ICP score a new target list before the team works it
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 scores a new account list against your ICP before anyone makes a call. Lusha pulls verified firmographic data for each account — headcount, industry, function size, active signals — and scores them into four tiers. The output is a prioritized list with the right first contact for every workable account and a clear reason to skip the ones that don’t fit.
The prompt
<context>
I have a new list of target accounts I want to score against our ICP before handing it to the team. I want to know which accounts are worth working, in what order, and why — based on verified firmographic data, not assumptions.
My list:
- Account list: [PASTE COMPANY NAMES OR DOMAINS — one per line]
- Our ICP criteria:
- Industry: [LIST TARGET INDUSTRIES]
- Company size: [HEADCOUNT RANGE — e.g. 100–1000 employees]
- Geography: [REGION OR COUNTRY]
- Function we sell into: [SALES / MARKETING / FINANCE / IT / OPS / OTHER]
- Signals that indicate fit: [RECENT FUNDING / HEADCOUNT GROWTH / NEW EXEC / OTHER]
- Disqualifiers: [CRITERIA THAT RULE AN ACCOUNT OUT — e.g. too small, wrong industry, acquired]
</context>
<task>
1. For each account on the list, use Lusha to pull verified firmographic data:
- Current headcount and headcount trend (growing, stable, contracting)
- Industry and sub-industry
- Geography and HQ location
- Headcount in the function we sell into
- Any active signals: recent funding, exec hire in the relevant function, M&A activity
2. Score each account against the ICP criteria:
- STRONG FIT: meets all core ICP criteria and has at least one active signal
- FIT: meets core ICP criteria, no active signal
- PARTIAL FIT: meets some criteria, missing one or two — flag which ones
- DISQUALIFIED: fails a disqualifier criterion — flag why
3. For each STRONG FIT and FIT account, identify the right first contact:
- The person most likely to own the problem we solve, given the function we sell into
- Verified title, work email, and direct phone where available
- Flag if no verified contact found in the relevant function
4. Return a scored list:
- STRONG FIT accounts first, ranked by signal recency within the tier
- FIT accounts second
- PARTIAL FIT accounts third — include what's missing and whether it's worth a manual check
- DISQUALIFIED accounts at the bottom — one-line reason each
5. Return a summary:
- Total accounts scored
- STRONG FIT / FIT / PARTIAL FIT / DISQUALIFIED counts
- Top 5 accounts to work first and why
- Any pattern in the disqualified accounts worth feeding back to whoever built the list
</task>
<constraints>
- Score only against the ICP criteria provided. Don't add criteria the rep didn't specify.
- DISQUALIFIED is a clean output — don't soften it to PARTIAL FIT to avoid wasting the list. A disqualified account worked by a rep is wasted time.
- If Lusha can't find firmographic data for an account, flag it as UNVERIFIED — don't score it.
- The top 5 recommendation must explain why each account ranks where it does — not just repeat the score.
</constraints>What you'll get back
The situation: A RevOps lead receives a list of 15 accounts from a recent industry conference. ICP: B2B SaaS or FinTech, 150–800 employees, North America, selling into Sales and Revenue Operations teams. Disqualifiers: under 100 employees, acquired as target entity, non-English-speaking markets.
Output: 4 STRONG FIT, 5 FIT, 3 PARTIAL FIT, 2 DISQUALIFIED, 1 UNVERIFIED. Top 5 identified with specific reasons. One pattern flagged in the disqualified accounts.
Summary
15 accounts scored · 4 STRONG FIT · 5 FIT · 3 PARTIAL FIT · 2 DISQUALIFIED · 1 UNVERIFIED
ICP: B2B SaaS / FinTech · 150–800 employees · North America · Sales and RevOps function. Run: May 19, 2025.
Top 5 accounts to work first
1. Halcyon Ventures — STRONG FIT 310 employees, SaaS, US. New CRO hired 18 days ago — strongest signal on the list and most recent. RevOps team of 12. First contact: J.K., CRO · j.k@[halcyon].com ✓ · mobile verified. Lead with the new CRO angle — 60-day audit window is open now.
2. Bright Arc Systems — STRONG FIT 420 employees, FinTech, US. Series B closed 6 weeks ago, sales headcount up 28% in Q1. RevOps team growing. First contact: M.T., CRO · m.t@[brightarc].com ✓ · mobile verified. Series B + headcount growth = budget unlocked and data layer under pressure.
3. Waverly Digital — STRONG FIT 190 employees, SaaS, Canada. New VP of Sales 22 days ago — first VP Sales hire, previously founder-led. First contact: S.R., VP Sales · s.r@[waverly].com ✓ · mobile verified. First VP Sales at a founder-led company is the highest-intent signal for sales tooling.
4. Kestrel Labs — STRONG FIT 155 employees, HR Tech (adjacent ICP — sales motion applies), US. New VP of Sales 27 days ago. First contact: D.P., VP Sales · d.p@[kestrel].com ✓ · email verified, mobile not found. Signal is fresh. HR Tech is adjacent — worth the outreach given the exec hire timing.
5. Finova Group — FIT 180 employees, FinTech, US. No active signal but strong ICP match — size, industry, function. First contact: A.M., VP Sales · a.m@[finova].com ✓ · mobile verified. No signal today — run this account again in 30 days or use the expansion signals scan to monitor.
STRONG FIT — work this week
| Account | Size | Industry | Signal | First contact | Details |
|---|---|---|---|---|---|
| Halcyon Ventures | 310 | SaaS | New CRO 18 days ago | J.K., CRO | j.k@[halcyon].com ✓ |
| Bright Arc Systems | 420 | FinTech | Series B + 28% headcount growth | M.T., CRO | m.t@[brightarc].com ✓ |
| Waverly Digital | 190 | SaaS | New VP Sales 22 days ago | S.R., VP Sales | s.r@[waverly].com ✓ |
| Kestrel Labs | 155 | HR Tech | New VP Sales 27 days ago | D.P., VP Sales | d.p@[kestrel].com ✓ |
FIT — work this week after STRONG FIT
| Account | Size | Industry | First contact | Note |
|---|---|---|---|---|
| Finova Group | 180 | FinTech | A.M., VP Sales ✓ | No live signal — monitor |
| Thornwick Media | 220 | SaaS | C.S., VP Revenue ✓ | Strong ICP, no signal |
| Dune Analytics | 280 | Data/Analytics | R.C., CRO ✓ | Adjacent industry, strong size fit |
| Corelink SaaS | 310 | SaaS | J.L., VP Sales ✓ | Clean ICP match |
| Aster Platforms | 175 | FinTech | S.B., Head of RevOps ✓ | No signal, solid fit |
PARTIAL FIT — manual check before working
| Account | What fits | What’s missing |
|---|---|---|
| Novela Group | Industry, geography | 94 employees — below 100 minimum. Check if headcount has grown recently before assigning. |
| Pallet Dynamics | Size, geography | Logistics software — outside target industry. Worth a manual check if the sales motion applies. |
| Bridgewater Labs | Size, industry | HQ in UK — outside North America. Check if they have a US office before assigning to a North America rep. |
DISQUALIFIED — do not assign
Cartway Technologies — acquired as target entity 9 weeks ago. Contracts under integration review. Remove from list.
Sola Systems — 41 employees. Below minimum headcount threshold. Remove from list.
UNVERIFIED
Meridian Co (not Meridian Logistics) — Lusha cannot confirm firmographic data for this entity. Two companies with similar names returned. Manual check required before scoring.
Pattern in disqualified accounts
Both disqualified accounts appear to have come from a conference list compiled by company name without size or status verification. Worth flagging to whoever sourced the list — adding a headcount filter and an M&A status check at the point of list creation would remove these before they reach RevOps.
All firmographic data and contacts validated via Lusha connector, May 19. Names masked to initials, emails abbreviated for privacy.
Why use Lusha in Claude
A list of 15 accounts handed to a rep without scoring means 15 accounts worked in arbitrary order — some strong fits, some wrong industry, some too small. Lusha in Claude scores every account against the ICP in one pass, identifies the right first contact at every workable account, and tells the rep exactly why the top 5 are top 5. The DISQUALIFIED output is as useful as the STRONG FIT output — it stops two accounts from consuming rep time that should go to the four accounts with live signals. The pattern flag at the bottom feeds back to whoever is sourcing lists, which improves the next list before it arrives.
Data drawn from 300M+ verified contacts under GDPR, CCPA, SOC 2, ISO 27701, ISO 31700, and TRUSTe.
FAQ
What if our ICP has more criteria than the prompt covers?
Add them to the ICP criteria field — the prompt scores against whatever criteria you provide. If you have five ICP dimensions, list all five. The scoring adjusts to match. The four tiers stay the same regardless of how many criteria you specify.
How is PARTIAL FIT different from DISQUALIFIED?
DISQUALIFIED means the account fails a hard rule — a disqualifier you specified up front. PARTIAL FIT means the account meets most of the ICP but is missing one or two criteria that aren’t hard rules. The distinction matters for how you handle it: DISQUALIFIED gets removed from the list, PARTIAL FIT gets a manual check before you decide.
What if most of the list comes back DISQUALIFIED?
That’s the most useful output the play can return — it means the list was built without ICP filtering and needs to go back to the source. The pattern flag in the summary tells you why, which feeds directly back to the marketing, events, or data team that built it.
Can I use this for inbound leads too?
Yes — the scoring logic works the same way for an inbound list. Replace “conference list” with “inbound leads from last week” in the context field. The output is the same: which ones are worth a same-day call, which ones need a manual check, which ones should be routed to a different team or market.
What if Lusha can't find data for several accounts on the list?
UNVERIFIED accounts are flagged and excluded from scoring — the prompt doesn’t guess. If a significant portion of the list comes back UNVERIFIED, that’s a list quality problem: company names may be misspelled, too generic, or not indexed. Providing domains alongside company names improves Lusha’s match rate significantly.
Who should run this — RevOps or the sales manager?
Either works. RevOps typically runs it before distributing a new list to the team. A sales manager runs it when a list arrives from marketing or an event and needs to be assigned quickly. The output format is the same — tiered, with contacts attached — so whoever runs it can hand it to reps immediately.
Ready to build this?
Get started with Lusha and set up this play in minutes.
