How to get phone numbers at scale for cold calling (API recipe)

Getting phone numbers at scale takes four calls: size the segment, filter to people who have a number, look up the numbers, export the list. Sizing is free. You pay one record for each person you actually pull.

TL;DR

• Use hasPhone: true in a search so you never pay for people without a number. Size the segment first with a free count.
• Takes about 20 minutes. The test pass spends about 110 records of the 1,000 on the Free plan. Bulk CSV export needs Solo ($97 a month) or above.
• Pull one number with GET /v2/people/phone, or pull thousands at once with POST /v1/exports (up to 50,000 rows a request, 500,000 a day).

Prerequisites

• A LeadOcean API key in $LEADOCEAN_API_KEY. The Free plan gives you one, with 1,000 records for life and no card.
• curl and jq.
• A number to start from: a job level, a function and a country.

Step 1: Size the segment for free

Count first. A count spends no records, so run it as often as you like.

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

count=true goes in the query string, not the body. meta.total is capped at 100,000 on this route, so use the MCP tool leadocean_count_leads when you want the exact figure. Filters are flat arrays. Enum values are exact strings, so check them with leadocean_list_enum_values before you guess.

Step 2: Search for people who have a number

Add limit (1 to 100) and page with cursor. Each person returned is one record.

curl -s https://api.leadocean.io/v1/people/search \
  -H "x-api-key: $LEADOCEAN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "jobLevel": ["VP", "Director"],
    "jobFunction": ["Sales & Business Development"],
    "country": ["US"],
    "hasPhone": true,
    "limit": 100
  }' > page1.json

jq '.meta.nextCursor' page1.json

Pass nextCursor back as cursor for the next page. One search pages to 10,000 rows. When meta.depthCapped is true, narrow the filters (add a title, a company size or a city) instead of walking deeper. Test with 100 rows before you scale.

Step 3: Look up the numbers

A search row says a number exists. The phone endpoint returns it. Send the person id from the search row.

curl -s "https://api.leadocean.io/v2/people/phone?person_id=PERSON_ID" \
  -H "x-api-key: $LEADOCEAN_API_KEY"

It costs one record per call, including a call that finds nothing, so filter with hasPhone first. Numbers come back best first, and the array position is the priority. Each number has a phone and a type such as mobile, direct or office. Treat an unfamiliar type as unknown.

The v2 response carries no do-not-call flag. Scrub every number against your own DNC lists before you dial.

Step 4: Export the whole list as CSV

For thousands of rows, skip the per-person loop. One export call writes the CSV for you.

curl -s https://api.leadocean.io/v1/exports \
  -H "x-api-key: $LEADOCEAN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "us-sales-leaders-with-phone",
    "filters": {
      "jobLevel": ["VP", "Director"],
      "jobFunction": ["Sales & Business Development"],
      "country": ["US"],
      "hasPhone": true
    },
    "limit": 5000,
    "preset": "full",
    "reveal": "email_phone"
  }'

The full preset has phones 1 and 2 as phone_1_number and phone_2_number. Poll GET /v1/exports/{id}, then fetch GET /v1/exports/{id}/download. Each row written is one record. The ceilings are 50,000 rows an export and 500,000 a day. You can run the same export from the Exports page in the app (app.leadocean.io), which shows the record price before you start.

Step 5: Feed the dialer

LeadOcean has no dialer or sequencer. Load the CSV into the tool your reps already use. Keep the phone_1_number column as the first dial and phone_2_number as the fallback. Re-run the export on a schedule, since the dataset refreshes monthly and each record carries its own fetched_at date.

What you get

Here is a free count run on 2026-10-21 with the MCP tool leadocean_count_leads. It filtered on job level VP or Director, job function Sales & Business Development, country US and hasPhone true. It spent no records.

{
  "total": 864742,
  "totalIsExact": true,
  "distinctCompanies": 296626,
  "credits": 0
}

That is a count of people with a phone number on file, not a mailable count and not a promise that every number connects.

The phone lookup returns this shape (response shape from the API reference, placeholder values):

{
  "success": true,
  "data": {
    "person_id": "1234567890",
    "found": true,
    "phones": [
      { "phone": "+1 415 555 0133", "type": "mobile" },
      { "phone": "+1 415 555 0199", "type": "office" }
    ],
    "best": { "phone": "+1 415 555 0133", "type": "mobile" }
  }
}

Troubleshooting

FAQ