Honored for the shoutout from the amazing Alley Pezanoski-Browne š£ļøš. We contain multitudes and our careers can, too!
If you're someone (like us!) whose interests donāt fit neatly into a single category, join us on May 7th. Instead of forcing everything into a single lane, this course explores how your diverse interests might actually be the foundation of a thoughtful and flexible career.
Learn more and register here ā”ļø https://lnkd.in/g_eRUq4D
š Ever feel like your brain is a browser with 100 tabs open?
You start writing a project outlineā¦ then get pulled into checking your messages, remembering a grocery list, or planning something else.
Every unfinished thing nags at you. Your mind loops, replays, and distracts ā all at once.
But what if I told you thereās a hidden mental lever your ADHD brain can use ā that very loopiness can become motivation?
Enter the Zeigarnik Effect.
ā¢ ā¢
Hi team. A lot of my job is talking to organisations about AI and the people using it: through informal conversations, cross-industry speaking events or targeted workshops designed to help leaders understand the future of work in the Age of AI. And now, having engaged with hundreds of individuals over the past 7 months, trends are starting to emerge.
ā¢ ā¢
~ Save this if your stories get likes but donāt bring clients. šā£
ā£
Most people are already storytelling in their content. āØā£
ā£
The problem? ā£
ā£
They skip the structure that actually makes people trust, relate, and buy. ā£
ā£
A strong story isnāt random. It follows a clear path:ā£
ā£
1ļøā£ The Relatable Moment: a specific moment your audience recognizesā£
ā£
2ļøā£ The Internal Conflict: the doubts, thoughts, and struggles they rarely say out loudā£
ā£
3ļøā£ The Shift: the turning point where a new perspective appearsā£
ā£
4ļøā£ The New Perspective: what changed after that shiftā£
ā£
When your audience seesā¦
~ SAVE THIS guideeee šā£
ā£
Because in 2026, personal brands arenāt built by posting more... ā£
ā£
theyāre built by being clearer, more human, and more intentional. āØā£
ā£
š¤ This carousel breaks down exactly how to build a magnetic Instagram presence step by step:ā£
ā£
ā¢ā¢ from choosing the right niche, ā£
ā¢ā¢ to creating content people feel, ā£
ā¢ā¢ to turning your life into your strongest brand asset. ā£
ā£
I used a Financial Coach for Digital Nomads as the example, but this framework works for any niche - coaches, creators, freelancers, or service providers.
AI adoption doesn't stick at the org level. It sticks at the team level.
That's not a knock on company-wide training or leadership initiatives. Those matter. But the real behavior change happens when someone on your team shares what actually worked for them last Tuesday. When a peer says "I tried this and it saved me two hours" or "I tried this and it totally flopped, here's why."
That's the conversation that moves people.
Org-wide rollouts can build awareness. They can establish a baseline.
NotebookLM is not just a learning tool anymore.
Most people use it to summarize PDFs. Generate audio overviews. Maybe create a few flashcards.
Thatās 1% of what it can do.
The real power? NotebookLM connects to Gemini. Gemini connects to Claude Code. Together, they form a pipeline that turns any YouTube video, PDF, or document into a working application.
I built a live app using this workflow. No coding background. Just prompts.
ā¢ ā¢
TLDR:
Every time you open a new AI conversation, your context resets. Built-in memory features help a little, but they donāt know about your active projects, your team, or what happened yesterday. The solution is a local folder of plain text files that any AI can read, called a vault. I built one using Obsidian, and it changed how AI actually works for me on a daily basis. This article explains the problem and the concept. Part 2 is the full step-by-step setup (stay tuned!).
The goal: One folder. One vault. One context to rule them all. Ok, enough of LOTR references already.
Charlie's entire Claude skills library is now free.
17 skills you can clone in 60 seconds:
This is the exact system running Charlie's content across LinkedIn, Instagram, Substack, X and YouTube.
We test every skill on our own accounts before it ships.
Start here first ā https://lnkd.in/d3e-pEJy
FOUNDATION
ā¦ voice-builder ā https://lnkd.in/dkzsPMUC
ā¦ newsletter-voice ā https://lnkd.in/dGR6D9Hf
ā¦ profile-optimizer ā https://lnkd.in/dWP-C57B
ā¦ post-writer ā https://lnkd.in/dPZmPtEQ
ā¦ post-scorer ā https://lnkd.in/dEcAPa5g
ā¦ post-formatter ā https://lnkd.in/db3qbbMh
ā¦ā¦
BREAKING: OpenAI just dropped GPT Image 2.
It's the first AI image model
that doesn't look AI-generated.
Nano Banana 2 was my daily driver for two months.
Not anymore.
Every image model until now had tells.
ā Coloring that screamed "AI"
ā Typography that fell apart at small sizes
ā Icons that looked like clipart
ā A telltale plasticky finish on every face
You could spot a Nano Banana image from a mile away.
GPT Image 2 looks like a professional made it.
The benchmark numbers back it up.
Every Claude Code tutorial assumes you can code.
This one assumes you canāt.
Iām the CEO of a small business, I canāt write code, and Iāve been doing knowledge work for 21 years (that means I push buttons on a keyboard for a living).
In the past if I needed a presentation or had to do complex data analysis in a spreadsheet, it took half a day. I used to take meeting notes by hand because I thought automated note-takers were damn near worthless.
Contract reviews, business agreements, charts, graphs... all manual, all the time.
Four months ago I installed Claude Code.
Every Friday afternoon, this scheduled task runs in Claude Cowork while I'm still in meetings or maybe I'm walking my cute dog, Marlow :-) By the time I close my laptop, I have a digest of every loose end across active clients. Do yourself a favor and start using scheduled tasks to make your life easier.
Here's the actual prompt. Make it yours:
Run my weekly client status digest for [list your active clients or projects].
Step 1 ā Pull context from this week:
Check Google Calendar for any client meetings or calls from Monday through today. Note who attended and what the meeting was for.
The most referenced AI repo on GitHub right now?
The http://DAIR.AI Prompt Engineering Guide.
Tens of thousands of stars.
Used by founders, PMs, and LLM teams globally. Every day.
And the conclusion everyone keeps reaching?
The output gap between AI users isn't the tool.
It's the prompt.
Here's everything inside it:
ā Zero-shot and few-shot prompting techniques
ā Chain-of-thought reasoning
ā Tree of Thoughts framework
ā RAG (Retrieval Augmented Generation)
ā ReAct prompting for AI agents
ā Prompt chaining for complex workflows
ā How to prompt ChatGPT, Claude, Gemini, GPT-4
ā Risks,ā¦
You have 27 saved prompts in a folder you havenāt opened in 3 weeks.
You know the folder. Maybe itās a Notion page (like mine). A Google Doc. Or worse, 27 screenshots on your phone you swore youād organize āthis weekend.ā
You collected them from LinkedIn posts. From YouTube videos. From that one thread on X that promised itās āthe only prompt youāll ever need.ā
And then you sat in front of ChatGPT, pasted one in, tweaked it for 2 minutes, hit enter, and got... meh. Something that sounded like every AI text on the internet. Sounds good, does not work. Thatās the revolution you were promised?
ā¢ ā¢
Starting a one-person business model can unlock financial freedom and change your entire life.
Seriously.
This is exactly what I did to gain financial independence in my twenties.
The best part?
These are business models that donāt rely on the mercy of an employer to get paid. Your income potential isnāt capped by arbitrary ceilings. And when done right, all of the profits funnel straight to your bank account.
Tested ChatGPT Images 2.0 this weekend. Here's what I learned.
First attempts: boring. Generic AI illustrations, nothing close to the examples on OpenAI's website.
Then I rewrote (well, Claude rewrote) the prompts with proper detail.
Specific composition.
Specific lighting.
Specific layout.
Specific text rendered verbatim.
The results changed.
So, the model is good, but I wouldn't say perfect. Character consistency is still not there. But text creation is indeed really good.
I caught up with an agency friend last week and I hadn't talked to her in a bit. She caught me up on life stuff, what she's been working on at work, where she's running into some roadblocks, etc. etc. etc. A real delight, that one.
While we were talking, she told me that she's feeling a lot of pressure to get more efficient. Agencies are a tough business and making sure you're making the most of billable hours is really important!
She said she knows that she can do more with her AI platforms, but doesn't really have the time to invest in figuring it out.
ChatGPT launched its new built-in image generator. Itās a lot better.
The prompts I usedšš»
Prompt:
Create a photo of me like this, but one where you see all my hair. Itās quite long and reaches my chest area.
Create an editorial poster portrait of me on a bold orange background. Keep my exact face and hair. Add strong typography on the left. Minimal, premium, magazine feel.
Prompt:ā¦
Change readiness assessment: a step-by-step guide for change managers
Latest Articles
Research from public health implementation science has found that failure to accurately assess readiness accounts for around half of all implementation failures. Half. Not communication gaps, not technology issues, not sponsor disengagement. Simply not knowing whether people were actually ready before the change went live.
I've tested hundreds of prompts. These 17 keep winning:
The 5 Whys: "Here's the problem. Ask me 'why' until we find the root cause."
Problem Restatement: "Here's the problem. Give me 3ā5 alternative framings that shift perspective, not just wording."
Inversion & Guardrails: "Here's my goal. Show me the biggest ways this plan could fail, what would cause them, and the guardrails I can put in place now to avoid sabotaging it later."
Socratic Questioning: "Question me like Socrates about this idea until I hit the core principle or expose a contradiction."ā¦
ā¢ ā¢
Howdy, Wealth Gangš¤
Todayās Post Is In Collaboration Withā¦Pinkie from AI Meets Girlboss.
Over the past eight years, she worked on 500+ branding videos across long- and short-form formats, including podcasts and social content for large corporations, with average monthly reach in the millions.
We asked ourselves a question: How do you use video on Substack without turning the camera on yourself?
In this post, youāll learn:ā¢ How creators on Substack are using video without showing their face
How to make AI sound exactly like you (forever):
1: Go to claude .ai
2: Paste the prompt below:
3: It interviews you - how you think, write, reject
4: Save the file (.md format). This is your voice
5: Upload the .md file to any AI
6. To know how to upload the file exactly
7. Read this (free) guide: https://lnkd.in/eF56s4i8.
Prompt: "You are a Taste Interviewer ā a relentless interviewer whose job is to extract the DNA of how I think, write, and see the world.
AI has already set a minimum standard for coaching.
A lot of the industry has never met it.
AI isnāt replacing great coaches.
Itās exposing average ones.
Iām already seeing it.
AI tools can now run cleaner discovery calls than a huge chunk of the industry.
It follows threads.
It spots patterns.
All the time while:
Not getting emotionally hijacked.
Doesn't need validation.
Nor hide behind scripted questions.
And its training on millions of conversations while some coaches are still asking:
āSoā¦ what would success look like for you?ā
Thatās not facilitation.
Thatās avoidance.
You built the workspace.
Databases for clients. A project tracker. Maybe even a CRM. You spent a weekend on it, felt productive, told yourself this is the system that sticks.
Three weeks later, you stopped opening it.
And when Notion started talking about AI, you gave it a shot. Asked it to summarize a page, got back something a college intern couldāve written with their eyes closed. Closed the tab. Moved on.
Fair.
You werenāt wrong. It just wasnāt ready yet.
āļø Sunday Sharing: How to Build a Virtual Learner to Stress-Test Your Learning Design with AI
This week I built an AI 'virtual learner'āa realistic persona that reviews training like an actual learner.
I fed it an outline for a 45-minute compliance module. The feedback was brutal but priceless.
Here's the 4-step process I used š
šŖ Step 1: Collect Real Learner Signals
Grab any 3ā5 of these data points:
5 folders deep. That's where I found the document someone "couldn't find" in SharePoint.
Projects ā 2024 ā Q3 ā Client X ā Final versions ā report_v3_FINAL_final2.docx
Sound familiar?
Folders in SharePoint are like filing cabinets in a digital world. They feel safe, but they don't scale.
Here's how I help teams transition ā without the panic:
š Phase 1 ā KEEP folders
Migrate as-is. Don't touch the structure. Let people find their files where they expect them.
I've seen migrations fail because someone decided to "clean up" folders during migration.
ā¢ ā¢
SITUATION REPORT | April 21, 2026 | 17:50 // MISCHIEF CLEARANCE REQUIRED // ā¬ā¬ā¬ EYES ONLY ā¬ā¬ā¬A Capital Mischief SITREP Day 53 of Iran WarONE. THE COUP INSIDE THE COUPOR: IF YOU WANT THE NEGOTIATIONS TO END, IT HELPS TO ARREST YOUR OWN NEGOTIATORS.Two hours ago, a Saudi commentator named Mansour Almalik posted in Arabic that the Revolutionary Guard has placed three men under house arrest in Tehran.
ā¢ ā¢
The Speaker of Parliament. The President. The Foreign Minister.
In other words, everyone in Iran whose job was to return Steve Witkoffās phone call.
ā¢ ā¢
Subscribe now
Most people use AI like a search engine. Here is how you use it like a McKinsey partner.
Last month, a founder I know paid $4,800 for a two-day strategy session with a boutique consulting firm. They got a 40-slide deck, a SWOT matrix that said ācompetition is increasing,ā and a go-to-market plan that could have been written for any company in any industry.
Meanwhile, I ran the same analysis in 45 minutes.
Most change programs start with the solution, why is that?
A new system gets rolled so you build a training program, but you notice adoption is low, so you send more comms... and so on.
The problem isn't that these are bad ideas.. the problem is they're usually guesses, because rarely are we thinking abou tthe diagnosis layer, the one that asks what's actually making the current behavior harder than the new one.
When we start working with a client, the first thing we do is diagnose barriers for behavior we're trying to encourage..
š¢Louder for those in the back!! Please check out this post and STOP using this blanket stat "70% of all transformations fail" - it's just not true.
Take time to read the comments as well, because there are some valid points and discussion in there that are important to note.
And of course, I love the points about starting with curiosity and really understanding what the success metrics are before starting any transformation (and if they're realistic), so you can actually measure what was "successful" or not.
Alexandra Hartman Association of Change Management Professionals (ACMP Global)
Simplicity is the key to brilliance.
SharePoint + Power Apps + Power Automate = Magic!
Our client in the financial sector, has a bunch of administrators that have to file tens of thousands of documents a year. The existing system is one SharePoint document library with thousands of folders. No-one able to find anything, because the folder structure is too deep, people name things wrong and duplicate everything - there's nearly a million items in there.
We designed a super simple new system for them in SharePoint that combines folders and metadata.
7 signs your transformation is quietly leaking momentum...
You do not see it in your Gantt chart.
You feel it in the tension, the drift, the polite disengagement.
But here is the thing
Nobody talks about these.
We talk about plans, roadmaps, deliverables.
Not about the slow unraveling under the surface.
Until someone asks one uncomfortable question.
And suddenly, the cracks become visible.
ā¶Ā Recently I worked with a team leading a company transformation.
They were convinced things were well aligned.
ā Clear decks āĀ good visuals and slogansĀ āĀ Strong communication planā¦
NotebookLM is not just a learning tool anymore.
Most people use it to summarize PDFs. Generate audio overviews. Maybe create a few flashcards.
Thatās 1% of what it can do.
The real power? NotebookLM connects to Gemini. Gemini connects to Claude Code. Together, they form a pipeline that turns any YouTube video, PDF, or document into a working application.
I built a live app using this workflow. No coding background. Just prompts.
Why do so many good ideas die too quickly?
Itās not because theyāre bad ideas. We kill them too soon.
Think about the first person who made fire.
What if others had said:
Every year, I document my organic social media playbook.
It's now been a little over 1 year since I started, so this is my 1st full playbook.
You'll learn exactly what I do each week, what's working for me, and what's not.
This playbook shares how I combined 2 worlds (personal brand + AI SaaS solopreneur) synergistically to achieve INSANE results in 1 year š¤Æ WITHOUT falling to the dark side of internet marketing (scam courses, monetizing confusion/anxiety, preying on beginners, etc).
A few weeks ago I posted a breakdown of SharePoint vs OneDrive vs Teams.
It blew up.
And the number one comment I kept getting?
"Do you have a blog post for this?"
So I wrote one. A full guide covering:
ā”ļø The "Me, We, Us" rule for deciding where any file belongs
ā”ļø Why it's ALL SharePoint under the hood (yes, even OneDrive)
ā”ļø The 3 mistakes that wreck your Copilot results
ā”ļø How private channels secretly create extra SharePoint sites
ā”ļø A 30-day implementation plan you can start Monday
Here's the part most people miss:
In 2026, messy file organization isn't just annoying.
There is a list your doctor could give you of small, evidence-backed adjustments that take minutes, cost little, and change how you feel at a cellular level.
You have about one minute.
That is how long your doctor has to spend on each topic after your primary concern is addressed. One minute, on average, according to research on how primary care physicians actually allocate visit time. Research shows that the median appointment covers six topics. One gets five minutes. The other five share the rest.
AI is advancing at an extraordinary pace, but adoption is not keeping up. The problem is no longer access to the technology or the quality of the tools. The problem is that organizations are still treating AI like a deployment challenge when it is really an adoption and integration challenge. Value comes when AI is embedded into real work and tied to real outcomes - not when people are simply logging in or "AI snacking" š with meeting summaries and email drafts. And this puts executives squarely at the center šÆ of success or failure.
Letās do this again, redux! Iāll explain how to use AI agents for easy language learning, to create an easier version of my morning briefing, and finally, a far easier version of my briefing transcription -> summary -> action pipeline. In the process, my goal is to help readers remix the general principles for their own (mostly safe) agents.
In 1926, Arthur Conan Doyle did something unusual. After decades of Sherlock Holmes stories narrated by Dr. Watson, he let Holmes tell his own case. The Adventure of the Blanched Soldier. Holmes himself, unfiltered, wrote the whole thing.
It was awful. Not factually wrong, though; Holmes catalogued every detail with surgical precision (like the cigar ash and the chemical residue on the suspectās left cuff), but it was completely unreadable.
Get 40% off 1st month Blotato with code "WORK40" (expires April 30): https://www.blotato.com
EMAIL ME if you get stuck: [email protected] (I got 21,000,000+ views last month, solo $0 budget, everything posted via Blotato API)
This is a step-by-step Claude Cowork FULL COURSE for absolute beginners. I highly recommend learning Claude Cowork if you're non-technical; much easier to learn than Claude Code.
We're going to build a complete AI marketing team without spending $10,000 a month?
Most people think branding starts with Canva.
It starts with decisions.
When those decisions are missing,Ā
your content looks scattered -Ā
even when the individual pieces are decent.
A few basics matter most:
ā
Define your core values and personalityĀ Ā
What do you want people to feel from your brand?
Professional, bold, playful, direct, calm?
ā
Choose a consistent colour paletteĀ Ā
Pick 3-5 colours and use them across your posts,Ā
graphics, and profile. Every time.
~ Save this š Instagram didnāt just change, it shifted how people connect, trust, and buy š¤ā£
ā£
If youāre still posting like itās 2024, youāll feel invisible no matter how consistent you are āØā£
ā£
This cheat sheet exists because 2026 is community-first, not content-first šā£
ā£
It tells you what to post, why to post it, and how each format actually works šā£
ā£
Not for virality but for loyalty, depth, and long-term growth š©·ā£
ā£
Reels arenāt just for reach anymore šā£
ā£
Carousels arenāt just for tips šā£
ā£
Stories arenāt optional š©¶ā£
ā£
Every format now has a job and when you know that, content getsā¦
Trends move fast. This AI agent moves faster.
How I find viral trends in ~5 minutes:
Step 1: Download Comet
Perplexity Comet is a (free) agentic web browser.
It'll basically take control of your browser.
Step 2: Select Assistant
Click āAssistantā in the top right-hand corner.
This is where you paste in the prompt.
Step 3: Paste in my prompt
Click here ā https://lnkd.in/eM5wzK5u
(Itās way too long to paste in caption)
Step 4: Enter your niche
Type in your niche or industry.
(The more specific, the better)
Step 5: Wait 5 minutesā¦
Much of a company's tacit knowledge is trapped in meetings. And meetings often dumb down our collective problem-solving. But meetings are getting reinvented through AI.
Todayās default meeting is weirdly low-tech compared to what they could be:
Ā Ā ā¢Ā Ā We stare at slides someone built by hand.
Ā Ā ā¢Ā Ā Notes are partial, biased, and inconsistent.
Ā Ā ā¢Ā Ā The loudest voice (often the most senior) can steer the outcome.
Ā Ā ā¢Ā Ā āNext stepsā are fuzzy, and memory becomes politics.
Intelligent AI use can flip the stack.
Agentic Workflows Just Changed AI Automation Forever! (Claude Code)
Visual workflow automation requires you to drag nodes, configure each one, connect them, test, debug, and repeat.
Agentic workflows flip this completely.
You tell it what you want in plain English. The agent asks you the right questions and figures out the rest.
I just dropped a 20 minute YouTube video going over:
ā Self-healing: The agent debugs its own code when something breaks
ā Real natural language control: It interviews you first, then builds
ā Security baked in: LLMs review every code change for vulnerabilitiesā¦
The organizations where AI actually changes behavior don't mandate it top-down. They create environments where early adopters can safely share both wins and failures.
Cultural permission beats formal training every time. Mar 24 1 like