How to Test a B2B Lead Gen Platform's Integrations (Before You Commit)

A platform has strong integrations if you can reach every feature from code, a no-code tool and an AI client with one key. Logo walls prove none of that. Six requests do.

This guide uses LeadOcean as the worked example. Run the same six checks against any vendor you are comparing.

TL;DR

• Strong integrations mean one key, documented endpoints, predictable errors and a path for each client type: REST, no-code tool, MCP and file export.
• The test takes about 30 minutes and spends about 5 LeadOcean records. Counts, filter lookups and account checks are free.
• Anything you cannot do from the API but can do in the vendor's UI is a lock-in signal. Ask about it before you pay.

Prerequisites

• A LeadOcean account on the free plan: 1,000 records for life, no card, 5 requests a second. Sign up.
• Your key exported as $LEADOCEAN_API_KEY.
• curl and jq. Optional: n8n, Make, Zapier or Clay for step 5.
• A second vendor's trial key, if you want a side-by-side.

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

Step 1: Check auth and limits with one free call

A good API tells you who you are, what plan you are on and how much you have left. Without that, no integration can alert you before it breaks.

curl -s https://api.leadocean.io/v1/account \
  -H "x-api-key: $LEADOCEAN_API_KEY" | jq

GET /v1/account and GET /v1/usage are free. Look for the plan (data.account.plan) and the records left (data.records.remaining). If a vendor hides these behind a dashboard, your scripts cannot watch them.

Step 2: Discover filters from the API, not from a PDF

Integrations break when filter names drift from the docs. Pull the live list instead.

curl -s https://api.leadocean.io/v1/filters \
  -H "x-api-key: $LEADOCEAN_API_KEY" | jq '.data.filters | keys'

curl -s "https://api.leadocean.io/v1/enums" \
  -H "x-api-key: $LEADOCEAN_API_KEY" | jq -r '.data.enums[].name'

Both calls are free. LeadOcean exposes 50 people filters and 34 company filters. Enum values are exact strings, so jobLevel takes VP, not vice president. A wrong value returns an empty result, not an error. Always read the enum list first.

Step 3: Size the list before you spend anything

Ask for a count with count=true as a query parameter and limit=1. The body stays flat JSON and filters are arrays.

curl -s -X POST "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"],
    "jobFunction": ["Sales & Business Development"],
    "country": ["US"]
  }' | jq '.meta'

meta.total is capped at 100,000 on this route, and the response sets meta.countOnly to true. For an uncapped number, use the free MCP tool leadocean_count_leads, shown below. For companies, GET /v1/companies/count is free and returns breakdowns.

A vendor that charges to size a list pushes you to guess. That is an integration cost too.

Step 4: Pull one page and walk the cursor

Real integrations paginate. Pull five rows, then follow the cursor.

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"],
    "limit": 5
  }' | jq '{next: .meta.nextCursor, rows: (.data | length)}'

Pass nextCursor back as cursor for the next page. limit runs 1 to 100. One search pages to 10,000 rows. The last page sets meta.depthCapped to true and has no nextCursor. Narrow the filters instead of walking further.

This step spends 5 records. A search page of 25 would spend 25, so keep limit small while testing.

Step 5: Connect the tool you already run

Any tool that sends an HTTP request with an x-api-key header works with the REST API. Put the key in a credential, not in the node.

• n8n: an HTTP Request node, or the MCP Client Tool node for agents. See LeadOcean + n8n.
• Make, Zapier, Clay: HTTP modules against the same endpoints. See LeadOcean + Make, LeadOcean + Zapier and LeadOcean + Clay.
• Cold email and verification: ReachHQ and VerifyHQ take LeadOcean lists as input.

All recipes are on the integrations hub. If a vendor's no-code connector exposes fewer filters than its API, note the gap. You will hit it in month two.

Step 6: Test the AI client and the file path

Two more client types matter: an AI assistant and a spreadsheet.

The hosted MCP server lives at https://api.leadocean.io/mcp. It signs in with OAuth 2.1, so you paste no key. It exposes 13 tools, and seven of them are free to call: leadocean_count_leads, leadocean_count_companies, leadocean_list_filters, leadocean_list_enum_values, leadocean_get_account, leadocean_get_export and leadocean_list_exports. Clients can self-register at https://api.leadocean.io/oauth/register.

Ask your AI client: "Count VPs of Sales in the US, then show me the breakdown." It should call leadocean_count_leads and spend nothing.

For files, POST /v1/exports takes up to 50,000 rows a request and 500,000 a day. Bulk CSV export is not on Free. The app at app.leadocean.io has an Exports page with a filter builder, a column picker, a price preview and a CSV download. That page is a convenience, and the same job runs over the API.

What you get

Output of leadocean_count_leads on 2026-10-21 with jobLevel: VP, jobFunction: Sales & Business Development, country: US. This is a mailable count: verified, catch_all_valid and catch_all addresses only.

{
  "total": 129494,
  "totalIsExact": true,
  "totalAll": 350517,
  "distinctCompanies": 58121,
  "distinctCompaniesIsExact": false,
  "credits": 0,
  "unsupportedFilters": []
}

totalAll is the same filters with no email condition. The call cost zero records.

Response shape from the API reference for one person from enrich, with placeholder values:

{
  "success": true,
  "data": {
    "profile_data": { "profile_full_name": "Jane Doe" },
    "contact_data": {
      "has_email": true,
      "email_status": "verified",
      "contact_emails": [{ "email": "jane@acme.com", "type": "work", "status": "verified" }],
      "contact_phones": [{ "phone": "+14155550133", "type": "mobile" }],
      "contact_current_experiences": [{ "job_title": "VP Sales", "company_name": "Acme", "company_domain": "acme.com" }]
    },
    "resume_data": { "experiences": [], "educations": [] },
    "meta": { "source_ids": { "person_id": "1234567890" } }
  },
  "meta": { "credits": 1 }
}

Enrich takes exactly one of linkedin_url, email, phone or person_id. Since October 2026 it also returns career history in resume_data: past roles, education, certifications and awards.

Troubleshooting

FAQ