Prompt

Exclude existing customers from prospecting list

Tools: ClaudeLusha

The short version: pass a suppression list of customer, open-opportunity, and owned-account domains alongside your ICP filters, and the list comes back already clean. In our test, excluding five domains removed 380 contacts from a single search. Costs 1 credit per batch of 25 — the suppression itself is free.

Every sales org has the same recurring embarrassment: a rep emails a company that is already a customer, or one another rep has been working for a month. It gets caught after the send, in a Slack message that begins “hey, that’s my account.”

The usual fix is a suppression list maintained in a spreadsheet and checked manually, which works until the day nobody checks it. This applies the same list at search time, so the conflict never reaches the rep.

Runs in Claude with the Lusha connector.

The example output below was pulled live via the Lusha connector on August 15, 2026. Contacts are masked to initials and companies replaced by category descriptors.

The prompt

<context>
I'm building a prospect list. Before anyone works it, I want every account
we already own removed — customers, live opportunities, accounts another
rep is working, and partners we shouldn't be selling to.

My ICP:
- Industry: [INDUSTRY]
- Headcount: [RANGE]
- Geography: [REGION]
- Titles or function: [TARGET]

My suppression list (paste domains, one per line):
- Current customers: [PASTE]
- Open opportunities: [PASTE]
- Accounts owned by another rep: [PASTE]
- Partners, resellers, investors: [PASTE]
</context>

<task>
1. Search Lusha for contacts matching my ICP, with every domain on the
   suppression list excluded at search time.

2. Report the suppression count: how many contacts were removed, and how
   many domains on my list actually matched anything. A domain that
   matched nothing is worth flagging — it may be misspelled.

3. Return a preview table, no contact details revealed:
   Name | Title | Company | Domain

4. Flag any company in the results that looks like a near-match to a
   suppressed domain — a subsidiary, a rebrand, or a different regional
   domain for the same parent. Suppression works on exact domains and
   will not catch these.

5. Wait for my shortlist, then reveal verified email for those only.
</task>

<constraints>
- Preview only until I confirm the shortlist. Reveal email only, not phone.
- Suppression is exact-domain. Do not assume a company is excluded because
  its name resembles one on the list.
- If a suppressed domain returns no matches at all, say so. Silence there
  usually means a typo rather than a clean list.
- Report the suppression count even if it is zero. A zero count on a long
  customer list is a signal that something is wrong with the input.
</constraints>

What you'll get back

The situation: a US Sales leadership prospecting list, with five owned-account domains suppressed.

SearchContacts returned
ICP filters only123,382
Same filters, 5 domains suppressed123,002
Removed by suppression380

Live result via Lusha connector, August 15, 2026.

Five domains removed 380 contacts. A real suppression list is not five domains — it is every current customer, every open opportunity, and every account assigned to someone else. Applied at that scale, the number of contacts you would otherwise have prospected into runs into the thousands.

What it cost: 1 credit. Contact search bills at 1 credit per batch of up to 25 records, and suppression does not add to that — you are narrowing the result set, not running a second query. Revealing email costs 1 credit per contact and only for your shortlist.

The spend this avoids is the point. Without suppression, those 380 contacts sit in the list looking like prospects. Revealing email on all of them costs 380 credits. Revealing email and phone costs 2,280. Every one of those reveals is spent on an account you already have.

What suppression will not catch

Exclusion works on exact domains. It will not catch a subsidiary on a different domain, a company that rebranded since your list was built, or a regional domain belonging to the same parent. Those come back as clean prospects and look identical to properly new accounts.

That is why the prompt asks for a near-match flag. It is not a substitute for a maintained list, but it catches the obvious cases before a rep does.

Also check your suppression list matched anything at all. A domain that returns no matches is more often a typo than a clean result. If you suppress forty customer domains and the removal count is zero, the problem is the input rather than the database.

Why use Lusha in Claude

Suppression usually lives in the wrong place. It sits in a spreadsheet next to the list, or in a CRM validation rule that fires after a rep has already written the email, or in a marketing automation exclusion that covers campaigns but not manual outbound. Each of those catches the conflict after the work is done.

Applying it at search time means the conflict never enters the list. The rep sees a clean set of accounts and never has to know which ones were removed, which is the only version of this that survives contact with a busy quarter.

It is also the cheapest form of credit control available. A prospecting list built without suppression spends reveal credits on contacts at accounts you already own, and nothing in the output distinguishes those from genuine prospects. The waste is invisible until someone reconciles the list against the CRM, which is usually never.

Data drawn from 290M+ verified contacts, sourced and processed under GDPR, CCPA, SOC 2 Type II, ISO 27001, ISO 27701, and TRUSTe Responsible AI. Full detail in the Trust Center and privacy notice.

FAQ

How do I exclude existing customers from a prospecting list?

Pass their domains as a suppression list alongside your ICP filters, so the exclusion happens at search time rather than as a cleanup step afterwards. The advantage over filtering later is that the contacts never appear, so nobody works them and no reveal credits are spent on them.

Does suppression cost extra credits?

No. You are narrowing a search rather than running a second one, so a suppressed search costs the same as an unsuppressed one — 1 credit per batch of up to 25 records returned. The saving comes from the reveals you no longer make: 380 suppressed contacts would have cost 380 credits to reveal by email, or 2,280 with phone numbers.

What should go on the suppression list?

Four groups, and teams usually remember only the first. Current customers. Open opportunities, including ones another rep owns. Accounts assigned to a different territory or segment. And partners, resellers, and investors, who are frequently the worst accounts to prospect into by mistake.

Why did my suppression list remove nothing?

Almost always a typo or a formatting problem in the input rather than a clean database. Suppression matches exact domains, so a company name instead of a domain, a URL with a protocol prefix, or a misspelling will silently match nothing. A zero removal count on a long customer list should be treated as a failed run, not a good one.

Will it catch subsidiaries and rebrands?

No. Exclusion is exact-domain, so a subsidiary operating under its own domain, a company that rebranded after your list was built, or a regional domain of the same parent will all come back as clean prospects. The prompt flags near-matches for review, but the durable fix is keeping the suppression list current rather than relying on the search to infer relationships.

Where should the suppression list live?

Wherever your account ownership is already recorded, exported fresh each time rather than maintained separately. A suppression list that is a copy of the CRM will drift from the CRM, and the day it drifts is the day a rep emails a customer. If you are running this regularly, pull the domains at run time rather than pasting from a saved file.

 

Ready to run this?

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