steal what
already works.
a live radar of proven linkedin posts and lead-magnet plays worth borrowing before you build your own.
build yours freebrowse 50 proven posts from the Discover library. free preview: 50 items.
😲 Le prix d'un 𝐚𝐮𝐝𝐢𝐭 𝐆𝐄𝐎 vient de chuter de 90 % grâce à 𝐉𝐞𝐯 J'ai vu passer des audits GEO facturés jusqu'à 10 000 €. Maintenant, il y a Jev. Et ca change le GAME. Je viens de le tester et il est vraiment hors norme. Tu lui donnes une URL. En quelques minutes, il te sort le tableau de bord complet : 1/ 𝐓𝐚 𝐯𝐢𝐬𝐢𝐛𝐢𝐥𝐢𝐭𝐞́, 𝐦𝐨𝐭𝐞𝐮𝐫 𝐩𝐚𝐫 𝐦𝐨𝐭𝐞𝐮𝐫 : Combien de fois tu es cité sur ChatGPT, Claude, Perplexity, qui te dépasse, pourquoi ... 2/ 𝐓𝐨𝐧 𝐒𝐜𝐨𝐫𝐞 𝐆𝐄𝐎 𝐠𝐥𝐨𝐛𝐚𝐥 : Où tu en es, et le pourcentage de gains qui dort encore sur la table. 3/ 𝐓𝐚 𝐜𝐨𝐮𝐯𝐞𝐫𝐭𝐮𝐫𝐞 𝐝𝐞𝐬 𝐢𝐧𝐭𝐞𝐧𝐭𝐢𝐨𝐧𝐬 : Info, prix, comparatif, avis, achat : sur quels sujets l'IA te trouve, et où tu es absent. 4/ 𝐓𝐚 𝐜𝐢𝐭𝐚𝐛𝐢𝐥𝐢𝐭𝐞́ 𝐩𝐚𝐫 𝐭𝐲𝐩𝐞 𝐝𝐞 𝐩𝐚𝐠𝐞 : Quelle page a une vraie chance d'être reprise, et laquelle ne sera jamais citée. Le genre de diagnostic qui prenait une semaine à un consultant SEO. C'est divisé par 10 maintenant. Mais un audit te donne le QUOI. Pas le COMMENT. Alors j'ai écrit un 𝐏𝐋𝐀𝐘𝐁𝐎𝐎𝐊 gratuit : comment utiliser Jev de A à Z pour passer d'une page invisible à une page que l'IA va chercher. La méthode exacte, corrections prioritaires en premier. Check l'image pour y accéder 👇
i (actually) turned Claude in a construction team. → hire #1: the drawing reader turns a PDF drawing set into a database you can question. ↳ 89-page set. "how many doors?" → 147, straight from the schedule on A601 ↳ every number tagged HIGH or LOW → hire #2: the spec reviewer reads the whole spec before you bid. ↳ real tender spec: 18 of 19 bid-killers found ↳ $2,500/day LDs. 10% bid security. bid due 2:01pm. ↳ every finding cited to the exact page → hire #3: he bid hunter watches SAM.gov for federal contracts. ↳ live feed, filtered to 28 general-contracting NAICS codes ↳ scores every contract 0-100% against your company ↳ deadline countdown + a bid draft pre-filled with the contract's details ↳ every number tagged HIGH or LOW → hire #4: the estimator reads the BIM model and prices it. ↳ IFC in → priced BOQ in Excel out ↳ every line classified to CSI, flagged when unsure → hire #5: the BIM coordinator audits the Revit model before it goes out. ↳ grades it A to F ↳ missing parameters, bad classifications, clickable failing elements → hire #6: the compliance checker checks the Revit model against the written spec. ↳ 379 checks, each one cited to the spec section + page ↳ 0 fails. but 57 doors with no fire rating entered at all ↳ PASS, FAIL or MISSING. missing is never called compliant. → hire #7: the cost controller loads the BOQ into Primavera P6. ↳ 1,332 BOQ lines. 45 seconds. 91% loaded automatically ↳ totals checked to the cent ↳ weak matches never loaded. it waits for the planner → one honest limit none of them signs anything off. agents do the work. humans approve it. that's the whole team. ps: comment 'TEAM' and i'll send you all 7. free.
I built a complete Claude team for LinkedIn. And I'm giving all the 17 agents system for free ↓ ( THIS IS THE SYSTEM – too many comments, no need to comment anymore: hope it helps: https://lnkd.in/gttpMtZB ) Over the last few months, I've been studying where Claude can actually improve LinkedIn content. It obviously can't replace your thinking or your ideas. But it can remove the repetitive work slowing you down. Here's what the team handles: 1. AI clone Builds your 3,000-word AI clone. Maps tone, rhythm, banned words, and sentence patterns from your LinkedIn, website, and X. 1. Research & trends Scans Reddit, X, and news daily. Surfaces angles in your ICP's language and finds what's trending before everyone else. 1. Hook generation Pulls from 17,836+ viral patterns across 100M+ impressions. Matches every hook to your voice profile. 1. Content writing Drafts op-eds, lead magnets, announcements, and build-in-public posts that sound like you. 1. Infographics Turns any idea into a finished infographic from 41+ visual styles. One prompt. 1. Analytics & tracking Reads your LinkedIn analytics via official API. Tracks 40 viral creators weekly and flags what's spiking. 1. Scheduling & distribution Queues a full month via LinkedIn's official API. Alternates formats. No manual publishing. Everything connects: AI clone sets the foundation → Research finds the angles → Hooks stop the scroll → Writing delivers the post → Infographics make it visual → Analytics show what's working → Scheduling keeps it running. Each Claude agent handles one specific job you're already doing manually. That way, Claude handles the repetitive work. You focus on the ideas worth sharing. 8 specialized agents. 17 LinkedIn skills. One complete Claude team. Get it here 👇
Real Estate focused #Investment Management Vehicles: SPVs/ Portfolios/ Holding Companies/ Platforms/ GP-LP #Funds - #GLOBAL Footprint Are you a General Partner/ Sponsor/ Principal/ Manager looking to raise #Capital for your Real Estate Vehicle ? * Cheque Size: $5M to $50M, and in some cases based on merit, up to $100M * Participation depends on: GP Legacy, Past Performance, Anchor Investor/s, Current Committed Capital, Investment Thesis, Target selection Criteria, Investment Committee structure, Exit Strategy, Holding Period, Hurdle/Coupon Rate, Projected IRR, Legal & Tax Jurisdiction * Target Geographies: Tier 1: #USA, Canada, #UK, Ireland, West Europe Tier 2: #Australia, NZ, #UAE, KSA, Qatar, India, SE Asia * Preferred Themes: #Residential/ #Commercial/ #Hospitality/ #Industrial/ #Retail/ Data Centers/ Specialized #Housing Projects * Comment Capital to get access
ChatGPT is catching up to Claude. I've been opening it more and more inside my own marketing system. And honestly it's pretty great. So I wrote the complete guide: ChatGPT for Marketing 2026. When I test both, Claude still wins 8 out of 9. Claude still runs my operation. But the gap is closing. And ChatGPT is the tool most teams already pay for. The problem is how they use it. - Like a search bar -One prompt - Generic answers - Start over tomorrow Every section of the guide is a workflow with a time budget, its own prompts and template, and a checklist to run before anything ships. Heres whats inside: → Marketing strategy and planning → Market and customer research → ICPs, segmentation and positioning → Brand messaging and offers → Content strategy and production → SEO, AEO and AI search visibility → Social media and LinkedIn → Email, lifecycle marketing and CRM → Paid advertising and creative testing → Landing pages and conversion optimisation → Analytics, attribution and reporting → ChatGPT workflows and marketing automation Every section gives you something to use today: ↳ Editable templates ↳ Reusable prompts ↳ Checklists ↳ Step-by-step workflows Do the setup, then pick the one section costing you the most right now. Most teams pay for ChatGPT, but skip the setup, and wonder why every draft sounds the same. How are you using ChatGPT for marketing? If you want the free guide, just follow me and comment "ChatGPT Guide" and I'll send it to you.
Hyper-realistic POV hooks are breaking ad feeds right now. The fastest way to hold viewer attention I built an agent in MakeUGC that builds these ultra-realistic concepts in minutes: - Feed it a simple core angle or creative direction - You approve the vision - It renders the final cut with Seedance 2.5 on complete autopilot What do you guys think of this output? Drop a comment below and I'll send you 2 full guides: - How to write prompts that nail hyper-realism - How to build ad templates in any style (Disney, LEGO, video games, UGC, cartoons, and more)
I talk to Claude like it owes me money. Half sentences. Wrong order. Sentences like 'that thing in that folder, see attached'. It knows what I mean. Every time. Not magic. Just a skill I built. Meet The Prompt Optimizer. Nobody has time to prompt properly. And bad prompting is bad for business, especially on big models like Fable 5.1 and Opus 5.5. So you ramble. In writing or, like me, out loud using Wispr Flow. Full sentences or crumbs. It does not judge. It finds 'whatever you are trying to say' and turns it into something intelligent. Then it hands Claude a prompt that makes the machine sing. So you get the answer you actually wanted. First time. Less back and forth. Fewer wasted tokens. Better results. I've turned the skill into a playbook so everyone can have it. For free. Read the instructions. Installation takes a few minutes. (Less if you don't ramble.) 👉 Like the post, comment 'PROMPT', and I'll send it over. 🤟 Make sure we're connected, or it can't find you. Using OpenAI's ChatGPT? Say so below. If enough of you ask, The Optimizer will learn a second language. Anthropic #AI #Claude #PromptEngineering #GenerativeAI ------------------ I'm Rebecca Rowntree - Creative Consultant, AI Creative Technologist, Founder of Get Sh*t Done. I speak on world stages. I run an 800+ member creative community. I host events that sell out in a day. I lead AI workshops for big brands. I build AI tools and iOS apps, direct AI films and more. 👉 Follow for more AI stuff. Funny. Useful. Never boring.
Claude Sonnet 5.5 is out. I made 20 Claude Skills for paid ads to run on it. Free. Open source. A smarter model still does not know your account. It reads the same exports, and without rules it just guesses faster. So I keep the judgment outside the model. 1/ Paid Ads Account Triage 2/ Cross-Channel Budget Reallocation 3/ Campaign Structure Debt Audit 4/ Wasted Spend Finder 5/ Search Terms and Intent Hygiene 6/ Creative Fatigue and Angle Review 7/ Landing Page Conversion Review 8/ Offer and Message Match Audit 9/ Tracking and Signal Quality Audit 10/ Lead Quality Feedback Loop 11/ Performance Anomaly Explainer 12/ Bidding and Learning Sanity Check 13/ Performance Max Control Review 14/ Advantage Plus Control Review 15/ Product Feed and Catalog QA 16/ Comment and Objection Miner 17/ Competitor and Auction Pressure Review 18/ Experiment Readout 19/ Weekly Paid Ads Operator Readout 20/ Scale Readiness Check Each one has required inputs, a workflow, decision rules, an output format and guardrails. They work across Google Ads and Meta together, which is the part that usually breaks. Budget moves between channels, tracking sits underneath both, and most tools make you answer the question twice. They read exports. They do not log into your account. They do not change campaigns. They do not touch budgets. Two of them will refuse to answer. When the window is too short to separate noise from a real shift, the anomaly explainer will not name a cause, and the experiment readout will not declare a winner the data cannot carry. Swap the model underneath and the rules stay the same. That is why I write them down. They help you decide: scale it, fix it, monitor it, or ask for data.
The LinkedIn Pipeline Guide to Claude is here. 3 agents, 1 MCP, 9 steps, and the exact prompts 👇 Content to CRM-logged reply - without opening a dashboard, manually importing a prospect, or switching tabs once. This is what the PDF is about: 1/ The research agent that finds what your ICP is already complaining about Reddit threads, LinkedIn comments, niche communities - scoured and ranked into 5 content ideas, all in your ICPs real language. Youre not guessing what to post. Youre writing what theyre already searching for 2/ The voice builder + copywriter agent One skill file that knows how you write. One agent that tightens every draft before it goes live. The post sounds like you, because it was trained on you. 3/ Signal harvesting from your own content Your post goes live. The comment section fills up. Claude pulls every commenter into a campaign - people who just told you they care about this exact topic, in public, unprompted. 4/ Enrichment before anything sends Job title, company, recent activity, connection degree - all enriched per prospect before the sequence runs. If you have LeadMagic connected, verified emails get pulled at the same stage as a fallback channel. 5/ AI personalization that isnt a mail merge One variable, one prompt. Every person gets a unique icebreaker written from their actual activity - not a template with their first name dropped in 6/ The sequence built in one conversation Connection request, follow-up, final message. Written, queued, and launched without opening the dashboard. 7/ The inbox handoff - and what happens after Morning summary of every reply, classified. Tag a conversation as hot lead and Zapier fires: CRM contact created in HubSpot, Pipedrive, or Salesforce, Slack pinged, email sequence triggered. Plus: the one step you dont think about - and why skipping it is how campaigns go sideways. LMK if you want the full guide - every agent prompt, the voice skill setup, all MCP prompts in one doc, and which int
GOODBYE ACCENTURE. HELLO CLAUDE OPUS 5.5. Accenture reports its year-end results this morning. Everyone will watch the AI numbers. I'd watch something else. The work a big consulting project team spends twelve weeks producing now fits in a library of Claude skills. So I built that library: 150 Claude skills across 15 practices, the full scope of a McKinsey, Deloitte or Accenture engagement 👇 1. Strategy: problem framing, issue trees, market sizing, growth, pricing, the board memo 2. Finance & risk: FP&A, cost and margin, controls, M&A diligence, org design 3. Operations: process redesign, supply chain and procurement, technology and AI rollout, customer service, program delivery and the benefits tracker Every skill is a real SKILL.md you save in a folder. It tells Claude what to ask you for, the method a senior consultant runs, the exact output format, and the checks to run before it hands anything back. Five minutes to install. Built for Claude Opus 5.5, which reads a million tokens at once, so the whole data room goes in. Run one against the next deliverable on your desk. Judge it the way you'd judge a sharp analyst's first draft. The skills produce the deck, the model and the memo. Someone still has to stand with the team the week the new process goes live. Comment the word on the card and I'll send you all 150 skills, free.
I built a sales team out of AI models that only answer yes or no. And I'm giving the full build away for free. A decision model takes a question and a fixed set of answers, then picks one in milliseconds. One founder checked 25,033 LinkedIn connections against his ICP in 71 seconds. 435 came back warm. I turned that into 10 sales decisions. Each one comes three ways: → A prompt you paste into Claude or ChatGPT → An agent that sorts a whole list for you → An n8n workflow that runs it automatically The 10: → Is this new connection someone we sell to? → Is this commenter a buyer, a peer or a competitor? → Is this reply interested, not now or no? → Is this company hiring for our buyer? → Did this company just raise money? → Is this lead already in the CRM? → Call or nurture email for this inbound lead? → Can this person sign off the budget? → Is this deal going quiet? → Which rep owns this lead? You also get: - The answer set for each, already written - Which model to run each on: Jev, OpenAI or Laya - A test sheet to check them on your own data - Setup steps for non coders Run your first one in 10 minutes. FREE ACCESS TO THE FULL BUILD 👇
GOODBYE 10-DAY CLOSE. HELLO CLAUDE OPUS 5.5. The Complete Claude Finance Playbook is here. 20 Claude agents that run the whole finance cycle, from the budget to the board pack 👇 Five stages. Four agents in each. 1. Forecast & Budget: agents that build the plan and keep it current Driver-Based Budget Builder, Rolling Forecast Updater, Scenario Planner, Headcount & Comp Planner 2. Cash & Working Capital: agents that bring cash in faster and pay it out on purpose 13-Week Cash Forecaster, Collections Prioritizer, Payables Timing Planner, Spend Leak Finder 3. Month-End Close: agents that tie out the books and explain every variance Close Calendar Runner, Account Reconciler, Journal Entry Reviewer, Flux Explainer 4. Controls & Audit: agents that have the evidence ready before the auditor asks Controls Tester, Audit Request Pack Builder, Expense Policy Checker, Revenue Recognition Reviewer 5. Reporting & Decisions: agents that turn the numbers into decisions Management Report Writer, Capital Request Evaluator, Price-Volume-Mix Analyzer, Tough Questions Rehearser Each one is a copy-paste Claude skill. It asks for your exports, runs the method a senior finance person would run, and hands back the tables with its own checks already done. Opus 5.5 reads a full trial balance in one pass, so you can hand it the real export. You still sign the reconciliation and make the call on the estimate. The agents do the hours of prep that come before it. Comment the word on the card and I'll send you all 20 agents, free.
I built our entire GTM research function as 12 Claude skills, and Im handing over the whole vault. GTM teams running LinkedIn research today do it manually. One competitor scroll, one gut call on what to post, no connection between what the market is actually saying and what ends up on the feed. The content stays random because the research is random. So I packaged the research layer into one skill per job Claude can load by name: 01 Competitor Post Breakdown - dissect whats driving any competitors best posts 02 Hook amp; Opening Line Analysis - study which first lines stop the scroll 03 Content Gap Analysis - find what nobody in your niche is covering yet 04 Engagement Pattern Analysis - diagnose why a post is or isnt taking off 05 Audience amp; ICP Mapping - map exactly who engages, by title, seniority, and company 06 Industry Conversation Mapping - locate the live debates your buyers are already in 07 Content Calendar Reverse-Engineering - pull apart a competitors full cadence and content mix 08 Format Performance Analysis - know when text, carousel, video, or poll actually wins 09 Creator amp; Influencer Benchmarking - find whos growing fast and extract the pattern 10 Comment Mining for Content Ideas - turn comment sections into a content pipeline 11 Personal Brand Positioning - find the angle that separates you from every other voice in the space 12 Posting Cadence amp; Timing - post when your ICP is actually on the feed That is the GTM Research Stack I actually run: competitor intel, hook research, gap mapping, engagement diagnostics, audience data, industry conversations, calendar deconstruction, format logic, creator benchmarks, comment mining, brand positioning, and timing 12 skills, built for any agent that reads SKILL.md. Stop guessing what your market wants to read. Own the research. Its yours below, free. LMK if you want the full vault.
😐 This tool killed Photoshop It might be the best AI Image Tool I've seen And I've seen a lot.... I want to show you how to master it Of course I'm talking about ChatGPT Images 2.5 Why it's a big deal: Reference fidelity: your product & faces stay identical across new scenes. Precision editing: change one colorway, prop, or label, and nothing else moves. Multi-turn consistency: edit 6 times in a row without quality degrading. Text & layout control: exact copy, safe zones, and defined layouts. Fast, pro turnaround: up to 50% faster, so variant sets take minutes. Sketch-to-image: rough a banner or PDP section, get a finished concept. On-brand sets: 1:1, 9:16, and banners that all read as one campaign. For marketing teams Ecommerce sellers Brands Agencies This is a MUST learn That's why I've pulled together A simple yet comprehensive guide to ChatGPT Images 2.5. It includes Prompts Tips Use Cases 👉 If you need this guide, let me know. I pulled it the help my clients, but I am sharing it here as well. And btw: do yourself a favour and save this post. I'll respond later with a link to your comment, which will make it much simpler to get your guide.
I built a complete Claude Fable 5.1 LinkedIn system. It's dangerously good at helping you win attention and leads 👇 ( THIS IS THE SYSTEM – too many comments, no need to comment anymore: hope it helps: https://lnkd.in/gttpMtZB ) Over the last few months, I've been studying where AI can actually help creators with LinkedIn. Here's what I found: It can't replace your thinking or your ideas. It CAN remove the repetitive work that eats up hours every week: 1. Researching what to post 2. Finding trending angles 3. Matching hooks to your voice 4. Formatting and editing 5. Scheduling and tracking Here's what it does better: 1/ Clones your voice Analyzes your LinkedIn, website, and X. Builds a 3,000-word AI clone that captures tone, rhythm, and sentence patterns. 2/ Finds your best angles Scans Reddit, X, and news daily. Surfaces trending topics in your ICP's real language, not yours. 3/ Writes hooks that stop the scroll Pulls from 17,836+ viral patterns across 100M+ impressions. Matches every hook to your voice profile. 4/ Drafts full posts in your voice Op-eds, lead magnets, announcements, build-in-public. Not generic advice. Content your audience actually needs to hear. 5/ Creates infographics that get shared 41+ visual styles mapped to content types. Idea in, finished infographic out. One prompt. 6/ Tracks what's working Reads your LinkedIn analytics via official API. Tracks 40 viral creators weekly. Flags what's spiking. 7/ Schedules and publishes on autopilot Queues a full month via LinkedIn's official API. No extensions, no manual publishing. It doesn't recommend random LinkedIn tactics. It builds the complete content system: clone → research → hook → write → design → post → analyze → improve The free full system — 8 agents, 17 skills, one loop — is below 👇
I built a Claude 5 workflow that replaced 12 hours of weekly LinkedIn lead magnet work. Check this out! Last week, I shipped 6 fully-researched lead magnets with this workflow. Here's the exact workflow: 1️⃣ Pull winning lead magnets from your niche inside Claude Claude collects every lead magnet post in your niche from the last 30 days. Extracts hook patterns, keywords, resource formats and comment counts. Get a structured swipe report in 1 chat. 2️⃣ Map angles by buyer pain + build the hook bank Claude clusters winning hooks by ICP pain point. Cross-references your buyers' own words with the highest-frequency hook patterns. Output: 5 ranked angles + a 20-hook bank organized by angle. 3️⃣ Generate the post stack + capture plan Claude turns each angle into a finished post: hook, body, resource outline, keyword. Ranks all variations by predicted comments based on niche signal. Outputs the delivery DM, follow-ups and trigger setup for each one. The results? Lead magnet research compressed from 12 hours to 15 minutes. 6 post-ready lead magnets per chat instead of 1 per week. Capture setup tied directly to live comment signal. Now I'm giving away access for free 👇
I Built 200 Claude Agents To Run Your Entire LinkedIn And I'm Giving The Complete Collection Away For Free👇 Most teams still use Claude like a chatbot. I wanted to see what happens when you build an actual LinkedIn team inside it. So I created 200 specialized agents across: → Coaches → Recruitment agencies → Agencies → GTM teams → B2B sales Including agents for: → ICP from closed-won → Proof librarian → Profile optimization → List building → Signal finding → Fit filtering → Opener writing → Sequence building → Engagement sessions → Comment scoring → Reply classification → Objection handling → Next-step closing → Campaign diagnostics → Weekly reporting Each agent is built around one specific job, and every one has a NOT YOUR CALL line so they stop doing each others work. So instead of opening Claude and figuring out what to prompt every time, you can pick the agent you need and put it to work. 200 agents. 5 verticals. One complete library. Every agent drafts. None of them send. Free access 👇
A RevOps leader last week: "but Janis, I've used Claude for months now, why do I still struggle to make it work for me?" I know why: Because you haven’t set it up correctly in the first place 𝟭. 𝗬𝗼𝘂'𝗿𝗲 𝘂𝘀𝗶𝗻𝗴 𝘁𝗵𝗲 𝘄𝗿𝗼𝗻𝗴 𝗺𝗼𝗱𝗲 - Chat: one-off tasks. Paste a pipeline export and ask what's at risk this week. - Projects: recurring work that needs your context. A "RevOps [Company]" project with your stages, ICP and definitions loaded once. - Cowork: delegated multi-step work. Every Friday, pull pipeline metrics, build the weekly report and send it. - Code: technical builds. Write the SOQL query, validation rule or data cleanup script. - API: always-on automation. Score every new lead against your ICP the moment it hits Salesforce. 𝟮. 𝗬𝗼𝘂'𝗿𝗲 𝗻𝗼𝘁 𝗴𝗶𝘃𝗶𝗻𝗴 𝗶𝘁 𝘁𝗵𝗲 𝗿𝗶𝗴𝗵𝘁 𝗰𝗼𝗻𝘁𝗲𝘅𝘁 Claude doesn't know your business. Give it these once, in Project instructions: - Company and ICP - Segments - Sales process and stage exit criteria - Methodology (MEDDIC, etc.) - Metrics and targets - Team structure and territories - Fiscal calendar - Tech stack and who reads the output 𝟯. 𝗬𝗼𝘂𝗿 𝗽𝗿𝗼𝗺𝗽𝘁𝘀 𝗵𝗮𝘃𝗲 𝗻𝗼 𝘀𝘁𝗿𝘂𝗰𝘁𝘂𝗿𝗲 Use: role, data, numbered tasks, format. For example: "You're a RevOps analyst. Below is our [SEGMENT] pipeline export as of [DATE]. Summarize: (1) total pipeline value and deal count by stage, (2) deals with no activity in 14 days, (3) top 3 deals at risk with reasoning, (4) recommended actions for this week's forecast call. Be concise. Use bullet points. [PASTE DATA]" 𝟰. 𝗬𝗼𝘂'𝗿𝗲 𝗻𝗼𝘁 𝗮𝘀𝗸𝗶𝗻𝗴 𝗳𝗼𝗿 𝘁𝗵𝗲 𝗲𝘅𝗮𝗰𝘁 𝗼𝘂𝘁𝗽𝘂𝘁 𝘆𝗼𝘂 𝘄𝗮𝗻𝘁 Output should be different based on who receives it: - CRO / Board: number first, 3-5 bullets, no jargon - VP Sales: deal name, stage, risk reason, next action - Finance: precise numbers with variance explained - Salesforce admin: exact field names, inline comments - RevOps team: owners, deadlines, reusable templates - Enablement: plain langua
I updated my "Hidden Job Market" spreadsheet: 125+ companies that DON'T post their jobs on LinkedIn, plus 48 niche job boards to try and 39 networking communities to join... In a tough market, Greenhouse's CEO gave this advice last week: "apply to less-known companies" That's what this list is for, and I checked 1,000+ employers to build it Like + Comment for the link! *PLEASE CONNECT AND DM to speed things up- I'm out of connection requests for the week
I ran 210 of my own LinkedIn posts through an AI check. It took under a minute and cost less than 5 cents. Now every draft gets a score before I post it. I'm testing JEV by TypeSafe AI. It doesn't write anything. It answers 55 questions about every post: the hook, how concrete it is, the call to action, whether there's a story. Then it matches those answers to how each post actually did. For me, top 20% means 11,000+ impressions. The bottom 20% stays under 600. What it found: Concrete posts hit my top 20% six times more often than abstract ones. I ended 78 posts with an open question. 3 of them made the top 20%. Good to know, and a bit painful. The test: show it one of my top posts and one of the rest. It picks the right one 88% of the time. By hand, at 10 seconds per question, that's 32 hours. So I never did it. Then I wrote my first post about JEV. It scored 88 out of 100 and predicted 16,000 impressions. Want this post checker? Let me know in the comments and I'll send it.
I’ve started a small WhatsApp community for founders and growth practitioners. We’re calling it B2B Growth Lab. The idea is simple: discuss real growth challenges, share experiments, and learn from what actually happens—not just what looks good in a marketing playbook. We have 10+ members so far. In our first session, we talked about founder-led content, outreach, niche events, partnerships and how to choose the right channels for a specific buyer. I’m building Nexshift.ai, so I’ll also share what we’re learning about lead engagement and customer conversations. This is a space for everyone to contribute, ask questions and challenge ideas. If you’re actively working on B2B growth and would like to join, comment “Growth” or send me a DM. I’ll share the group link.
Adam Robinson hit $32M ARR across 3 companies using JUST LinkedIn I spent 3 weeks reverse engineering exactly how and built a Claude Skill Folder, get free access below built on the exact playbook Adam runs to turn one personal brand into three separate revenue streams. here is everything inside: 1/ Content engine - 8 skills - covers hook architecture, daily posting cadence, topic rotation across three audiences, pattern-interrupt openers, and scroll-stopping first-line frameworks Adam has posted every single weekday for 3+ years and each post cross-feeds pipeline into all three companies at once 2/ Revenue transparency system - 7 skills - covers public mrr milestone posts, honest loss breakdowns, founder narrative arcs, audience trust sequencing, and vulnerability-to-credibility conversion frameworks Adam's "$1M ARR in 90 days" rb2b post alone generated more inbound than their entire outbound motion that quarter 3/ Audience growth playbook - 8 skills - covers follower compounding mechanics, viral post anatomy, comment section amplification, repost triggers, and algorithm-aware scheduling for maximum organic reach Adam grew from 12K to 180K linkedin followers while simultaneously scaling three separate companies to combined $32M ARR 4/ Multi-company launch system - 7 skills - covers pre-launch content sequencing, waitlist drop strategy, founding member framing, cross-brand audience migration, and launch day post structure Adam pre-sold moltsets to 600+ waitlist members using only linkedin posts before a single product feature shipped publicly 5/ Inbound pipeline conversion - 8 skills - covers comment-to-DM qualification routing, CTA placement within posts, buyer intent signal reading, follow-up sequencing, and demo booking frameworks from organic traffic Adam's comment sections across all three companies convert inbound interest at 3-5x the rate of any cold outbound channel he's ever run 38 claude skills, adam robinson's entire linkedin playbook
Most AI procurement case studies out there are just marketing hype. 2 years ago, I started looking for verifiable proof… real companies, named tools, and clear financial ROI. Here is what I found across 50 real world deployments: 1️⃣ Nestlé started with a single, highly focused use case (contract analysis) rather than a massive company wide rollout. 2️⃣ Walmart launched a small pilot in Canada first before scaling AI negotiations to 75% of their suppliers. The pattern is clear: Real transformation happens when you solve one specific pain point first not when you have the biggest budget. I compiled all 50 verified case studies (including company names, tools, timelines, and measurable outcomes) into a complete PDF breakdown. If you're working on integrating AI into your procurement workflows, tell me what specific process you're currently trying to optimize in the comments and I’ll gladly share a copy with you 😊 📍Want more practical AI procurement breakdowns? Follow Asmaa Gad and Supply Chain AI Pro for AI procurement playbooks and case studies.
I looked at the top 100 Shopify Subscription and Buyboxes so you don't have to. It's all in one free Google Doc, with a Figma file to match. If you’re an eCommerce manager of 7-9 Figure DTC Brand, looking to optimize your offer or buybox - then this is for you. Here's what you get: ↳ 100 real buybox screenshots from subscription and consumable brands, one per brand. ↳ A link to every live page so you can click through the actual offer. ↳ Sorted into four offer types: 13 starter kits and gifts, 25 bundles and quantity offers, 41 subscription vs one-time setups, and 21 purchase choice layouts. ↳ Notes on every buybox: how the offer is built and what to compare against yours. ↳ Grüns, Nutrafol, OLIPOP, LMNT, Ritual, ARMRA, Primal Queen and 93 more. ↳ The Figma file with all 100 laid out side by side so you can zoom in and compare. Use it to pick what to test next. Full guide in the first comment.
AI for self-improving agentic outbound 7 loops that turn buying signals into qualified opportunities 👇 1/ Start with intent, not a giant database Instead of scraping thousands of prospects and hoping something sticks, the system starts with signals that indicate someone might actually be in-market. 2/ Track what prospects actually do Hiring, fundraising, competitor engagement, LinkedIn activity, new posts, comments, follows, keywords… Small signals become much more powerful when combined. 3/ Find the people behind the signal Once intent is detected, the system identifies the right decision-makers around it. No guessing. No random lists. 4/ Understand why they're relevant Every lead gets context around the signal that triggered it. Who they are. What happened. Why it matters. What changed. 5/ Turn intent into a message Instead of sending the same pitch to everyone, the signal becomes the reason for the outreach. The message starts from something that actually happened. 6/ Let the agent prioritise Not every lead deserves the same attention. Strong signals get prioritised. Weak signals get filtered. New signals continuously enter the pipeline. 7/ Close the loop Replies, meetings and outcomes become feedback. The system learns which signals actually correlate with opportunities and which ones create noise. The biggest shift isn't sending more outbound. It's knowing who is worth contacting before you contact them.
I put 5 AI models into one LinkedIn Ecosystem. Claude. ChatGPT. DeepSeek. Mistral. Qwen. Built for agencies and creators that already writing with all five, but losing hours somewhere between the draft and the publish button. So I put every one of them behind one persona and one publish button you can run inside Sawify. Here's what's inside: 1. Persona - The Voice Doc ↳ Up to 40,000 characters on how you write and who you're writing for, and every model writes from it 2. Search and Deep Research - The Fact Check ↳ Searches the web with sources shown, or goes deeper and hands you a full report you can download 3. Strategy - The Content Plan ↳ Tell it how far ahead you want to go, 7, 14, 21, or 30 days, and it drafts a full post for every day 4. Planner - The Calendar ↳ A board and a calendar you drag posts around, and the chat can file a post there with the date already on it 5. Scheduling - The Publish Button ↳ Post now, or set a time and it goes out even if your laptop is closed, with a retry if LinkedIn hiccups 6. Comments - The Five Options ↳ Five drafts every time, with a slider for short, medium, or long 7. Analytics - The Numbers ↳ Upload your LinkedIn export and see impressions and followers against the previous period Here's how it works: → Set up a persona (paste in how you actually write, not how you'd like to) → Pick a model or leave it on Auto → Ask for a plan as long as you want and tick the hooks you'd say out loud → Read every post and cut whatever sounds like a model before you hit schedule This is exclusive But I'm giving access to the Ecosystem. For FREE. PS - Repost this to get it first.
Everyone optimises the account but, not many optimise how they run it. So here's 15 Claude skills that help you do exactly that. You have rules for the account. Naming conventions, testing cadence, budget caps. But the week around may run on instinct. So these fifteen give it the same treatment. Here's six of the skills: → /number-check, computes five numbers and returns one verdict, so you know which one is weak → /where-money-went, traces spend down to ad level, so the share that produced nothing has a number → /spend-ceiling, reads your own history at each spend level, so scaling stops being a guess → /target-reset, tests the target against unit economics, so a bad one arrives with its replacement → /decision-log, records every change with its reason, so "why did we do that" has an answer → /onboard-audit, runs the checks that invalidate everything else first, so week one surfaces the deal breaker The other 9 cover forecasting, priorities, handoffs, pitch numbers, your tool stack and the hours that are being wasted. Each one is a slash command to work the entire skill. I'm giving away all 15 operator skills! Want access? 1. Make sure we're connected 2. Comment "OPERATOR" below I'll send you the link.
The Claude for CFOs Ultimate Bundle (everything you need to master Claude in Finance) 👉 Comment "Claude" and I'll send you the full bundle I'm limiting this to the first 100 comments, so move fast Every time, I hear the same thing from CFOs and controllers: "Nicolas, I know Claude is powerful, but I don't know where to start" So I put together the exact bundle I wish every finance team had since day one: 📘 Claude Finance Playbook How to build financial models in Excel, use Voice Mode from your phone, and create interactive scenario planners 📋 Top 100 Claude Tips 100 prompts for Excel, Cowork, dashboards, and best practices you can use today ⚡ Top 100 Claude Shortcuts The fastest way from prompt to finished output, no wasted steps ✅ Claude Checklist So you never skip a step when building a model or a report 📖 Claude 101 Dictionary Every Claude term explained in plain finance language 🎓 Claude AI Finance Course A structured path to go from zero to confident, at your own pace 🔑 The Mega Claude Collection Collection of Infographics and carousels about Claude On top of that, this bundle covers the whole journey: understand it, use it, master it, teach it to your team 👉 Comment "Claude" and I'll send you all the resources Bonus: you'll also get a free seat to my AI Finance Masterclass 🔁 Repost so your finance team stops guessing and starts using Claude the right way
If you're out of the loop on Jev, here's the simplest workflows for marketing. Seven jobs that are usually manually now only cost 29 cents. Jev is a new AI that only makes decisions. It can't write a sentence. But every one of those seven jobs is the same decision repeated thousands of times. So you write the questions once and they run over every row you have: 1/ Sorts your search terms Code pulls the full report, all 12,000 rows. Jev labels each one buyer, researcher, competitor or junk, and the negatives file writes itself. 2/ Grades every ad Code joins last week to this week across Meta and TikTok. Jev returns scale, hold, cut or refresh on each one, and says how sure it is. 3/ Catches fatigue early Code finds the ads whose click rate fell while frequency rose. Jev says whether the creative is tired or the offer is dead. Opposite fixes. 4/ Scores every lead Code sends each form fill with your ideal customer next to it. Jev rates the fit, and the score goes back to Google and Meta so they learn. 5/ Checks the ad against its landing page Code pairs every ad with the page it points at. Jev scores whether the page keeps the promise, and names where it breaks. 6/ Scores briefs before you shoot Code sends the hooks with your brand rules. Jev rates each one, so only the ten worth making get made. 7/ Triages every comment Code pulls the comments under your ads. Jev sorts the buyers from the trolls, so the buying questions get answered first. Every run ends in a file you open. Nothing touches your account. I'm giving away the seven question sets and the runner.
here's our entire off-market deal sourcing system, built for Claude/ChatGPT every owner in a market swept, scored, and named before a single deal hits the market not locked to one asset class: multifamily, retail, industrial, land. same system, you just swap the filters point it at a market and it runs the off-market desk end to end: - Deal Sourcer: sweeps every off-market owner in the market and scores each one against your buy box. kills the misses. - Sell-Signal Scorer: flags the owners about to sell, on long hold, loan maturity, tax delinquency, absentee ownership. - Owner Resolver: names the real person behind the LLC, from lawful public records. - Underwriter: income, debt, 5-year return, and a deal score out of 10 no broker blast, no stale list, no monday spreadsheet. you get to owners before they list, while everyone else fights over the 3% that's on the market comment 'SOURCED' and i'll send you the system (make sure we're connected so i can dm you). PS: Repost this for priority access. Linkedin limits how many I can send
If you are running outbound with Grok bot, Claude code, or some advanced workflow and expect the tool/workflow will help you book more meetings you are wrong. No one cares about your super unique first line personalization especially if its not relevant to the prospect/offer. Your offer is what actually books you meetings, not the advanced workflow that mentions something the prospect did 15 years ago in college 🤣. I will send you over the exact template I would use based off your offer to book more qualified meetings with outbound. Just comment: "offer"
I dare you to try this and you’ll be surprised. A Hook is like ginger in chai. Han bina adarak ke bhi chai bna sakte ho aap. But add the right amount of ginger, and suddenly the whole taste changes. Your post works the same way. You can have: → Great knowledge → Valuable information → A strong story But if your Hook is weak, people may never read the rest. A good Hook makes people stop scrolling, feel curious, and read the next line. So before posting, ask yourself: “Would I stop scrolling for this?” If the answer is NO, your Hook needs work. Day 6 of my 30 Days LinkedIn Learning Series. Follow me for Day 7, where I'll show you how I write my posts from Idea to Upload. And if you want the PDF of my personal LinkedIn hook, comment “PDF” or DM me. I'll send it to you 🔥
Canva and Figma should be worried. Here's why. Anthropic launched Claude Design. I've been testing it. And I need to talk about it. This is not another AI image generator. This is not another template tool. This is a full design workspace Where you describe what you want And it builds it. Landing pages. Pitch decks. Animated videos. Email layouts. Campaign visuals. All on brand. All in minutes. Here's what blew my mind: ✅ You upload your brand guide ONCE and every project inherits it ✅ Landing pages in 10 minutes — responsive, mobile-first ✅ Pitch decks with speaker notes ✅ You can iterate with chat, inline comments, or live sliders ✅ Export to PDF, PPTX, HTML, or send straight to Canva for polish The breakthrough: You don't need to be a designer. You don't need templates. You just need to describe what you want. And the output doesn't look like AI slop. It looks like your brand. I've boiled the whole thing down into the infographic below 👇 And if you want to go deeper, I've built a full step-by-step Notion board - the setup, the prompting, and 5 ready-to-use marketing workflows with copy-paste prompts, all laid out so you can work straight through it. Just let me know in the comments and I'll send you the link. 💡 PRO TIP: Save this post so the infographic is there whenever you need it.
Last November, the brands we manage grew ad spend 186% YoY. That spend generated $93M+ in revenue across the month, with Black Friday alone delivering $8.2M at a 3.96x ROAS. We've now been through 7 BFCMs as an agency, with £170M+ in annual ad spend and £600M+ in total revenue generated, across brands like Mindful Chef | B Corp, Free Soul, Heights, TRIP, Spacegoods, and Smol. This year, we packaged everything we've learned from our last 7 BFCMs into one resource: The 2026 BFCM Playbook. Meta, TikTok, Creative, and full-funnel strategy, all in one place. Here's what you’ll find inside: 1. 2026 BFCM trends and predictions built from $93M in revenue data, plus the Growth Compass framework for diagnosing where your brand sits before peak hits. 2. Andromeda and what it means for BFCM targeting, pre-peak persona research, lead-gen and demand capture before the sale goes live, campaign structures that scale, landing page essentials, and partnership ads with creator whitelisting. 3. Why you should use TikTok Shop for BFCM, foundations and setup, offer mechanics and margin protection, building the affiliate engine, LIVE shopping strategy, and how organic, paid content, and paid amplification work together 4. How to build a BFCM creative toolkit, formats that convert during peak, creative forecasting, hook fundamentals, angles that drive incremental reach, and lessons on creators, partnerships, and UGC. 5. Full teardowns of Kind Patches and Bloom Nutrition, showing exactly how the strategy played out in real accounts. This isn't a generic "get ready for Q4" checklist. It's exactly what worked from a team that's lived through it 7 times. Want a free copy? Comment "BFCM" and I'll send it over via DM. (make sure we’re connected so I can send)
JEV ranked 500 leads in 10 seconds!!!🤯 This means you can finally focus on prospects who are actually going to buy Jev has taken the AI world by storm, and along with our tool, it 1/ Scores every lead against your ICP We pull in the leads. JEV judges each one on title, company stage and buying trigger, then ranks who to contact first. 2/ Reads the buying signal behind each lead We the post, comment or competitor engagement. JEV decides whether it shows real intent or was just a passing like. 3/ Matches each prospect to the best message You write a few versions with different hooks, offers and CTAs. JEV picks the one most likely to land with each lead. 4/ Predicts how each message will perform JEV gives every message a confidence score before it goes out. You catch lead-message mismatches before anything sends. 5/ Sorts every reply in your inbox JEV tags each reply as interested, not now, or wrong person, with a confidence score. Low-confidence replies go to a human before anyone sends a Calendly link. 6/ Finds which intent signals booked your demos JEV goes through thousands of past outreach messages at once. You see which signals led to meetings and which ones wasted your sends. 7/ Picks which campaigns to scale JEV ranks your campaigns by how likely they are to perform, based on real data. Claude then writes new sequences only for the ones that passed. 500 leads. 10 seconds. $0.09. Qualifying them one by one with Claude used to cost us minutes per lead. Not anymore though. Top sales teams are moving from “find leads” to “find reasons to contact them.
I built a complete LinkedIn Claude Content System. And I'm giving the entire system away for free 👇 I built a collection of specialist Claude tools to support every stage of writing, posting and converting on LinkedIn, from finding what to say to turning comments into booked calls. It covers: → The voice profiler, so your posts sound like you and not a template → The daily trend scanner pulling from Reddit and X, so you're never writing into a dead topic → The content matcher that maps your ideas to 12 proven post formats, with image direction included → The post writer that drafts full posts, lead magnets and case studies in one pass → The performance tracker that logs what books calls and adapts any post in one click, so you stop guessing what to write next → The lead magnet builder that creates the actual resource, not just the post about it → The scheduling engine that tells you exactly when to post each piece → The comment-to-DM converter that sorts everyone who engaged into buyers vs. browsers and writes the first message for each, so warm leads don't go cold in your notifications → The hook tester that writes 10 openers for every post and ranks them by what stops your buyers mid-scroll, so your best ideas don't die in the first line → The buyer language miner that pulls exact pain points and phrases from your sales calls, DMs and client notes, so every post speaks the way your buyers already talk → The repurposing engine that turns one winning post into a carousel, a newsletter and a week of follow-up angles, so your best ideas keep booking calls instead of dying after 48 hours Inside, you'll find 11 specialist tools, each built to solve a specific problem across the entire content and pipeline process. The goal isn't to post more content. It's to build a LinkedIn that consistently books calls with the right buyers, without you staring at a blank doc every morning. It's the same system I used to book 25 meetings last month. Whether you're a founder,
The Ultimate Fundraising Guide to Raise Millions🔥 I've invested in tens of startups that have raised over $100M together. After years of writing about fundraising on our newsletter, everything useful ended up spread across dozens of editions. So I compilded the full guide and collected every resource behind it in one place. Inside you'll find: * Investor Update * +90 Pitch Decks * Cap table Template * The Financial Model * Data Room Template * The Family Office List * Safe Dilution Template * The Angel Investor List * The Al Unicorn Angel List * The LP Investor Database * Valuation Model Template * The Startup Incubator List * The Runway and Burn Template * The Corporate Venture Capital Fund List * The 300 Most Active Seed VCs Investors * The KPI and Metrics Dashboard Calculator Comment RAISE and I'll send you the link!
I spent 6 months prompting Claude for LinkedIn. RESULT: 13 views / post. Then everything changed 👇🏾: ( THIS IS THE SYSTEM – too many comments, no need to comment anymore: hope it helps: https://lnkd.in/gyEfYqXF ) I rebuilt it as a specialist team — and it changed the game entirely. Here's what I learned: Claude becomes significantly more powerful when you stop treating it like a chatbot and start running it as a department. One agent for voice cloning. One for hooks. One for research. One for editing. One for scheduling. Each agent has a specific job. Specific context. Specific skill set. No more "write me a LinkedIn post" into a blank chat window. Here's what's inside the system: → 8 Specialized LinkedIn Content Agents → 17 LinkedIn Content Skills → 17,836+ Viral Hook + Content Pattern Skills → 41+ Infographic Styles → 40 Viral Creators Tracked Weekly Plus: → 3,000-Word AI Clone of your voice → Full Scheduling System via Official LinkedIn API → Complete setup with just 2 links This is the exact system I use daily. Built over 1+ years of testing. Get it here. 👇🏾
2 years ago I was all-in on Clay. Then n8n became my main tool. Now I run most of my outbound from Claude Code. Clay still has the best UI in GTM. If a client already uses it, I’m happy to work in it. But for most of my clients (and for my own) outbound, the economics stopped making sense. I stopped recommending it to founders. Some even ask me to leave Clay out of the proposal lol. So this is the stack now: Research - Granola: every call recorded - Attio: CRM Build - Claude Code: builds and runs everything - Conductor: to run sessions with Claude Code (I'm very happy with Opus 5.5) - Trigger.dev: runs the pipelines - Neon Postgres: my own DB, 20M+ accounts Data - Deepline + Crustdata (YC F24) - Treg.to + Icypeas: mostly for people search & emails - Context.dev (YC S26) Score / Qualify - Jev by TypeSafe AI: qualifies every company and tiers every reply Outreach - Salesforge.ai 🔥 or Instantly.ai: email - Salesforge.ai 🔥 or HeyReach: LinkedIn I mapped all workflows into 6 diagrams, showing when each tool runs and what each step costs. 1. Send me a connection request 2. Comment "DOCK" or send me a DM ps. this animation video was fully generated by opus 5.5
These 3 Claude Skills Run My LinkedIn (It booked me 10 calls this week) I ran a test, because I saw dozens of posts talking about: "These 32 AI Agent closes me $15K deals in my sleep" So I used my proven frameworks, systems & 100+ post templates. And built a LinkedIn Operator Agent, broken into 3 claude skills: → Content Ghostwriter: Writes viral posts in my voice & context → Outbound SDR: Personalises DMs inbounding to my inbox → Graphic Designer: Designs my banner, featured & about sections This outperforms the typical LinkedIn ghostwriter or Claude "Prompt". All because of the context, most agents don't have the correct context. The Setup: 1. Download the repo (comment "Skills" for it) 2. Upload the folder into Claude 3. Have the Agent onboard you 4. Start drafting viral content + DMs It runs a context profile generator which builds a JSON file of your story, business case studies & voice from the Agent's onboarding features. Then it uses this to draft viral posts, successful DMs, all within minutes. If you want to recieve the video breakdown on how to install this for yourself, comment "Skills" & I will send it over. (Must be connected first)
R.I.P LinkedIn Ghostwriters GPT-6 Sol just got released and transformed the LinkedIn content game. I built the GPT-6 Sol LinkedIn Content System. One system that plans, writes and designs your entire week of content. It analyzes your ideal client → Finds the angles they respond to → Pulls the LinkedIn profiles your content should reference → Analyzes their top-performing posts → Analyzes your top-performing posts → Identifies exactly what your underperforming ones are missing → Generates a full week of content personalized to your offer and ICP → Writes the posts → Creates the images to go with them. The system runs through 6 stages: 1. Audience & input It takes your offer, your ICP and your brand voice, then maps what your ideal client actually responds to before a single word gets written. 2. Competitive analysis The system pulls the LinkedIn profiles you want to reference, reads their top-performing posts and extracts the patterns driving their reach. 3. Content diagnosis It runs your own posts through the same lens, flags what the top performers share and tells you precisely what the underperforming ones are missing. 4. Production & output The system generates a full personalized week of posts tied to your offer and ICP, then produces the images for each one. 5. Lead magnet creation It builds a lead magnet around the exact problem your ICP is stuck on, then writes the post that promotes it and the image for it, so every download turns into a warm conversation in your inbox. 6. Performance loop The system tracks which posts drove profile views, comments and inbound DMs, then feeds those results back into next week's content so every batch gets sharper at pulling in leads. Instead of paying a ghostwriter for every post, you get a system that learns what works and does it again every week. I'm giving away the complete playbook. Free Access 👇
I ran 25,033 of my LinkedIn connections through an ICP check. It took 71 seconds and cost less than a dollar. 435 came back warm. Those are decision makers at the kind of company I sell to, and they were in my network the whole time. I'm testing JEV by TypeSafe AI It answers one question per connection: does this person fit my ICP? It only sees job title, company, headline and whether someone engaged with my posts, never their name. My criteria: owner, CEO, MD, COO or head of operations. Company in or near the Netherlands, likely 50+ employees. Engaging with my posts is a plus. The result: 435 warm 1,982 maybe warm 5,856 cold 16,756 not relevant Two thirds of my network isn't my buyer. Good to know, and a bit painful. By hand, at 10 seconds a profile, that's almost 70 hours. So I never did it. The first run found 2 warm people out of 25,000. That was on me. I had made engaging with my posts a requirement, and only 205 of my connections had done that in the last 90 days. I fixed the rules and ran it again. Want this ICP checker? Let me know in the comments and I'll send it.
I'm now sharing 84 real Statement of Purpose (SOP) samples that helped students secure admissions and scholarships in: UK, USA, Germany, Switzerland, and other top destinations. These are not generic templates, these are actual successtul SOP structures you can learn from and use to build your own. If you're planning to apply for: • Scholarships • Universities abroad • Graduate programs This will help you significantly. If you want the SOP samples, comment your email below - I'II send them to you.
Most people assume social proof always works, but our data says otherwise. We've run 4,000+ A/B tests across 250 European DTC brands. 3 out of 4 social proof tests in our last study failed or went flat. You're paying to bring people to your store, and somewhere between landing and checkout, you're losing most of them. It happens because the way your store communicates isn't matching how your customers' brains actually make decisions. Users didn't respond to "others bought this." They responded to "I want to look like that." I documented 4 psychological triggers with 18 real tests behind them, including which ones backfired and why. What's inside: - Real ARPU uplifts per test (up to +23.74%) - The pattern behind each trigger cluster - Why the same principle can win or lose depending on context 𝗜𝗳 𝘆𝗼𝘂'𝗿𝗲 𝗿𝘂𝗻𝗻𝗶𝗻𝗴 𝗮 𝟳-, 𝟴-, 𝗼𝗿 𝟵-𝗳𝗶𝗴𝘂𝗿𝗲 𝗘𝗰𝗼𝗺 𝗯𝗿𝗮𝗻𝗱 𝗮𝗻𝗱 𝘄𝗮𝗻𝘁 𝗶𝘁: Comment PSY
I put together a complete guide to using ChatGPT to launch A/B tests on Shopify. If you're an eCommerce manager looking to spin up a few tests yourself, this free guide is for you. It uses the ChatGPT Codex desktop app, Figma and Convert MCP. Here's what it walks you through: ↳ Import your current page into Figma and keep it as the Control. ↳ Build a clickable variant prototype. ↳ Save a dev brief and hand it to a new development thread. ↳ Connect Convert MCP, build the test and save it as a draft. ↳ Check it on your phone, desktop and tablet, then fix any bugs. ↳ Launch from the same development thread once the checks pass. Full guide in first comment.
Agencies charge anywhere between $3,000 - $15,000 for a brand style guide. Here's how you can create your very own professional-grade doc in Claude Design. Here's what it includes: Brand Story & Mission Core Values Voice & Tone Logo & Lockups Clear Space & Minimum Size Logo Misuse — Never Do This Color Palette & Ratios Typography & Scale Layout & Grid UI & Component Styling Imagery & Screenshots Surface Specs — Social, Deck, Docs Signature, File Naming & Brief Block Drop this PDF into Canva's Brand Kit. Hand the table to your designer. Paste it into your Claude project and every asset you make from here pulls from the same brand, without thinking about it. Then test it: "Write a LinkedIn post in this brand voice." "Create an Instagram reel based on my brand style guide on topic [x]" That's how you know the guide works. Make every piece of content you make follow the same rules. If you want the full step by step and the prompts, follow me and comment "Prompt" and I'll send it to you.
take my claude + linkedin system that closed 5 deals in a week. 7 agents, 1 MCP, 5000 free credits. founders running linkedin outbound right now still look like this: → export a list from sales nav → paste the same "hey {first_name}" opener for the hundredth time → get a 12% acceptance rate and call it a linkedin problem → give up before the system even had a chance and then wonder why AI isn't doing anything useful 🫠 if that's you - here's what the actual system does, step by step: 1/ finds your ICP where they're actually active. competitor pages, communities, events - not just job titles in a filter. then enriches each profile with 10+ data points before you spend a single minute on them. 2/ fixes your profile before it sends a single message. your banner, headline, about section - rebuilt around what your ICP needs to see to say yes. the profile is the landing page. if it doesn't convert, nothing else will. 3/ writes connection requests that sound like you. not a name-drop opener. an actual reason tied to what this specific person posted, commented on, or shared. one client hit a 78% connection rate with this alone. 4/ builds two types of posts every week. a lead magnet post that puts free value in front of your ICP. an authority post that makes you the go-to in your niche. both written in your voice, both calibrated to do a job. 5/ tracks every warm reply before it goes cold. drafts the response, handles the hidden objection, flags the exact moment a conversation is ready enough to ask for the call. 6/ pulls a clean report at the end of the week. what worked, who to follow up with, where to double down instead of guessing. this is the workflow you're getting: Research → Fix → Connect → Content → Follow up → Report everything's inside 👇
I built an entire LinkedIn sales department inside Claude. And you can get the whole thing for FREE 👇 It includes: → 60 Sales Agents → 50 Sales Skills → 100 Sales Prompts → 30 Sales Workflows → 10 Sales Playbooks Covering the entire LinkedIn sales process: → ICP & buyer research → Positioning & profile → Content & post writing → Lead magnets & resources → Keywords & comment capture → Prospecting & buying signals → DM outreach & personalisation → Follow-ups & nurture → Lead scoring & qualification → Discovery & call prep → Objection handling → Proposals & closing → Pipeline & CRM → Analytics & forecasting → Scaling across multiple accounts Each resource is built to handle a different part of the sales process. The goal is simple: Reduce repetitive sales work, improve your targeting and preparation, and give you a complete system for moving from post → conversation → closed deal. Whether you're a founder, SDR, AE, consultant or agency owner, there should be something useful inside. I'm giving away The Ultimate Claude LinkedIn Sales Department. Free access 👇
Claude helped me generate 16 LinkedIn posts, pull 10 inbound leads, and book 3 demos before I woke up. Get it here, free. 👇 ( THIS IS THE SYSTEM – too many comments, no need to comment anymore: hope it helps: https://lnkd.in/gttpMtZB ) I was asleep. It was working. Here's what I gave it before I slept: • 1 LinkedIn profile • 1 website url • 1 ICP • 5 lines explaining who I help That's it. No prompts. No copy-paste. No 2am editing. Here's what it did overnight: 1. Researched what my ICP was actually talking about 2. Found trending angles from Reddit, X, and news 3. Wrote posts in my voice using my 3,000-word AI clone 4. Generated hooks from 17,836+ viral patterns 5. Built infographics from 41+ styles 6. Scheduled the full week via LinkedIn's official API 7. Tracked 40 viral creators and flagged what's spiking I woke up to a full content week and a booked calendar. The full LinkedIn content system: ► The voice profiling system that builds your 3,000-word AI clone ► AI workflows for trend research, hook generation, and writing ► The exact agents that write in your voice, not generic AI slop ► The infographics engine with 41+ styles built in ► The analytics dashboard reading your LinkedIn via official API ► The scheduling system that posts a full month on autopilot Get it here, free. 👇
LinkedIn agencies just got AUTOMATED with official apis. Fable replaced them all: Get it, here. 👇🏾 ( THIS IS THE SYSTEM – too many comments, no need to comment anymore: hope it helps: https://lnkd.in/gttpMtZB ) Before this, you had 2 terrible options: → Spend 2 hours daily writing posts manually → Or automate everything with ChatGPT And AI for content? It was terrible. Yes, it WAS. Because with Claude's latest models, you can now combine: ☑ Content at scale ☑ Deep personalization But here's where it gets CRAZY: Claude runs the entire content engine for you. → It fetches your last 100 posts and builds a 3,000-word voice profile in 60 seconds → It scans Reddit and X daily for what your ICP actually wants to read → It generates 10 scroll-stopping hooks per post trained on 100M+ viral impressions → It creates viral infographics designed for maximum engagement → It writes full posts that sound exactly like you with 3 QA rounds - zero AI slop → It extracts your hidden stories through 1-1 podcast-style interviews → It schedules + gets analytics of 1+ year via LinkedIn's official API → It learns what performs and updates your preferences after every post All in your exact tone and voice. All on its OWN. Like a top 1% agency working 24/7 for a fraction of the cost. Get free link. 👇🏾
Want the next batch of proven lead magnets?
You reached the free Discover cap: 50 proven items. Drop your email and we will send the next batch when the extended library opens.