More Resources
Prompt

Verify a B2B email address before you send

Built by: Lusha
Time to build: 1 min
Difficulty: Easy
Tools: Claude ▪ Gmail ▪ Lusha

Images on this page are for illustrative purposes only. Example outputs in this play are illustrative — the structure, fields, and format reflect real Lusha connector and Gmail output, but were not pulled from a live session. Run the prompt with your own data and connectors to see live results. Personal details in any live examples are masked or abbreviated for privacy.

Sending to an unverified email address risks a bounce, a spam flag, or a message landing with someone who left the company six months ago. This Claude prompt checks any B2B email address against live Lusha data — confirms the address is current, checks Gmail for any prior bounce history, and flags any title change worth knowing before the email goes out.

The prompt

Open in Claude Open in ChatGPT 
<context>
Before I send an outreach email, I want to verify the email address is valid and deliverable — not a guessed format, not a role address, not a contact who has left the company. I want to confirm the address belongs to the right person at the right company before anything goes out.

My contact:
- Name: [NAME]
- Company: [COMPANY NAME OR DOMAIN]
- Email I have: [EMAIL ADDRESS OR "FIND IT"]
- Title on file: [TITLE OR "UNKNOWN"]
</context>

<task>
1. Use Lusha to look up this contact by name and company:
   - Does Lusha have a verified work email for this person?
   - Does it match the email I have on file?
   - Is the contact still at the company?
   - What is the current verified title?

2. Check Gmail for any prior send history to this address:
   - Has an email been sent to this address before?
   - Did it bounce?
   - Did the contact reply from this address?

3. Return a verification result:
   - VERIFIED MATCH: Lusha email matches what I have — safe to send
   - VERIFIED DIFFERENT: Lusha has a different email — use the Lusha address
   - VERIFIED NEW: I had no email — Lusha found one
   - DEPARTED: contact is no longer at the company — don't send
   - UNVERIFIED: Lusha cannot confirm — manual check before sending

4. If DEPARTED: find the most likely replacement in the same function via Lusha.

5. Return:
   - Verification status
   - Correct verified email to use
   - Current title (update CRM if different)
   - One flag if anything should change about the outreach
</task>

<constraints>
- Only return Lusha-verified emails. Don't suggest format guesses.
- If Gmail shows a prior bounce to this address, flag it even if Lusha returns the same address.
- The outreach flag must be specific — not "update your records" but what specifically changed.
</constraints>

What you'll get back

The situation: A rep has P.M. at Meridian Analytics on file as “VP of Sales” with email [email protected]. The address was added to the CRM 14 months ago. Running the verification before adding to a campaign.

VERIFIED DIFFERENT — title changed

  • P.M. confirmed at Meridian Analytics ✓
  • CRM email: [email protected]
  • Lusha verified email: p.m@[meridian].com ✓ (same domain, confirmed active)
  • CRM title: VP of Sales
  • Current verified title: Senior Vice President of Sales
  • Prior Gmail send: 1 email sent 11 months ago — no bounce, no reply

Outreach flag: Title changed VP → SVP since the email was added to the CRM. Update the title field and the {{title}} personalization token before this contact goes into any campaign sequence.

Illustrative example. Run with your own data to see live results.

Why use Lusha in Claude

Sending to an unverified address is one risk. Sending with the wrong title in the personalization token is a second — a SVP who receives an email addressed to “VP of Sales” knows immediately the sender hasn’t checked. Both problems appear in a single verification run. The Gmail check adds the bounce history layer: if the address previously bounced and Lusha returns the same address, that combination is worth flagging before sending again — the address may be technically valid but practically unreachable.

The five-status output — VERIFIED MATCH, VERIFIED DIFFERENT, VERIFIED NEW, DEPARTED, UNVERIFIED — is what makes the result actionable. Each status maps to a specific next step. DEPARTED means stop and find a replacement. UNVERIFIED means research before sending. VERIFIED MATCH means proceed. No ambiguity.

Data drawn from 300M+ verified contacts under GDPR, CCPA, SOC 2, ISO 27701, ISO 31700, and TRUSTe.

FAQ

  • What if I don't have an email address at all?

    Set the email field to “FIND IT” — Lusha searches by name and company and returns the verified address. The result is the same structure: VERIFIED NEW status with the address to use. No email format guessing required.

  • What if Lusha returns UNVERIFIED?

    UNVERIFIED means Lusha has no verified record for this contact at this company. Before sending: check LinkedIn to confirm the person is still there, then re-run the prompt with an updated company name or domain if anything has changed. Don’t guess the format and send anyway — a guessed format that lands wrong is worse than a delayed send.

  • How is this different from an email validation tool?

    An email validation tool checks whether an address exists at a domain. This prompt checks three different things: whether the specific person is still at the company in the same role, whether the email format Lusha has differs from what’s in the CRM, and whether prior Gmail send history shows a bounce. Three checks, not one, in a single 60-second run.

  • Should I run this for every contact before sending?

    Run it for any contact whose CRM record is more than 6 months old, any contact going into a high-priority sequence, and any contact on a list that’s never been verified. For validating a full campaign send list rather than one contact at a time, the campaign list cleaning prompt covers the full list in one pass.

  • What happens if the contact departed and Lusha can't find a replacement?

    The prompt returns DEPARTED with a note that no replacement was found in Lusha for that function at that company. That’s a clean output — it tells the rep the account needs manual research before outreach proceeds, rather than sending to a dead address.

Related Prompts

Ready to run this?

One data connection. Works in Claude, ChatGPT, your CRM, or any agent you build.

Start for free Read the Docs