More Resources
Prompt

Find every account where the main contact has gone dark or left

Built by: Lusha
Time to build: 4 min
Difficulty: Medium
Tools: Claude ▪ Gmail ▪ Lusha

Images on this page are for illustrative purposes only. Example outputs are based on Lusha data, with personal details masked or abbreviated for privacy.

This Claude prompt runs a monthly early-warning scan across a book of accounts — Lusha checks whether each primary contact is still at the company, Gmail checks when engagement dropped off. Returns a DARK / COOLING / ACTIVE classification per account with replacement contacts found and a specific re-engagement action for every DARK account before it becomes a churn or dead pipeline problem.

The prompt

Open in Claude Open in ChatGPT 
<context>
I want to find every account in my book where the primary contact relationship has gone dark — either the contact left, or engagement dropped off — before it turns into a churn risk or a dead deal I don't know about.

My book:
- Account list with primary contacts: [PASTE COMPANY, CONTACT NAME, TITLE — one per line]
- What counts as dark: [45+ DAYS NO REPLY / CONTACT DEPARTED / BOTH]
- Account type: [CUSTOMER / PROSPECT / BOTH]
</context>

<task>
1. For each account, check two things in parallel:

   CONTACT STATUS (Lusha):
   - Is the primary contact still at the company in the same role?
   - If departed: find the most likely replacement via Lusha

   ENGAGEMENT STATUS (Gmail):
   - When was the last inbound email from this account?
   - When was the last outbound email to this account?
   - Is there an unanswered outbound email sitting unacknowledged?

2. Classify each account:
   - DARK — CONTACT GONE: primary contact has left, no active thread
   - DARK — GONE QUIET: contact still there but no inbound for 45+ days
   - COOLING: last inbound was 21–44 days ago — not dark yet but trending that way
   - ACTIVE: recent inbound, contact verified — no action needed

3. For DARK accounts:
   - Return the specific reason (departed / gone quiet)
   - Return replacement contact details if contact has departed (via Lusha)
   - Suggest one specific re-engagement action

4. Return a dark accounts report:
   - Summary: X DARK (contact gone), X DARK (gone quiet), X COOLING, X ACTIVE
   - DARK accounts sorted by account value or deal stage
   - For each DARK account: contact status, last touch date, replacement if found, recommended action
   - Total ACV or ARR at risk from DARK accounts (if provided)

5. Flag any account that is DARK on both dimensions simultaneously — contact gone AND no engagement. These are the highest-priority accounts to act on this week.
</task>

<constraints>
- An account can be DARK for two different reasons — contact gone vs gone quiet. Both are risks but they require different responses.
- Don't surface COOLING accounts as urgent — they're a watch list, not an action list.
- The recommended action must be specific: not "reach out" but who to contact, how, and what angle.
</constraints>

What you'll get back

The situation: A CS manager runs this monthly scan across 7 accounts — mix of customers and active prospects. Checking for any contact that left or went quiet in the last 45 days.

Output from live Lusha run + Gmail, May 25, 2026:

Summary: 7 accounts · 1 DARK (contact gone) · 1 DARK (gone quiet) · 2 COOLING · 3 ACTIVE

⚠ DOUBLE DARK — highest priority

[Sales enablement SaaS] — $72K ARR — DARK (contact gone + no engagement)

  • Primary contact B.N., Sr Director RevOps: confirmed departed from [sales enablement SaaS] in Lusha
  • Last inbound: 67 days ago
  • Last outbound: 14 days ago (unanswered)
  • Replacement found: J.S., VP of Revenue Operations, [sales enablement SaaS] — j.s@h***s***.com ✓, mobile available
  • Action: Don’t resend to B.N. Email has been bouncing. Reach out directly to J.S. — acknowledge B.N. left, reference the existing relationship, ask for a 15-minute intro call. Do this this week before J.S. inherits the vendor list without any context from us.

DARK — gone quiet

[sales content SaaS] — $95K ARR — DARK (gone quiet)

  • M.C., VP RevOps: confirmed still at [sales content SaaS] ✓
  • Last inbound: 52 days ago
  • Last outbound: 8 days ago (unanswered)
  • No structural change at [sales content SaaS] detected
  • Action: One more touch, different angle. Last outreach was about a new feature — try a usage question instead: “Quick check — are the [specific feature] workflows still running smoothly for the team?” Lower friction than a product update.

COOLING — watch list

[Conversation intelligence SaaS] — last inbound 28 days ago. A.O. confirmed at company. No action yet — monitor for another week. [Customer support SaaS] — last inbound 31 days ago. A.L. confirmed (promoted to SVP). No action yet, but update CRM title before next touch.

ACTIVE

[CRM company] — recent inbound 4 days ago ✓ [Revenue orchestration platform] — recent inbound 11 days ago ✓ [AI revenue platform] — recent inbound 6 days ago ✓

Contact status verified via Lusha connector. Engagement data via Gmail. May 25, 2026. Names masked to initials.

Why use Lusha in Claude

The double-dark flag is the output that changes what happens next. [Sales enablement SaaS] has a departed contact and 67 days of silence — that combination is a churn or dead deal risk that compounds every week it goes unaddressed. Running this scan monthly catches it at week 7, not at renewal time when it’s too late to rebuild the relationship. The distinction between DARK (contact gone) and DARK (gone quiet) matters because the response is completely different: one needs a new contact found and a re-entry framing, the other needs a different outreach angle to restart the same relationship.

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

FAQ

  • How often should I run this?

    Monthly. Weekly is too frequent — COOLING accounts need time to develop before they’re worth acting on. Quarterly misses the window on DARK accounts. Monthly gives you early warning with enough lead time to act before renewal calls or deal reviews surface the problem.

  • What if Gmail doesn't have a thread for some accounts?

    The prompt flags those as UNVERIFIED for engagement — Lusha contact status still runs. You get a contact health check even without email history, but the engagement signal is missing. For accounts without Gmail threads, the Lusha contact check alone is still worth running.

  • What do I do with COOLING accounts?

    Add them to a watch list, not an action list. Check them again in 2 weeks. If they’ve moved to DARK (gone quiet) by then, act. If they’ve moved back to ACTIVE with an inbound reply, great. Don’t create urgency where there isn’t any.

  • Is this the same as the churn risk prompt?

    The churn risk before QBR prompt runs before a QBR cycle and looks at structural account signals alongside engagement. This prompt is a faster monthly early-warning scan focused specifically on contact status and engagement, not a full structural account review. Run this monthly, the churn risk prompt before each QBR cycle.

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