup.
- Protect information supplied by the buyer or a sales rep. An enrichment match should not automatically overwrite a manually verified CRM field or establish the person's current employer.
What can a work email establish?
An inbound form may contain only a name, email address and a short question. An enrichment API can use that input to look for a person or company record and return context such as role, employer or company size. That context can help route the enquiry, but it is a proposed match, not a replacement for what the person actually told you.
A work domain can point to a company, but it does not prove the sender's current role, buying authority or the correct subsidiary. A personal email domain does not identify an employer at all. A good workflow treats ambiguous or missing results as unknown and lets the buyer proceed without forcing a guess.
This guide starts with an existing email address. Finding an unknown address from a name and domain is a different job, and verifying whether an address may receive mail is different again. We compare the APIs from vendor documentation; no provider accuracy test has been performed. The broader B2B data API guide covers a wider selection of company and person endpoints.
Email enrichment APIs at a glance
| API | Starting use case | Important response or workflow distinction | Buying question |
|---|---|---|---|
| People Data Labs | Structured person match and attributes | Match strength and person fields must be interpreted for the use case | Field bundle, match billing and minimums |
| Dropcontact | B2B contact enrichment and verification | Asynchronous processing and input completeness matter | Eligible inputs, credits and result timing |
| Tomba | Person or combined person-company response from email | Some no-match or invalid-input outcomes return an error object within HTTP 200 | Endpoint, fields, plan limits and credits |
| Apollo | Person matching with prospecting context | Match confidence and optional asynchronous phone results | Plan eligibility, credits and waterfall configuration |
No field is guaranteed for every email. Request the same sample and schema from each supplier, and check the purpose and rights attached to the data.
1. People Data Labs: for structured person enrichment
People Data Labs' Person Enrichment API matches supplied identifiers to person records and returns structured attributes. A work email can be one input; additional reliable identifiers can help assess whether the match is the intended person.
Before using the response for routing, inspect the match signal, employment details and field coverage for your actual audience. A profile's current company or title may be missing or stale. Keep the data's source and retrieval time where available, and do not let a populated field silently displace an explicit answer from the form.
PDL's input documentation lists email alone as a valid request input and sets min_likelihood to 2 by default. The provider recommends a threshold of at least 6 for uses requiring high accuracy, while noting that email-only requests rarely score above 6. A higher threshold can reduce weak matches and also reduce returned coverage. Select and test the threshold against labelled records for the actual workflow rather than equating any returned profile with a confirmed identity. The response documentation describes likelihood as the provider's confidence in the person match; it does not certify that every returned employment field is current.
PDL is relevant for a team that needs a structured person-data contract and can manage matching rules in its own application. Check the current API reference, field-bundle pricing, minimums and no-match treatment for the intended plan. A lower-cost company-only requirement may be better served by a company endpoint rather than paying for an unnecessary person profile.
2. Dropcontact: for asynchronous contact enrichment
Dropcontact describes finding or verifying professional email and adding company and other information to B2B contacts. Its workflow can be asynchronous, so a form submission and a completed enrichment are not necessarily the same event. Ask which fields are returned from an email alone and when name or company inputs are needed for the result you want.
Its input reference permits an email as an identifier, but accepting the input is not a promise that it can identify the person or return every requested field. The credit policy distinguishes finding an email from submitting one for verification: when an email is supplied, it says one verification credit is consumed even if the address is invalid. For the 5,000-form example below, submitting all 5,000 addresses would therefore imply 5,000 verification credits under that published rule, not a charge only for successful person matches. Confirm eligibility and the fields returned before committing a production workload.
Its error documentation states that a 200 response can still carry a not-ready or other unsuccessful state. Persist the request ID, check the application-level response and use the supported polling or webhook path. Avoid marking a lead unqualified merely because enrichment has not finished.
Dropcontact is a candidate for teams comfortable with a queued contact-enrichment workflow. Confirm plan, credit and result-retrieval terms, and test how a returned company differs from the one provided by the prospect. A sales owner should have a route to resolve a conflict rather than having the API decide it silently.
3. Tomba: for person and company responses
Tomba documents person, company and combined enrichment responses, including endpoints that accept an email and return person and company fields. Its documentation notes that some no-match or invalid-input outcomes return HTTP 200 with an errors object rather than a data object. Inspect the response body, not only the status code.
Tomba describes Clearbit-compatible field names for certain endpoints, but similar keys do not prove identical coverage, freshness or semantics. Test the fields your application actually consumes, such as employment domain, title and company size. Check the plan's rate limits, credit rules and access to the desired endpoint before promising a response time to sales.
For example, Tomba's Person API reference identifies several Clearbit-shaped fields that it always returns as null, including employment seniority and sub-role. A CRM rule that depends on those fields needs another source or a review path; a schema-compatible response alone is insufficient. This statement applies to that documented person response, not necessarily every Tomba endpoint.
It is a useful candidate when a team wants a combined response or is replacing a schema with similar field names. A migration still needs field-by-field validation and a policy for webmail, missing data and mismatched entities.
4. Apollo: for person matching and prospecting context
Apollo's People Enrichment endpoint uses supplied information to match a person and provides a match_confidence field. Its documentation says better input can improve the chance of a match; a successful HTTP response can still indicate that no person was enriched.
Apollo documents one to nine credits per person without waterfall enrichment, depending on returned credit-consuming data: one credit for demographics or email and an additional eight if a mobile is returned. It says a request with no credit-consuming data uses zero credits; demographic charging also depends on match_confidence, while email and mobile charges do not. Its pricing guidance says waterfall costs depend on the configured sources and may include charges for lookups without a result. Phone or waterfall results can also require a webhook or polling flow. Check the plan and configuration; do not apply the basic one-credit rule to every workflow.
For planning, Apollo describes email waterfall enrichment as typically 1–4 credits and phone waterfall as typically 8–25 credits, while warning that certain configurations or matches can exceed those ranges. Those figures are not a ceiling or a quote. The same guidance says legacy plans can have different credit consumption, so check the workspace's actual terms before applying the current-plan examples.
For phone enrichment, Apollo documents polling only when the phone status is pending, using the top-level request_id_str. A skipped or not_attempted status means no phone job was created; repeatedly polling it will not turn it into a pending result. Record that state separately from a completed lookup that found no number, and confirm the current API contract before implementation.
For the documented phone or waterfall flow, choose a webhook or poll_only=true with the returned request ID; Apollo says supplying both polling mode and a webhook URL causes a 400 error. The main person response and the later phone or waterfall result are separate events. Link them to the same request and make webhook processing idempotent so a retry cannot create a duplicate CRM update.
When batching, Apollo's bulk endpoint accepts up to ten people. It describes its billed-person count as distinct people who consumed credits, not the number of matches, and notes that credits_consumed can be fractional depending on plan pricing. Parse the amount as a decimal and reconcile the later waterfall usage separately rather than using the number of returned matches as the bill.
Apollo is relevant when enrichment will feed an existing Apollo prospecting or CRM process. It may be more capability than a minimal inbound qualification service needs. Test the confidence threshold, returned fields, employer conflicts and actual credit usage before enabling optional personal-email or phone retrieval.
What do 5,000 inbound forms cost to enrich?
Use a hypothetical month with 5,000 form submissions: 4,000 use work email domains and 1,000 use personal or generic domains. Suppose the 4,000 work-email records represent 2,000 distinct company domains. The counts are modelling assumptions, not observed traffic.
If a team requests company data once per work-email submission, it performs 4,000 company lookups. If its policy permits a fresh cached company result to be reused for repeat domains, it performs 2,000 company lookups, a reduction of 2,000. At a synthetic $0.10 per lookup, the company-data component changes from $400 to $200. This is not a price for any named supplier; cache duration, contract rights and changes to company data must be considered.
Person enrichment is different. Two employees at the same company can have different roles, so deduplicating by domain must not collapse them into one person. Suppose 3,000 of the 4,000 work-email submissions are distinct people and person enrichment costs a hypothetical $0.25 per request. That is $750 before no-result rules and other charges. Combining it with the cached company component gives $950, excluding operations, CRM and tax. If only 1,800 records pass the team's identity and freshness rules, the illustrative data cost is about $0.528 per accepted person record.
The 1,000 personal-email submissions should not be automatically rejected or mapped to a company from the email domain. Route them using the information the prospect supplied or a respectful follow-up. If a team chooses to enrich them, evaluate that separate input and privacy policy rather than treating a generic email host as the employer.
Compare actual provider quotes using requested fields, match and no-match billing, plan minimums, rate limits, retries and storage rights. An optional mobile or waterfall can materially change the unit cost. The waterfall enrichment guide helps identify when a second request is worth the added expense.
Which fields should enrichment be allowed to write?
Define a field contract before connecting the API to the CRM. For each property, identify the authoritative source, whether the enrichment can create or update it, how old the value may be and who resolves a conflict. A sensible default is to preserve the prospect's submitted information and any field manually verified by a rep.
| CRM field | Example rule | Reason |
|---|---|---|
| Submitted email and name | Preserve the original form values; store suggested corrections separately | The provider match may be wrong or normalised differently |
| Company and domain | Write to a proposed-match field until confidence and hierarchy rules pass | Parent, subsidiary and shared domains can be ambiguous |
| Job title | Fill a blank field or propose an update; do not overwrite a protected manual value | Employment may have changed |
| Company size or industry | Record source and retrieval date; refresh on a defined schedule | Firmographics can be estimates or stale |
| Owner and consent | Keep under CRM policy; never infer from an enrichment response | Routing and consent require independent rules |
A blank response is not an instruction to erase an existing value. Likewise, a newer API retrieval timestamp does not prove the underlying fact was updated that day. Preserve the vendor's observation or update date if provided and mark unknown when it is not.
If the same person submits again, attach the new enquiry to the correct record while retaining both events and their original answers. Review the HubSpot contact enrichment guide for the packaged CRM perspective, and the CRM data cleaning guide for ongoing record governance.
How do you test the inbound record?
Use synthetic records that exercise the actual handoff. The fixture is proposed; we have not run it against the providers.
- A prospect submits a work email and identifies their company. The provider returns the same employer and a current-looking title. Record the response, confidence, field sources and CRM write.
- A second prospect at the same domain submits a different role. Reuse company data only where allowed; keep the two person records distinct.
- A prospect uses a personal email and supplies a company name. Preserve the submission and route according to the business rule without treating the email host as the employer.
- A provider returns a parent company while the prospect names a subsidiary. Mark the match for review or apply a documented hierarchy rule; do not overwrite the submitted company automatically.
- A rep has manually verified a title. A later enrichment returns a conflicting title. Preserve the protected value and show the suggestion with its source and date.
- The API returns pending, no result, rate limit or an error object inside HTTP 200. Each needs a different state, retry and alert; none should be silently counted as a qualified match.
Check idempotency when the form sends the same event twice, and record the provider request ID and credit consumption. Keep an exception queue with a named owner and an age threshold. After the pilot, compare useful accepted records by segment and input type, not merely the number of non-empty responses.
Which email enrichment API should you choose?
Evaluate People Data Labs for structured person data, Dropcontact for a queued B2B contact workflow, Tomba for person and company responses, and Apollo for a matched person record within a broader prospecting stack. Use the same input sample and field-protection rules for each; select on usable context, operational fit and total cost.
FAQ
Can an API identify a company from a personal email address?
Not from the email domain alone. A provider may find a person match using other data, but that match should be assessed against the information the prospect supplied. Do not treat a consumer email provider as the employer.
Should enrichment overwrite a rep's verified CRM field?
Not by default. Store the new value as a suggestion with its source and date, then apply a documented review or precedence rule. Protect fields that the business has deliberately marked as authoritative.
Does a successful API response mean the lead is qualified?
No. A successful response may be incomplete, uncertain or unrelated to your qualification criteria. Qualification is a business decision using the prospect's needs and the team's rules, not merely the presence of a company or job title.





