Claude Prompt Library

Claude Prompts for Sales Scripts That Don't Sound Like Sales

20 copy-paste prompts

20 copy-paste Claude prompts for cold calls, discovery, objections, demos, and closing. Long-context Claude can ingest your call transcripts + spit out scripts that match your team's voice.

Cold + Outbound

4 prompts

Cold Call Script

1/20

Cold call script for [product] to [persona]. Output: 30-second opener (pattern interrupt + permission), 1-line problem (in their words), 1-line proof, single ask (not "demo" — "5 minutes to see if it's relevant"). Branch: if interested → next step, if not → respectful exit + permission to follow up.

Writes cold call scripts.

💡

Pro tip: 5-minute meeting ask > demo ask = 3x conversion. Demo asks = "no, I'm busy." 5-min asks = "fine, what is it?" Lower friction; more meetings; demo earned through interest.

Cold Email Sequence

2/20

Cold email sequence for [product] to [persona]. Output: email 1 (problem-led, not pitch), email 2 (case study or specific insight), email 3 (specific question), email 4 (different angle), email 5 (graceful break-up). Each under 100 words. No "Hope this finds you well."

Writes cold email sequences.

💡

Pro tip: Cold emails over 100 words = unread. Generic openers ("hope this finds you well") = filtered. Specific personal opener + 1 sentence problem + 1 ask = open + reply rates lift.

LinkedIn Outreach Variant

3/20

LinkedIn DM for [product] to [persona]. Different from email: shorter (under 60 words), more conversational, reference their content/profile, no pitch in first message. Output: DM 1 (rapport), DM 2 (light context), DM 3 (specific ask if engaged).

Writes LinkedIn DM sequences.

💡

Pro tip: LinkedIn DMs that feel like email pitches = ignored + worse, perceived as spam. Conversational DMs that respect platform = response rate 3-5x higher than email-pitch ports.

Voicemail Scripts

4/20

3 voicemail variants for [product/persona]: (A) intrigue-led (curiosity hook), (B) reference-led (mutual connection), (C) value-led (specific insight). Each under 30 seconds. Always include callback ask + email follow-up commitment.

Writes voicemail scripts.

💡

Pro tip: Voicemail callback rate is brutal (~5%). Voicemail still triggers email open + LinkedIn check. Plays alongside email = compounds. Solo channel = ignored.

XML tags are just the start. Learn the full Claude workflow.

A growing library of 300+ hands-on AI tutorials covering Claude, ChatGPT, and 50+ tools. New tutorials added every week.

Start 7-Day Free Trial

Discovery + Demo

4 prompts

Discovery Call Script

5/20

Discovery call for [product]. Output: opener (set agenda), discovery questions in order (current state, pain, impact, timeline, budget, decision process), what I want to learn, what I want them to learn about us, transition to demo if qualified or graceful exit if not. 30 min.

Writes discovery call scripts.

💡

Pro tip: Discovery call that pitches = bad discovery. Best AEs talk 30%; let prospect talk 70%. Question depth + active listening = qualified prospects + earned right to demo.

MEDDIC Discovery Mapping

6/20

[Paste discovery call notes]. Map to MEDDIC: Metrics (quantified pain), Economic buyer (named), Decision criteria (their evaluation framework), Decision process (steps + timeline), Identify pain (specific), Champion (internal advocate). Flag missing dimensions = next call focus.

Maps discovery to MEDDIC.

💡

Pro tip: MEDDIC gaps = forecast risk. Missing economic buyer = deal in jeopardy. Missing pain quantification = "nice to have" deal. Mapping discovery to MEDDIC = honest forecast.

Demo Script

7/20

Demo script for [product]. Audience: [decision maker / user / champion]. Output: 5-min opener (recap their stated needs, demo agenda), 15-min product walk (focused on their use cases ONLY, not feature tour), 5-min outcomes/social proof, 5-min next step. Don't feature-dump.

Writes targeted demo scripts.

💡

Pro tip: Feature-tour demos = lost deal. Use-case-focused demo = "they get me + their product solves my problem." Same product, opposite reception. Discipline to NOT show all features = winning move.

Multi-Threaded Demo

8/20

Demo to multi-stakeholder room: [list of personas]. Each cares about different things. Output: opener acknowledges each persona, demo flow that hits each persona's priorities, individual transitions ("Sarah, this addresses your team's X"), Q&A frame inviting each persona. No persona left unaddressed.

Demos multi-stakeholder rooms.

💡

Pro tip: Multi-stakeholder demo where one persona's needs ignored = they kill deal post-demo. Addressing each by name + need = each becomes champion or at least neutral. Inclusive demo = closeable deal.

Objections + Closing

4 prompts

Objection Handling Library

9/20

Build objection handling library for [product]. Common objections: [list — too expensive / not now / using competitor / no need / decision-maker absent]. Per objection: acknowledge, reframe (don't deny), evidence, redirect to next step. 3 variants per objection for different relationship depths.

Builds objection handling libraries.

💡

Pro tip: Objection responses that sound canned = read as canned = lose trust. 3 variants in different tones = pick most natural for the moment. Authentic > scripted.

"It's Too Expensive" Reframe

10/20

Prospect: "It's too expensive." Help me handle. Output: 5 questions to surface real objection (rarely actually price), reframe options (per ROI, per alternative cost of inaction, per competitor cost), where to flex on terms vs price, where to walk. Don't discount reflexively.

Handles price objections.

💡

Pro tip: "Too expensive" = 80% something else (timing, authority, value-perception). Reflexive discount = leaves money on table + signals weakness. Question-discovery → real objection = better path.

Closing Move Scripts

11/20

5 closing moves for [product]. Output: assumptive close, alternative close, urgency close (real urgency, not manufactured), question close, social proof close. Per close: when appropriate, exact words, what NOT to do. Match move to deal stage + relationship temperature.

Writes closing move scripts.

💡

Pro tip: Wrong close at wrong moment = damages relationship. Right close at right moment = closes deal. Reading the room > having all moves memorized.

Stalled Deal Recovery

12/20

Deal stalled at [stage]. Last contact: [days ago]. Help me re-engage: subject line variations (curiosity / value / direct), email body referencing last conversation specifically, single ask, what NOT to do (don't guilt-trip, don't fake-urgency). Get response or move to disqualify.

Recovers stalled deals.

💡

Pro tip: Stalled deals waste pipeline focus. Re-engage with respect → response means re-active deal; no response after 3 attempts = disqualify + free up time. Both outcomes useful; lingering = neither.

These prompts give you the what. Tutorials give you the why.

Learn when to use extended thinking, how to build Claude Projects, and workflows that compound. 300+ tutorials and growing.

Try AI Academy Free

Specialty Scripts

4 prompts

Inbound Lead Response

13/20

Inbound lead just submitted form: [info]. Help me respond fast (within 5 min). Output: opener referencing what they submitted, qualifying questions tactfully, suggested next step, CTA. Speed of response > polish; lead temperature drops 80% per hour.

Responds to inbound leads fast.

💡

Pro tip: Sub-5-minute response wins inbound leads. By 30 min, you're competing with everyone they also filled out. Speed + quality > perfection slowly.

Renewal/Expansion Conversation

14/20

Customer up for renewal: [account info]. Help me prep: usage data review, ROI articulation, expansion opportunity, renewal risk signals, decision-maker engagement plan, timeline + scripts per stage. Renewals are easier than net new; treat with attention.

Preps renewal conversations.

💡

Pro tip: Renewals taken for granted = surprise churn. Renewals worked like sales (proper engagement, ROI articulation, decision-maker check) = 95%+ retention. Treat as new sale.

Win/Loss Interview Script

15/20

Conducting win/loss interview with [former prospect/lost deal]. Help me script: thank them for time, no defensive tone, specific questions about decision factors, what we did well/poorly, competitor comparison, learnings for our process. Interview format, not pitch attempt.

Scripts win/loss interviews.

💡

Pro tip: Win/loss interviews that turn into pitches = you learn nothing. Pure-research framing = honest data. Most teams skip W/L interviews; the discipline is the edge.

Account Plan from Notes

16/20

[Paste 6 months of notes from major account]. Build account plan. Output: stakeholder map with influence, current state of relationship, expansion opportunities, threats (competitors, project at risk, churn signals), 90-day plan, success metrics. Long context lets Claude see whole account.

Builds account plans from notes.

💡

Pro tip: Account plans built from memory = miss patterns. Account plans built from full notes via Claude long-context = uncover relationships + risks invisible day-to-day.

Frequently Asked Questions

If you read them verbatim, yes. Use them as scaffolding — frame the call/email + adapt live. Best AEs internalize the structure; don't recite. Scripts as starting point > scripts as performance.
Long context. Multi-call account history + transcripts + notes all fit in Claude. ChatGPT context limits force chunking. For account-based work, Claude's long context is the unlock.
Don't paste contact PII or sensitive deal data into consumer Claude. De-identify or use Anthropic via Bedrock with BAA-equivalent controls. CRM data privacy may dictate.
No norm yet. Personalized AI-helped outreach indistinguishable from non-AI. Generic AI-helped outreach caught + creates trust loss. The personalization matters more than disclosure.
Claude Opus 4.7 for account planning, complex deals, nuanced messages. Sonnet 4.6 for daily script work + email drafts. Haiku 4.5 for high-volume simple outbound. Long context across all three.

Prompts are the starting line. Tutorials are the finish.

A growing library of 300+ hands-on tutorials on ChatGPT, Claude, Midjourney, and 50+ AI tools. New tutorials added every week.

7-day free trial. Cancel anytime.