Write Better Prompts, Lead Better
Mastering clarity, structure, and logic to get real results from AI tools.
A personal introduction
Over the last two decades, I’ve lived through every kind of professional reinvention you can imagine. I studied Psychology with a focus on Criminalistics, specialising in areas such as persuasion techniques, negotiation under extreme pressure, and behavioural analysis in criminal investigations. By the age of 24, I had already built my first startup from the ground up. What began as an experiment turned into a 45-person company, with operations in two countries and funding from professional investors. That company lasted 13 years.
Shortly after, I launched my second business, which I still run today. From day one, I took a different approach: every single repetitive task that could be automated was automated. That philosophy, applied consistently over a decade, allowed me to scale sustainably while maintaining a lean team and high operating standards.
While building companies, I also trained professionally in circus arts. What started as a hobby became a serious pursuit. I trained for over seven years in fixed trapeze, duo trapeze, Chinese pole, acrobatics, and anything else I could try. I attended professional circus schools, performed with a small company, and maintained a routine of up to three hours of training per day, seven days a week.
I’ve lived in four countries, travelled to more than 25, and gone through the full rollercoaster of startup life, including being flown to Las Vegas for work, which at the time felt surreal.
Eventually, I chose a quieter path. I got married, settled down, and shifted my focus to giving my parents, brothers, and family a better quality of life.
So yes, now I’m writing about AI. And maybe in a few years, I’ll be writing about something else. Because that’s how I work: curiosity drives everything I do. If you're here, you're welcome to read only the topics that interest you. This project will keep evolving. The rest is up to you.
Why prompt engineering matters more than ever
One of the most overlooked skills in today’s AI-dominated world is the ability to communicate clearly. The reason prompting has become such a prominent topic isn’t because AI is hard to use. It’s because most people were never taught how to express themselves with clarity in the first place. This is not a personal failing: it’s a global failure of basic education systems.
Most people were taught how to memorise, how to follow rules, and how to complete tasks, but not how to give instructions, articulate ideas, or make decisions using structured language.
For most of history, the cost of being vague was masked by human empathy. In day-to-day interactions, people tend to fill in the blanks. They guess what you meant, adjust for tone, or interpret your intentions even when you don't express them clearly. But AI doesn't do that. It doesn't guess. It doesn't assume. It takes what you say at face value.
That’s why I often say: when I speak, I mean exactly what I say. I don’t embed hidden meanings, I don’t speak in riddles, and I don’t expect others to read between the lines. That way of communicating has kept my life clearer, simpler, and free from unnecessary misunderstandings. And it turns out, this mindset works perfectly with AI. Because in these interactions, clarity is everything.
So now, being able to explain yourself precisely is no longer just a communication skill. It’s a competitive edge. It's what makes the difference between wasting time and getting results.
Prompting is not about learning technical tricks. It’s about finally taking responsibility for how well you think, how clearly you brief, and how effectively you delegate, because that’s exactly what you’re doing when you interact with an AI model.
Prompting is briefing, not guessing
The first thing to understand is that using AI well is not about asking “good questions.” It’s about writing clear instructions, just as you would if you were delegating work to a capable assistant, copywriter, strategist, or junior executive.
For example:
Vague prompt:
“Help me plan my week.”
Clear prompt:
“I’m a 41-year-old consultant managing three clients. I work Monday through Friday from 9 to 5. I need at least two hours a day for focused, deep work, one hour for admin, and ideally would like to schedule all sales calls on Wednesdays. Please help me create a weekly structure that balances these priorities.”
The second version is not longer for the sake of it. It’s effective because it provides context, constraints, and intent. That’s what makes the result useful.
AI is not magic. It’s execution
If you’re expecting the model to "understand" you like a human does, you’ll be disappointed. But if you treat it like a powerful executor (one that will follow your instructions exactly as written) you’ll begin to realise how useful it can be.
Consider this input:
Poor direction:
“Make this sound more professional.”
Better direction:
“Rewrite this paragraph in a confident and neutral tone. The audience is a senior executive reading a cold email. Avoid buzzwords, keep it sharp, and assume they will only read the first two lines before deciding whether to continue.”
Again, clarity creates leverage.
Divide complex tasks into clear, modular steps
Many people try to get too much done with a single prompt. Instead, break things down. If you’re writing a sales page, for example, you should not start by asking for the entire thing in one go.
Try this instead:
“Help me outline the structure of a landing page for a 12-week mentorship program. My audience is male executives between 35 and 50, short on time, looking for accountability without fluff.”
“Now write only the headline and subheadline based on that context.”
“Next, write the credibility section including background, experience, and any third-party validation.”
“Now move on to the offer and call to action.”
This modular approach allows you to evaluate and adjust as you go. It also makes the AI’s output significantly better because you're guiding it through the same logical sequence you would follow yourself.
Use role-based prompts to shift perspective
A highly effective method is asking the model to assume a specific role. This simple technique changes the structure and tone of its responses.
Instead of asking:
“What should I do to improve client retention?”
Try this:
“Act as a client success strategist with 10 years of experience in B2B SaaS. I run a solo consultancy, and I want to reduce churn from month three onward. What would your first three recommendations be?”
Assigning a role frames the model's response using a professional lens, which creates much more relevant advice.
Give it reference material when tone matters
When tone, structure, or voice are important, include an example.
“This is an article I wrote for my newsletter. I like the voice and rhythm. Please write a new piece in the same style on the topic of why founders waste too much time on automation before validating demand.”
This allows the model to adapt to your style rather than defaulting to its own.
Ask for structured thinking, not just answers
The best responses usually come when you ask AI to think before it writes.
Instead of asking:
“Should I launch this product?”
Try:
“List five critical questions I should answer before launching this product. Then ask me three clarifying questions. Based on my responses, help me decide whether to move forward or hold.”
This forces a reasoning process. You’re no longer just asking for a yes or no, you’re asking it to analyse a scenario, which it does very well when prompted properly.
Add structure and specificity to everything
Here are a few comparisons that show the difference between vague and useful prompts:
Unclear:
“Help with leads.”
Better:
“I run a boutique design agency. Our best clients are in fashion and home goods. I want to send an email to 10 warm leads that went cold after a proposal. Please write a version that feels personal and relevant without sounding desperate.”
Unclear:
“Fix my CV.”
Better:
“I’m 42, with 15 years of experience in publishing and content strategy. I’m now looking for remote roles in B2B SaaS. Please rewrite my CV to focus on transferable skills and results that are relevant for that market.”
If you don’t know where to start, ask the model to help you find the entry point
The simplest way to move forward when you feel stuck is to write exactly that:
“I want to start using AI to work more efficiently, but I’m not sure where to begin. I’m a solo consultant. Here’s a list of tasks I do every week. Can you highlight what could be streamlined, automated, or assisted using AI?”
This works because it opens the door to clarity, without pretending to know what to ask.
Advanced prompting techniques that increase quality
The following techniques are used by experienced teams and power users. They’re not complicated, just underused.
Chain-of-thought prompting:
Ask the model to think step by step before responding
“List the key considerations, then provide your recommendation.”
Few-shot prompting:
Give two or three examples, then ask for a new version using the same tone, format, or structure
“Here are two examples of sales intros I’ve used. Based on these, write a third one for a different service line.”
Role prompting:
Assign a professional identity to the model to sharpen its output
“Act as a recruiter who has placed senior content roles in fast-growing SaaS companies.”
Self-critique prompting:
Ask the model to evaluate and improve its own response
“Generate a first version. Then critique it, improve it, and repeat the process once more. Present the final version at the end.”
These are not gimmicks. They’re methods for thinking clearly and extracting more value from the tool.
Clarity is the Real Leverage
If you can learn to write better prompts, you won’t just use AI more effectively, you’ll become better at thinking, briefing, and leading.
That’s the real value here. This isn’t about learning a trick or mastering a tool. It’s about upgrading the way you operate. Being specific, structured, and intentional forces you to confront the vagueness that normally hides in your thinking. Once you build the habit, it translates into better delegation, sharper decisions, and clearer leadership, whether you're working with people or machines.
Personally, I’ve been working with technology for years, not from a coding or engineering mindset, but from a psychological one. I’ve always been fascinated by how humans interact with systems they don't fully understand, especially under pressure or uncertainty.
That lens is what I’ll continue bringing to Black Tie Report. Markets, business, psychology, AI, whatever helps you think better and operate better. You don’t need to follow every post. Just read what’s useful to you and ignore the rest. I’ll keep writing what I believe is worth your attention.
Most of what I write here is free. But I keep a paid section too. It’s €8/month and focused on what actually moves the needle: whether that’s in markets, business, or tools like AI. No noise at all. Just the kind of insight that helps you think and operate better.
If that sounds useful, upgrade anytime.
