Clean CSV email lists before import or outreach

The CSV is where dirty data hides

A CSV export is the messiest artifact in any pipeline. It carries whatever the source tool guessed, whatever a human typed with a fat finger, duplicate rows from two overlapping saved searches, and a header row that may or may not be where you expect it. Before that file becomes a CRM import or a campaign audience, every bad row in it is a future bounce, a duplicate contact, or a support ticket.

CSV email verification is the step that turns an untrusted export into a trusted list. Verifly reads the email column, runs a live SMTP mailbox check on each address, and hands back a per-row verdict you can join straight back onto the original file — keeping every other column intact so names, companies, and tags travel with the clean addresses.

There is no upload UI to learn and no file-size negotiation. You parse the column, POST it to the bulk endpoint, and get structured JSON back. For genuinely large exports the async pipeline handles up to 1,000,000 rows in a single job.

Parse, verify, and rejoin a CSV

1. Open the CSV and identify the email column (it is not always the first one).
2. Extract that column into an array and de-duplicate it so you do not pay to check the same address twice.
3. Register for 100 free credits at verifly.email/api/v1/autonomous/register.
4. POST the array to /verify/bulk (async) or /verify/batch (smaller lists).
5. Join the verdicts back onto the original rows by email, keep result === 'deliverable', and write out a clean CSV that still has all your other columns.

import fs from 'node:fs'

const rows = fs.readFileSync('contacts.csv', 'utf8').trim().split('\n')
const header = rows.shift().split(',')
const emailIdx = header.findIndex(h => /email/i.test(h))

const emails = [...new Set(rows.map(r => r.split(',')[emailIdx].trim()))]

const res = await fetch('https://verifly.email/api/v1/verify/batch', {
  method: 'POST',
  headers: { 'Authorization': 'Bearer vf_your_api_key', 'Content-Type': 'application/json' },
  body: JSON.stringify({ emails }),
})
const { results } = await res.json()
const verdict = new Map(results.map(r => [r.email, r.result]))

const clean = rows.filter(r => verdict.get(r.split(',')[emailIdx].trim()) === 'deliverable')
fs.writeFileSync('contacts_clean.csv', [header.join(','), ...clean].join('\n'))