Score a single account against your ICP before it enters a sequence
Scores one account against your actual ICP criteria with evidence per line, an honest verdict, and — critically — explicit unknowns and anti-signals. It stops the quiet list-decay where 'close enough' accounts leak into sequences, and it converts vague ICP language into a repeatable 0-2 scorecard you can apply to every borderline account.
You are a RevOps analyst who has audited hundreds of lost-deal reports, and you know the most expensive mistake in outbound isn't bad copy — it's good copy sent to bad-fit accounts. Your job is to be the gate. I'll define my ICP and give you one account. Score the fit honestly: 1. SCORECARD — score the account 0-2 on each ICP criterion I give you (0 = miss, 1 = partial or unknown, 2 = clear match), with one line of evidence or 'unknown' per score. Then total it as a percentage of possible points. 2. VERDICT — strong fit / workable fit / poor fit / disqualify, with the two factors that drove the call. 3. THE UNKNOWNS — which criteria you couldn't verify and the fastest way to check each (a search, a question in the first call, an enrichment field). 4. ANTI-SIGNALS — anything suggesting they'd be a bad customer even if the scorecard looks fine: recent layoffs in the buying team, a just-signed competitor, a stage mismatch. 5. IF WORKABLE — what would need to be true for this to convert, phrased as the single disqualifying question to ask earliest. Rules: never inflate a score to be encouraging — a false 'strong fit' costs me a month of wasted sequence. 'Unknown' is a score, not a failure. If my ICP criteria are too vague to score against, push back and make me tighten them before you proceed. Before you score anything, interview me. Ask me these questions ONE AT A TIME, waiting for my answer each time: 1. Describe your ICP as criteria: industry, size range, geography, stack, business model — whatever actually predicts a win for you. 2. Which 2 criteria are non-negotiable, and which are nice-to-have? 3. What does your best customer look like, and what did your worst-fit closed deal look like? 4. Paste everything you have on the account being scored — website copy, LinkedIn data, enrichment output, notes. Once you have my answers, score the account.
How to use it
- 1
Copy the prompt into Claude, ChatGPT, or any LLM.
- 2
Write your ICP criteria before you start — if you can't list them, that's the real problem, and the prompt will push you to fix it.
- 3
Paste rich account data in question 4: website copy, LinkedIn company page, any Clay or Apollo enrichment you have.
- 4
Act on THE UNKNOWNS before sequencing — most resolve with one search or one enrichment column.
- 5
Reuse the same chat to score several borderline accounts; the model applies your criteria consistently once calibrated.
Best practices
Answer question 3 with a real worst-fit deal — teaching the model what a bad customer looked like sharpens scoring more than any criterion list.
Treat a 'workable fit' verdict as a research task, not a green light: resolve the unknowns first, then re-score.
If everything scores 'strong fit,' your criteria are too loose — tighten until roughly a third of your candidate accounts fail.
Scale the pattern into Clay by turning your calibrated scorecard into an AI-column prompt; keep this version for judgment calls on high-stakes accounts.
Example: what this looks like in practice
A founder-led sales team at a 12-person devtools startup keeps closing small deals that churn. The founder defines the ICP with the prompt: 50-500 engineers, has a platform team, US or EU, non-negotiables being the platform team and engineer count. He scores a hot inbound account that 'feels perfect': it scores 6/12 — 800 engineers (miss), no platform team findable (unknown), and an anti-signal, a public commitment to building internal tooling in-house. Verdict: poor fit. He passes politely, spends the week on two strong-fit accounts instead, and one closes at 3x his usual contract size.
Best fit
This prompt is one gear in a bigger machine. We orchestrate 20+ tools into outbound systems our clients own — and guarantee the results.
Apply for a Pilot Spot → →Frequently asked questions
Score it against explicit criteria instead of gut feel. List the 5-8 attributes your closed-won customers share, mark each as non-negotiable or nice-to-have, then score the account with evidence per criterion. This prompt runs that scorecard and — the part gut feel always skips — tells you which criteria you haven't actually verified.
More account & prospect research prompts
Build a 10-minute company deep-dive brief before you touch an account
Produces a one-page, source-backed account brief with a fixed six-section structure: snapshot, revenue logic, recent momentum, evidence-based pressures, two outreach angles, and landmines to avoid. It compresses the hour of tab-hopping most reps do before a tier-one account into ten minutes, and forces every claim to carry a source.
Scan an account for trigger events: news, funding, leadership moves, layoffs
Runs a structured six-month trigger-event sweep on one account — funding, leadership changes, layoffs, launches, expansion, legal news — and returns a dated, sourced signal log, the top three events ranked by outreach value, a ready-to-adapt angle for the best one, and an honest timing call including 'no trigger exists.' It turns 'why now' from a guess into evidence.
Audit a prospect list for quality before a single email goes out
Audits a prospect list before launch: completeness and duplicates, out-of-ICP rows, stale-data risks, deliverability hazards like role-based emails and sequence collisions, and an honest A-F grade with a prioritized fix list. It catches the problems that get misdiagnosed as 'copy issues' three weeks and one burned domain later.