How to Build an Accurate Prospect List (Sales Intelligence Data)

TL;DR

• Accurate prospecting data is a pipeline, not a vendor claim: size the list, filter on verification, confirm each address on enrichment, then gate on status before you send.
• It takes about 20 minutes and spends 20 LeadOcean records. Counts are free.
• The Free plan (1,000 records for life, no card) covers it many times over.

Prerequisites

• A LeadOcean API key from app.leadocean.io, exported as $LEADOCEAN_API_KEY. Free gives you 1 key and 5 requests a second.
• curl and jq.
• Optional: the hosted MCP server at https://api.leadocean.io/mcp (13 tools) if you would rather size the list from an AI client.

Step 1: Size the audience before you spend anything

Count first. A free count tells you whether the filter is empty, too broad or worth pulling.

Send count=true as a query parameter, not in the body, with limit 1. The body rejects unknown fields.

curl -s -X POST "https://api.leadocean.io/v1/people/search?count=true" \
  -H "x-api-key: $LEADOCEAN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "jobLevel": ["VP"],
    "jobFunction": ["Sales & Business Development"],
    "country": ["US"],
    "employeeRange": ["201-500", "501-1000"],
    "limit": 1
  }' | jq '.meta | {total, totalIsExact, distinctCompanies}'

meta.total is capped at 100,000. meta.distinctCompanies tells you how many employers the people span, which is the number to compare against your account list.

Enum values are exact strings. Check them with GET /v1/enums/job_level instead of guessing.

Step 2: Filter on email status, and know what the filter means

Pick the status filter on purpose. The default mailable audience is verified, catch_all_valid and catch_all. Narrow to verified when the cost of a bounce is high.

Run the same filter both ways with the MCP tool leadocean_count_leads (free). On 2026-10-21 it returned this:

verified on a business address means LeadOcean's own verifier sent to the address and it accepted. A supplier saying an address is good does not earn the word. Those addresses are published as untested.

One caveat matters. People search carries no per-address verification provenance, so a search row can say verified and enrichment can say untested. Treat a search count of verified as an upper bound. The enriched answer is the correct one.

Step 3: Pull a page and confirm each address on enrichment

Pull a small page with the verified filter. Each person returned is one record.

curl -s -X POST "https://api.leadocean.io/v1/people/search" \
  -H "x-api-key: $LEADOCEAN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "jobLevel": ["VP"],
    "jobFunction": ["Sales & Business Development"],
    "country": ["US"],
    "employeeRange": ["201-500", "501-1000"],
    "emailStatus": ["verified"],
    "limit": 10
  }' > page1.json

Search rows are a thin projection with no contact points. Each row carries its id at meta.source_ids.person_id. List them, then enrich each one with reveal_email true.

jq -r '.data[].meta.source_ids.person_id' page1.json

Enrich takes exactly one of linkedin_url, email, phone or person_id.

curl -s -X POST "https://api.leadocean.io/v1/people/enrich" \
  -H "x-api-key: $LEADOCEAN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "person_id": "PERSON_ID_FROM_PAGE1", "reveal_email": true }'

Page with meta.nextCursor, passed back unchanged. One search stops at 10,000 rows (meta.depthCapped true). Narrow the filters instead of walking further.

Step 4: Gate on status and date before anything is sent

Keep a row only if the enriched status passes. The address object carries status and verified_batch_date. The date is the verification run, not a timestamp for the address, and it can be null.

curl -s -X POST "https://api.leadocean.io/v1/people/enrich" \
  -H "x-api-key: $LEADOCEAN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "person_id": "PERSON_ID_FROM_PAGE1", "reveal_email": true }' \
| jq '.data.contact_data.contact_emails[0]
      | select(.status == "verified" or .status == "catch_all_valid")
      | {email, status, verified_batch_date}'

Rules worth encoding:

• Send: verified, catch_all_valid.
• Send with care: catch_all. The domain accepts everything and the mailbox cannot be confirmed.
• Hold: risky, role, derived, untested, unknown.
• Drop: invalid, spam_trap, abuse, disposable.

Bounced emails are not refunded or credited back. Read email_status before sending, not after.

Step 5: Cross-check the account side and the role

Accurate contacts at the wrong accounts still waste a sequence. Size the company list for free with GET /v1/companies/count. It takes the company-search filters and returns an exact total with breakdowns by industry, headcount band, HQ country, HQ state and NAICS.

curl -s "https://api.leadocean.io/v1/companies/count?industry=Computer%20Software&employeeRange=201-500" \
  -H "x-api-key: $LEADOCEAN_API_KEY" | jq '.data | {total, totalIsExact}'

Compare that total with meta.distinctCompanies from Step 1. If the people span far fewer employers than the company list holds, your role filter is too narrow for the account list. If they span more, the title filter is pulling adjacent teams.

Enrichment also returns career history in resume_data: past roles with title, company name, dates and job level, plus education and certifications when held. Use it to catch a stale title. A person whose current role started last month and whose previous role matches your ICP is a better target than the row alone suggests. Past employers carry no company_id yet, so match them by name.

The same Step 1 and Step 2 counts work from an AI client. Ask the MCP tool leadocean_count_leads for the numbers, and spend records only when you move to Step 3.

A status check you can run on your own sample

Do not take any vendor's accuracy number on trust, ours included. Enrich 10 rows, save each response as enriched_1.json to enriched_10.json, send nothing, and tally the statuses.

jq -r '.data.contact_data.contact_emails[0].status' enriched_*.json | sort | uniq -c

If the verified filter returned 10 rows and enrichment confirms 6, your search-level count overstates the safe list by about 40 percent for that segment. Apply that ratio to the Step 2 total before you size a campaign. Repeat per segment, because the gap differs by company size and region. Ten records is a cheap test.

Why this order

Each step removes a different kind of error. Counts catch an empty or runaway filter before it costs anything. The status filter removes addresses nobody has tested. Enrichment replaces the search-level label with the per-address verdict. The gate keeps risky mail out of the sequencer. The account cross-check keeps good contacts at the right companies.

Skip a step and the error moves downstream, where it costs more. A bounce is paid for in sender reputation, and nothing is credited back.

What you get

This is the response shape from the API reference, with placeholder values. It is not a real record.

{
  "success": true,
  "data": {
    "profile_data": { "profile_full_name": "Jane Doe", "profile_headline": "VP Sales at Acme" },
    "contact_data": {
      "has_email": true,
      "has_phone": false,
      "email_status": "verified",
      "email_type": "work",
      "contact_emails": [
        {
          "email": "jane.doe@acme.com",
          "type": "work",
          "status": "verified",
          "verified_batch_date": "2026-09-20"
        }
      ],
      "contact_current_experiences": [
        { "job_title": "VP Sales", "company_name": "Acme", "company_domain": "acme.com", "start_date": "2025-06-01" }
      ]
    },
    "resume_data": {
      "experiences": [{ "job_title": "Director of Sales", "company_name": "Globex", "start_date": "2021-01-01", "end_date": "2025-05-31" }],
      "educations": []
    },
    "meta": { "source_ids": { "person_id": "123456789" }, "fetched_at": "2026-10-01" }
  },
  "meta": { "credits": 1 }
}

The sizing numbers in Step 2 are real, taken from leadocean_count_leads on 2026-10-21. A 6,375-person verified audience across 3,572 companies is a list you can work.

Troubleshooting

FAQ