How to Build Email Lists That Actually Convert (B2B Outbound)

A list converts for two reasons: the person reading it owns the problem, and the message reaches an inbox. Everything else is polish. This guide builds a list around those two tests with real requests against the LeadOcean API.

Lists fail in predictable ways. The audience is too wide, so the message fits no one. The addresses are unchecked, so mail bounces and the sending domain pays. The file is old, so part of it points at people who changed jobs.

Most list-building advice stops at "define your audience" and "keep it clean". Here each of those is a query. You size the audience for free, filter by role and company size, and refuse any address that is not mailable.

The example target is VPs of Sales in Canada. Swap the filters for your own market.

TL;DR

• It takes about 25 minutes to size an audience, pull a sample, reveal one address and export a small list.
• The walkthrough spends about 31 records: 5 from the sample search, 1 from the reveal and 25 from the export. Sizing and counts are free. CSV export needs Solo or above.
• Mail only verified, catch_all_valid or catch_all. Every filter you add makes the list smaller and the audience tighter, so size before you spend.

Prerequisites

• A LeadOcean API key from app.leadocean.io. The free plan has 1,000 records, one-off, and no card.
• The key in your shell as $LEADOCEAN_API_KEY.
• curl, jq and Miller (mlr) for the cleanup in Step 6.
• A way to send mail. Any sequencer works. This guide stops at the CSV.

export LEADOCEAN_API_KEY="paste-your-key-here"

Step 1: Define the buyer as filters, not adjectives

Write the buyer as filter values before you open any tool. A filter is a test a person passes or fails. "Senior sales leader at a mid-size company" is not.

{
  "jobLevel": ["VP"],
  "jobFunction": ["Sales & Business Development"],
  "country": ["CA"],
  "employeeRange": ["51-200", "201-500"]
}

Enum values are exact strings. VP works and Vice President returns nothing. Check a value first with the free leadocean_list_enum_values MCP tool, or GET /v1/enums.

Write down the one sentence your email will open with. If a filter does not help that sentence stay true for every row, drop it. If the sentence needs a filter you lack, such as a tool the company uses, find the matching one with GET /v1/filters or leadocean_list_filters. Both are free.

Three filters do most of the work: role (jobLevel plus jobFunction), geography (country or salesRegion) and company size (employeeRange). Add industry or technologyCategories only when your offer is specific to one.

Step 2: Size the list before you spend a record

Count how many people match, for free. Send count=true and limit=1 as query parameters. Arrays go comma-separated.

curl -s -G https://api.leadocean.io/v1/people/search \
  -H "x-api-key: $LEADOCEAN_API_KEY" \
  --data-urlencode "jobLevel=VP" \
  --data-urlencode "jobFunction=Sales & Business Development" \
  --data-urlencode "country=CA" \
  --data-urlencode "emailStatus=verified,catch_all_valid,catch_all" \
  --data-urlencode "count=true" \
  --data-urlencode "limit=1" | jq '.meta | {total, totalIsExact, countOnly}'

The call returns no rows and spends nothing, so run it as often as you like. Run it with and without each filter to see which one cuts the list most. A list under about 500 is a hand-written campaign. A list over 100,000 is too broad to convert, because meta.total stops at 100,000 and the filters are not doing any work yet.

Through MCP, the same call is leadocean_count_leads. It also defaults to mailable addresses, which is the number you care about.

Step 3: Pull a sample and read it

Read five real rows before you commit to the list. Each person returned costs 1 record, so keep limit small.

curl -s 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": ["CA"],
    "employeeRange": ["51-200", "201-500"],
    "emailStatus": ["verified", "catch_all_valid", "catch_all"],
    "limit": 5
  }' | jq '.data[] | {person_id: .meta.source_ids.person_id, contact_data}'

Read the titles and companies. If two of five are the wrong person, tighten title with an exact match in brackets, such as "[VP Sales]", or add excludeTitle. The check is the point of this step. You spend 5 records to learn whether the filters describe the right people, before you spend 25 or 25,000.

Search returns a thin row. It carries a person_id and flags such as has_email, not the address.

Step 4: Reveal only the addresses you will send

Send the person_id to enrich with reveal_email set to true. Enrich takes exactly one of linkedin_url, email, phone or person_id.

curl -s 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_STEP_3", "reveal_email": true}' \
  | jq '.data.contact_data | {email_status, contact_emails}'

This costs 1 record. A person we hold nothing for is a 404 and costs nothing.

The status is the gate. verified was tested. catch_all_valid means the mailbox was confirmed on a domain that accepts everything. catch_all means the domain accepts everything and the mailbox is unproven. Everything else, including risky, unknown and role, stays out. LeadOcean gives no refund for bounced emails, so read email_status before you send.

Step 5: Export the list as a CSV

Export the whole audience in one request. Bulk CSV export is not on Free: it is included on Solo ($97 a month) and Agency ($497 a month). On Free, reveal the addresses one by one as in Step 4. The body takes the same filters as search, a limit and a column preset. The walkthrough uses 25 rows.

curl -s https://api.leadocean.io/v1/exports \
  -H "x-api-key: $LEADOCEAN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "ca-vp-sales",
    "filters": {
      "jobLevel": ["VP"],
      "jobFunction": ["Sales & Business Development"],
      "country": ["CA"],
      "employeeRange": ["51-200", "201-500"],
      "emailStatus": ["verified", "catch_all_valid", "catch_all"]
    },
    "limit": 25,
    "preset": "full"
  }' | jq '.data | {id, status, requested, pollAfter}'

Poll GET /v1/exports/{id} until status is done, then fetch GET /v1/exports/{id}/download. Rows are deduplicated by person_id, and every row costs 1 record whatever the columns.

An export reserves up to limit records while it runs and bills each row it writes. One export takes up to 50,000 rows, and an account takes 500,000 a day. On a paid plan you can run the same export from the Exports page in the app. It shows the record price before you start.

Step 6: Clean the CSV before it reaches your sequencer

Drop what should not be mailed, then load the file. Three checks are enough. The first keeps only the mailable statuses, using the column names from the export header. The second drops anyone already in your CRM or on your suppression list.

mlr --icsv --ocsv filter '
  $email_1_status == "verified" ||
  $email_1_status == "catch_all_valid" ||
  $email_1_status == "catch_all"' ca-vp-sales.csv > mailable.csv

mlr --icsv --ocsv join --np --ul -j email_1_address -f suppress.csv mailable.csv > send.csv

Third, mail in small batches and watch the first bounces before the rest go out. Each record also carries its own fetched_at date, and the dataset refreshes monthly. A list you exported six months ago is older than the data behind it. Re-export before a new campaign instead of re-mailing the old file. The cost of decay is covered in email list churn.

What you get

Step 2 is free. We ran it on 2026-10-21 with leadocean_count_leads for VPs in Sales & Business Development in Canada, changing one filter at a time. Mailable rows use the tool's default (verified, catch_all_valid or catch_all) with no email filter passed.

Each row is a count of people, not a promise of replies. The mailable count is about 30% of everyone matching the role and country (5,574 of 18,800). The rest carry a weaker status or no email. Adding the size filter shrinks the list to 1,300 people you can message in one sentence.

A revealed person comes back in the shape below. This is the response shape from the API reference, with placeholder values.

{
  "success": true,
  "data": {
    "profile_data": { "...": "name, title, company" },
    "contact_data": {
      "has_email": true,
      "email_status": "verified",
      "email_type": "work",
      "contact_emails": [
        { "email": "jane.doe@acme.com", "type": "work", "status": "verified" }
      ]
    }
  }
}

Troubleshooting

FAQ