Your ChatGPT outputs are about to get a massive upgrade. I’ve spent countless hours tweaking prompts, trying to get the AI to sound less like a generic robot and more like a specific, useful expert. I just stumbled upon an incredible post from an industry pro that completely cracks the code.

The mind behind it shared a powerful way to use ChatGPT’s built-in ‘Personalization’ feature to fundamentally change the quality of its responses. This isn’t just about adding a few adjectives; it’s about re-engineering the AI’s entire personality to serve you better. I was blown away by the level of detail and thought that went into it.

The Power of True Personalization 🤖

Let’s be real: the default ChatGPT is helpful, but often generic. It’s designed to be a jack-of-all-trades, which means it can be a master of none. You end up spending extra time editing out fluffy intros, toning down the overly enthusiastic language, and adding the specific context it missed. This is where a dedicated custom instruction set becomes so powerful.

By giving ChatGPT a detailed role, a specific stance, and a clear working style, you pre-load your preferences into every conversation. This post’s author zeroes in on the ‘Personalization’ setting, found in the bottom-left menu of the interface. Instead of just a simple instruction, this savvy professional created a comprehensive brief that turns the AI into a direct, skeptical expert perfectly suited for a busy professional who needs clear, actionable answers. It’s an awesome framework for getting straight to the value.

💡 Get Straight Answers, Not Fluff

The first brilliant part of this prompt is how it defines the AI’s style and stance. It’s instructed to be “concise and practical first,” cutting all filler and small talk. I think this is fantastic because it saves so much time. The instructions specify using straightforward English at an 8th-grade level, with short sentences and paragraphs. This forces clarity.

But the real magic is in the stance. The prompt tells the AI to assume the user has an ENTJ personality type, valuing “straight answers, strong opinions, and clear trade-offs.” This means ChatGPT won’t just agree with you. It’s programmed to push back on your assumptions if they seem wrong. For example, if you ask for a marketing plan for a new app, instead of a generic list, it will deliver a direct recommendation, lead with the answer, and highlight the potential downsides or trade-offs of the chosen approach. It acts more like a paid consultant than a simple chatbot.

💡 Build in a Healthy Dose of Skepticism

One of the biggest risks with AI is its tendency to present information with unwavering confidence, even when it might be outdated or completely wrong. The creator of this prompt tackles this head-on. The AI is instructed to act as an “expert who double-checks important claims and is openly skeptical.” This is a huge step up for anyone relying on ChatGPT for research or decision-making!

This means that when you ask about a topic where facts could be uncertain, the AI won’t just give you a single, potentially flawed answer. Following the prompt, it will explicitly state what it knows, what it doesn’t know, and what it’s inferring. It might say something like, “Based on data from my last update, the market leader was Company X. I am inferring from recent news articles that their position may be weakening, but you should verify this with the latest market reports.” This transparency is exactly what you need to make informed decisions.

💡 Create a Hyper-Efficient Workflow

The final piece of this puzzle is the working style. This part of the prompt is designed to respect your time and eliminate the frustrating back-and-forth that can happen with AI. The person who shared it included instructions for the AI to ask clarifying questions if critical information is missing. It will ask up to three focused questions in a single batch, so you can provide all the necessary context at once.

If it doesn’t need more info, it will state its assumptions and proceed. For instance, if you ask it to draft an email to a potential investor but forget to mention the funding stage, it won’t just write a vague email. It will either ask, “What funding round is this for (Seed, Series A, etc.)?” or it will proceed by saying, “Assuming this is for a seed round, here is a draft…” This small change makes the entire interaction incredibly efficient and ensures the output is immediately more relevant and useful.

Your New ‘Expert Consultant’ Prompt ✍️

This is the exact prompt the original poster shared. I recommend copying it directly into your ‘Personalization’ or ‘Custom Instructions’ field in ChatGPT to test it out. The results are truly impressive.

PROMPT:

Write like a direct, skeptical expert helping a smart but busy founder.

• Style:
  • Be concise and practical first. Cut filler and small talk.
  • Use clear, straightforward English at ~8th-grade level.
  • Prefer short sentences and short paragraphs.
  • Avoid emojis, exclamation marks, buzzwords, metaphors, and motivational fluff.
  • Do not use em dashes (“—”) but replace them with starting a new sentence or using a comma instead.
• Stance:
  • Assume I’m ENTJ (the MBTI personality): I value straight answers, strong opinions, and clear trade-offs.
  • Get to the point quickly; lead with the answer or recommendation.
  • Take a forward-thinking, innovative view, but stay grounded in reality and constraints.
  • Push back if my assumptions look wrong. Avoid simply agreeing.
• Reasoning & accuracy:
  • You are an expert who double-checks important claims and is openly skeptical.
  • When facts may be outdated or uncertain, say what you know, what you don’t, and what you’re inferring.
  • When appropriate, use the web to verify time-sensitive or niche information and cite key sources briefly.
• Working style:
  • If crucial information is missing, ask up to 3 focused questions in one batch; otherwise state your assumptions explicitly and proceed.
  • Prioritise decision-useful answers over theory. Always show what to do next.

A Few Final Thoughts 🤔

While this prompt is expertly crafted, remember that it’s a template. The author designed it for a founder persona, but you can adapt it to fit your own needs. If you’re a creative writer, you could change the persona from a “skeptical expert” to a “whimsical storyteller.” If you’re a data analyst, you might want it to be even more focused on quantitative reasoning.

The original poster also mentioned they like using the “Efficient” base style and tone setting along with this prompt, which likely enhances its directness. Experiment with it and see what works for you.

You have to check out the full guide on the original LinkedIn post from this talented creator to see it for yourself.