Mastering New ChatGPT-5

In partnership with

Today’s Sponsor

Where the smartest investors start their day

The Alternative Investing Report (A.I.R.) helps you get smarter on alternative assets from crypto treasury companies to Pre-IPO venture secondaries to private credit and more.

Join 100,000+ investors to get the latest insights and trends driving private markets, a weekly investment pick from a notable investor, and bonus offers to join top private market platforms and managers. And it’s totally free.

Read AIR today for free

Are you getting lackluster results from ChatGPT-5? You're not alone. A staggering 95% of users are experiencing worse outputs, not because the model is weaker, but because they're relying on outdated prompting techniques.

OpenAI has fundamentally changed GPT-5's architecture, rendering old methods less effective. This article will equip you with the knowledge and skills to harness the full power of ChatGPT-5.

This comprehensive guide is designed for anyone who wants to improve their AI productivity, from marketing professionals and content creators to data scientists and software developers. We'll delve into the architectural changes that impact performance and, more importantly, provide five proven techniques to dramatically improve your outputs. Get ready to master ChatGPT-5 and unlock its true potential.

Understanding GPT-5's New Architecture

To truly master ChatGPT-5, it's crucial to understand the underlying changes OpenAI implemented. Two key updates have significantly altered how the model responds to prompts: model consolidation and enhanced instruction following.

The Consolidation Update

Previously, ChatGPT Plus users had access to a range of models. Now, the options have been streamlined to three core versions: GPT-5, GPT-5 Thinking Mini and GPT-5 Thinking. While fewer options seem like a user-friendly improvement, this consolidation introduced an invisible router system.

This router acts like a customer service representative, attempting to direct your request to the most appropriate model. However, this router doesn't always work perfectly. Sometimes, a simple prompt might be routed to a less capable model, resulting in suboptimal output.

Furthermore, the more powerful models are more expensive to run. This creates an incentive for OpenAI to route users to the fastest (and often less intelligent) option whenever possible. Understanding this routing system is the first step in optimizing your prompts for better results.

The Instruction Following Update

GPT-5 has been specifically trained with AI agents in mind, making it exceptionally precise in following instructions. This means that GPT-5 now adheres to our prompts with extreme precision.

While this sounds like a positive change, it presents a double-edged sword. Unlike previous models that could infer intent from vague prompts, GPT-5 struggles to "guess" what you want when your instructions are poorly constructed. If you stick with your old prompting techniques, you'll likely get worse results.

Essential Prompting Techniques for GPT-5

Now that we understand the architectural changes, let's dive into the practical techniques that will significantly improve your ChatGPT-5 outputs.

Router Nudge Phrases (Effort: Low)

The simplest and perhaps most impactful technique is using "router nudge phrases." By adding just a few words to the end of your prompts, you can influence the invisible router to select a higher-reasoning model.

These phrases effectively "nudge" the system towards a more thorough and thoughtful response. According to testing, the following phrases reliably trigger deeper reasoning:

• "Think hard about this"

• "Think deeply about this"

• "Think carefully"

These explicit phrases prompt GPT-5 to engage its more powerful reasoning capabilities.

Real-World Example:

Consider this scenario: you want to understand the best investment strategy for your cash. You ask ChatGPT-5, "What are the pros and cons of putting my cash in a low-cost index fund versus a money market account?"

Without a nudge phrase, the output might be generic and lack depth. However, by adding "Think hard about this" to the end of the prompt, you're more likely to receive a comprehensive analysis that considers second-order effects and provides actionable insights.

Best Practices for Implementation:

• Use nudge phrases for high-stakes tasks where missing second-order effects could have significant consequences.

• Experiment with different phrases to see which ones work best for your specific use cases.

• Remember that these phrases are most effective for free users, but Plus users can still benefit from them.

Verbosity Control (Effort: Low)

In addition to reasoning depth, ChatGPT-5's invisible router also controls output length, or "verbosity." Just as we can nudge the router towards deeper reasoning, we can use specific phrases to control how long or short GPT-5's outputs are.

After extensive testing, three power phrases for controlling verbosity:

• Low Verbosity Outputs: Ideal for when you need only critical information. Use phrases like, "Give me the bottom line in 100 words or less. Use markdown for clarity and structure."

• Medium Verbosity Outputs: Best when you need key takeaways plus context. Try, "Aim for a concise 3 to 5 paragraph explanation."

• High Verbosity Outputs: Great for comprehensive documents like project briefs or research summaries. Use, "Provide a comprehensive and detailed breakdown, 600 to 800 words."

Practical Examples:

• Slack Message to a CMO: Instead of a lengthy email, use a low verbosity phrase to get a concise update.

• Team Meeting Explanation: Use a medium verbosity phrase to provide key takeaways and context without losing your audience's attention.

• Project Brief: Use a high verbosity phrase to generate a comprehensive document for internal team kickoff.

Implementation Tips:

• Add these reusable phrases to your text expander app for easy access.

• Adjust the word count based on your specific needs. GPT-5 handles specific word counts much better than previous models.

Advanced Prompting Strategies

Once you've mastered the essential techniques, you can move on to more advanced strategies that require a bit more effort but offer significant returns.

OpenAI's Prompt Optimizer (Effort: Medium)

Did you know that OpenAI has an official prompt optimizer tool that rewrites your prompts for GPT-5? This tool can significantly improve the quality and effectiveness of your prompts.

The Prompt Optimizer Tool is designed to automatically analyze and improve prompts for better results with AI models like GPT-5. It works using a collaborative multi-agent system, where specialized AI agents detect and resolve common prompt issues including contradictions, missing or unclear format specifications, and inconsistencies between instructions and few-shot examples.

The tool offers features such as prompt rewriting for clarity and effectiveness, automated evaluation of improvements using Evals and Agent SDK, and targeted migration/adaptation for newer AI models including GPT-5.

The tool makes three consistent improvements:

1. Structure Addition: Breaks down unstructured text into logically distinct sections.

2. Vagueness Elimination: Clarifies ambiguous language and provides specific instructions.

3. Error Handling: Adds reminders to ask for clarification when prompts contain contradictions or missing information.

Free Workaround:

Creating a separate developer account with OpenAI and adding a payment method can be cumbersome. Fortunately, there's a free workaround: use a meta prompt with ChatGPT-5.

Here's an example of a meta prompt you can use:

This approach leverages GPT-5's ability to critique and improve its own instructions.

Integration with Text Expander Tools:

Add this meta prompt to your text expander app for quick access. This effectively gives you OpenAI's prompt optimizer tool for free.

XML Sandwich Technique (Effort: Medium)

If you've used the meta prompt from the previous section, you may have noticed that the optimized prompts sometimes include angle brackets wrapping different sections of the text. These are called XML tags, and OpenAI recommends organizing your instructions this way.

XML tags act as labeled boxes, explicitly labeling each component of your prompt:

• <context>: Background information

• <task>: The specific task you want GPT-5 to perform

• <output_format>: The desired output format

Instead of dumping everything into one paragraph and hoping GPT-5 figures it out, you're providing clear and structured instructions.

Implementation Guide:

Let's say you want GPT-5 to help you prepare for a product manager interview. Instead of saying, "Help me prepare for a product manager interview. Here's all the background information," you can use the XML sandwich technique:

<task>
Act as a hiring manager and, based on my resume and the job description, ask me three questions I'm likely to face.
</task>

<resume>
[Paste your resume here]
</resume>

<job_description>
[Paste the job description here]
</job_description>

This structured approach helps ChatGPT-5 better comprehend its task, leading to a better outcome.

Template Creation and Reuse Strategies:

Save a template in your text expander with defaults so you don't have to type it out every time. For example, you can include a <tone> tag with a default value of "user-friendly and conversational."

This structure works especially well for custom instructions in custom GPTs and ChatGPT projects. Since these are recurring use cases, it's worth investing the time to create the XML sandwich.

The Perfection Loop

The final and most advanced technique is the "perfection loop." This technique leverages GPT-5's ability to critique itself to achieve the best possible results.

Instead of accepting GPT-5's first response and manually asking for improvements, you tell it upfront to:

1. Create its own definition of excellence.

2. Grade its own work.

3. Keep iterating internally until it achieves the best result.

This process mimics hiring a designer who grades their initial draft against their own quality checklist and keeps redesigning until they score a 10 out of 10.

Implementation Steps:

1. Start with a clear prompt that defines the task.

2. Instruct GPT-5 to develop an internal rubric for what constitutes a world-class output.

3. Tell GPT-5 to internally iterate and refine the draft until it scores top marks against its own rubric.

Examples:

• "Write a market analysis report on the enterprise AI industry. Before you begin, develop an internal rubric for what constitutes a world-class market analysis report. Internally iterate and refine the draft until it scores top marks against your rubric."

• "Draft an outline for my QBR (quarterly business review) presentation. Before you begin, create an internal rubric with five criteria for a perfect QBR. Then use that rubric to internally iterate the outline until your response scores 10 out of 10."

Universal Perfection Loop Template:

Before providing your response, you must complete the following internal process:

Create an internal, detailed rubric that defines the characteristics of a 'world-class' (10/10) answer specifically tailored to the my request.

Internally generate and iterate your proposed answer, critically scoring each draft against the rubric until a perfect score of 10/10 is achieved.

Show only the final, perfect output. Do not display the rubric, the drafts, the scores, or any part of this internal thinking process.

When to Use This Technique:

The perfection loop works best for complex zero-to-one tasks, such as creating finished documents from scratch or writing production-ready code.

Practical Implementation Guide

Now that you're armed with these powerful techniques, let's discuss how to implement them effectively.

When to Use Each Technique

The best approach is to stack these techniques on top of each other. You can use nudge phrases with verbosity control, with XML sandwiches, with the perfection loop. It's not an either/or situation; it's a "yes, and" approach.

Best Practices and Tips

• Template Creation and Management: Create templates for common tasks and store them in a central location for easy access.

• Text Expander Integration: Use a text expander app to quickly insert phrases and templates into your prompts.

• Quality Control Measures: Always review GPT-5's outputs to ensure they meet your standards.

• Performance Monitoring: Track the performance of your prompts and make adjustments as needed.

Conclusion

Mastering ChatGPT-5 requires understanding its new architecture and adopting new prompting techniques. By implementing the strategies outlined in this article – router nudge phrases, verbosity control, OpenAI's prompt optimizer, the XML sandwich technique, and the perfection loop – you can dramatically improve your AI outputs and unlock the full potential of GPT-5.

These techniques are not just theoretical concepts; they are practical tools that you can use to enhance your productivity and achieve better results with AI. Start experimenting with these methods today and experience the difference for yourself. With the right approach, you can transform ChatGPT-5 from a frustrating tool into a powerful ally in your work and personal life.

That’s all for today, folks!

I hope you enjoyed this issue and we can't wait to bring you even more exciting content soon. Look out for our next email.

Kira

Productivity Tech X.

Latest Video:

The best way to support us is by checking out our sponsors and partners.

Today’s Sponsor

Run IRL ads as easily as PPC

AdQuick unlocks the benefits of Out Of Home (OOH) advertising in a way no one else has. Approaching the problem with eyes to performance, created for marketers with the engineering excellence you’ve come to expect for the internet.

Marketers agree OOH is one of the best ways for building brand awareness, reaching new customers, and reinforcing your brand message. It’s just been difficult to scale. But with AdQuick, you can plan, deploy and measure campaigns as easily as digital ads, making them a no-brainer to add to your team’s toolbox.

You can learn more at AdQuick.com

Ready to Take the Next Step?

Transform your financial future by choosing One idea / One AI tool / One passive income stream etc to start this month.

Whether you're drawn to creating digital courses, investing in dividend stocks, or building online assets portfolio, focus your energy on mastering that single revenue channel first.

Small, consistent actions today. Like researching your market or setting up that first investment account will compound into meaningful income tomorrow.

👉 Join our exclusive community for more tips, tricks and insights on generating additional income. Click here to subscribe and never miss an update!

Cheers to your financial success,

Grow Your Income with Productivity Tech X Wealth Hacks 🖋️✨