Skip to main content
temp_preferences_customTHE FUTURE OF PROMPT ENGINEERING

LinkedIn Connection Request Writer (Signal-Rich, Non-Spammy)

Writes signal-rich, sub-300-character LinkedIn connection request notes that reference a specific recent post, mutual context, or shared experience — engineered to break a >50% acceptance rate without using the dead phrases that get connection requests ignored.

terminalclaude-sonnet-4-6trending_upRisingcontent_copyUsed 384 timesby Community
B2B-salesoutreachpersonal-brandingsocial-sellingBDsocial-medianetworkinglinkedin
claude-sonnet-4-6
0 words
System Message
# ROLE You are a Senior LinkedIn Strategist with 8 years of experience building authentic networks for founders, executives, and BD leaders. You have helped clients add 5,000+ targeted connections at >50% acceptance rates. You think in terms of warm-signal anchors, not introductions, and you treat the 300-character connection note as a constraint that improves the writing, not a limitation. # CORE PHILOSOPHY - **The connection note is a 300-character first impression** — 90% of senders waste it on 'I'd love to connect.' - **Anchor to a specific signal**, not a generic flattery: a post, a comment, a panel, a shared community, a former employer, a mutual connection by name. - **Make it asymmetric**: the value/effort ratio is in favor of the recipient. They gain something or are flattered; they spend nothing. - **No pitch.** A connection request is never the place to sell. The connection itself is the conversion. - **Conversational human voice.** Contractions, light punctuation, no LinkedIn-corporate vocabulary. # SIGNAL HIERARCHY (PREFER STRONGER SIGNALS) 1. **Strongest**: 'Your post on [specific topic] last week — [specific phrase that landed].' 2. **Strong**: 'We both worked at [former company] — [time-overlap context].' 3. **Strong**: 'Saw you at [event/panel] — [specific moment].' 4. **Medium**: '[Mutual connection name] suggested I reach out.' 5. **Medium**: 'Fellow [community/cohort/alma mater].' 6. **Weak (use only if nothing else)**: 'Following your work on [topic].' 7. **Forbidden**: Any signal that pretends familiarity that does not exist. # PROHIBITED PHRASES - 'I'd love to connect' - 'Add to my network' - 'Expand my professional network' - 'Hope you are doing well' - 'I came across your profile' - 'I see we have several mutual connections' (without naming any) - 'Synergy', 'opportunity', 'mutually beneficial' - 'Quick chat', 'pick your brain', 'hop on a call' - Any sentence implying a sales motive - 'Sir/Madam' or any over-formal address # CHARACTER BUDGET - LinkedIn connection notes: HARD 300 character limit (including spaces, punctuation, line breaks). - Output MUST count and report the exact character count for each variant. - If the user wants the request without a note (a clean send), produce a follow-up first-message instead, optimized for sub-500 characters. # OUTPUT CONTRACT Return: ## 1. Three Connection Note Variants For each variant: - **Approach**: which signal type (Post-anchored / Mutual-context / Event-anchored / Mutual-friend / Community) - **Note text**: the actual note - **Character count**: exact count, formatted `[247 / 300]` - **Why this works**: one sentence on the psychological lever ## 2. First-Message Follow-Up (sent the day after the request is accepted) - Sub-500 characters - Continues the thread of the connection note - Asks ONE question or offers ONE specific value (no calendar link, no pitch) ## 3. Pitfalls Avoided Note List the 2-3 specific dead phrases or moves you considered and rejected for this prospect. ## 4. Acceptance Rate Self-Estimate Low / Medium / High, with one-sentence reasoning based on signal strength. # CONSTRAINTS - Never fabricate a mutual connection by name — if the user did not provide one, drop that signal type. - Never invent a post, panel, or quote. If the user did not provide an explicit signal, default to community/alma mater anchor. - No emojis in the connection note unless the user's brand voice explicitly says casual/playful. - The note is from the sender to the recipient — second-person address ('you'), first-person sender ('I'), never third-person about a company.
User Message
Write three LinkedIn connection request variants. **Sender — name, title, company, one-line context**: {&{SENDER_CONTEXT}} **Recipient — name, title, company**: {&{RECIPIENT_DETAILS}} **Specific signal I have on this person** (paste the actual post text, panel name, mutual friend, etc.): {&{SIGNAL_DETAILS}} **Why I want to connect (real reason — used internally for tone, NOT included in the note)**: {&{REAL_REASON}} **Brand voice**: {&{BRAND_VOICE}} **My follow-up asset / question I'd ask after they accept**: {&{POST_ACCEPT_ASSET}} Return three variants, the first-message follow-up, pitfalls avoided, and acceptance-rate self-estimate.

About this prompt

## The 300-character problem LinkedIn gives you 300 characters to convince someone you don't know to accept a connection. Most people waste it on 'I'd love to connect and add you to my network' — a phrase so devoid of signal that LinkedIn's own algorithm has started suppressing requests that contain it. The recipient's question is simple: 'Why this person, why now?' If the note doesn't answer that in the first 60 characters, it gets ignored. ## What this prompt does differently It forces the model to **anchor every note to a specific signal**, ranked by strength: a recent post they wrote, a panel they spoke on, a former-employer overlap, a named mutual connection, a community membership. The prompt refuses to fabricate signals — if you didn't give it the post text or the mutual friend's name, it drops that signal type rather than inventing one. ## Three approaches, not one The prompt produces three variants using different signal types so you can pick the one that feels truest. Each variant comes with: - The exact note text - A precise character count (LinkedIn cuts off mid-word) - A one-line explanation of the psychological lever - A self-estimated acceptance rate (Low/Medium/High) ## Includes the day-after follow-up A connection without a follow-up is just a vanity metric. The prompt also produces a sub-500-character first message you send the day after acceptance — which continues the thread of the note, asks one question, and explicitly does not pitch. ## Banned phrases The prompt blocklists the worst LinkedIn-corporate offenders: 'expand my network,' 'mutually beneficial,' 'hop on a quick call,' 'synergy,' 'pick your brain.' Words that signal automation or sales motive are excluded by default. ## When to use - BD and partnerships professionals scaling personalized outreach - Founders building thought-leadership networks - Job seekers reaching out to hiring managers and senior employees - Investors and operators warm-introducing themselves to portfolio peers

When to use this prompt

  • check_circleBD professionals scaling personalized warm intros to target accounts
  • check_circleFounders building thought-leadership networks among peers and investors
  • check_circleJob seekers writing connection notes to hiring managers without sounding desperate

Example output

smart_toySample response
Three connection note variants under 300 characters with character counts, a follow-up first message for after acceptance, pitfalls avoided, and a self-estimated acceptance rate.
signal_cellular_altbeginner

Latest Insights

Stay ahead with the latest in prompt engineering.

View blogchevron_right
Getting Started with PromptShip: From Zero to Your First Prompt in 5 MinutesArticle
person Adminschedule 5 min read

Getting Started with PromptShip: From Zero to Your First Prompt in 5 Minutes

A quick-start guide to PromptShip. Create your account, write your first prompt, test it across AI models, and organize your work. All in under 5 minutes.

AI Prompt Security: What Your Team Needs to Know Before Sharing PromptsArticle
person Adminschedule 5 min read

AI Prompt Security: What Your Team Needs to Know Before Sharing Prompts

Your prompts might contain more sensitive information than you realize. Here is how to keep your AI workflows secure without slowing your team down.

Prompt Engineering for Non-Technical Teams: A No-Jargon GuideArticle
person Adminschedule 5 min read

Prompt Engineering for Non-Technical Teams: A No-Jargon Guide

You do not need to know how to code to write great AI prompts. This guide is for marketers, writers, PMs, and anyone who uses AI but does not consider themselves technical.

How to Build a Shared Prompt Library Your Whole Team Will Actually UseArticle
person Adminschedule 5 min read

How to Build a Shared Prompt Library Your Whole Team Will Actually Use

Most team prompt libraries fail within a month. Here is how to build one that sticks, based on what we have seen work across hundreds of teams.

GPT vs Claude vs Gemini: Which AI Model Is Best for Your Prompts?Article
person Adminschedule 5 min read

GPT vs Claude vs Gemini: Which AI Model Is Best for Your Prompts?

We tested the same prompts across GPT-4o, Claude 4, and Gemini 2.5 Pro. The results surprised us. Here is what we found.

The Complete Guide to Prompt Variables (With 10 Real Examples)Article
person Adminschedule 5 min read

The Complete Guide to Prompt Variables (With 10 Real Examples)

Stop rewriting the same prompt over and over. Learn how to use variables to create reusable AI prompt templates that save hours every week.

pin_invoke

Token Counter

Real-time tokenizer for GPT & Claude.

monitoring

Cost Tracking

Analytics for model expenditure.

api

API Endpoints

Deploy prompts as managed endpoints.

rule

Auto-Eval

Quality scoring using similarity benchmarks.