10x Your ChatGPT Account With These 3 Tweaks
Improve an AI account with durable instructions, intentional memory, and organized projects that preserve the context needed for recurring work.
Get the next practical AI marketing episode wherever you listen.
Make Your AI Account Work Like It Knows You
An AI account becomes more useful when it has clear instructions, durable context, and an organized place for related work. These are not advanced prompting tricks. They are account-level habits that reduce repetition and make every conversation start closer to the quality you need.
The source identifies three practical levers: customize how the assistant works with you, deliberately build useful memory, and separate ongoing work into projects. Each one solves a different problem. Together, they turn a blank chat window into a more consistent working environment.
Give Clear Account-Level Instructions
Do not expect an assistant to infer preferences you have never stated. If you want it to question weak ideas, ask clarifying questions, use a certain tone, avoid a formatting habit, or follow specific blog rules, write those expectations down in the account’s customization area.
The most valuable instructions tend to come from repeated friction. Every time the assistant does something you consistently dislike, decide whether it is important enough to prevent in future work. A simple rule such as avoiding emojis in titles or limiting a style choice can save a small annoyance across hundreds of outputs.
The same applies to your role and work context. Describe the work you do, the audience you serve, and the outcomes you care about in enough detail for the assistant to make better default assumptions. This is not a substitute for project-specific context, but it gives the account a useful baseline.
Build Memory Around What Matters
Memory should hold information that improves many future interactions: core values, long-term goals, decision-making preferences, professional strengths, and stable characteristics that shape the kind of help you need. It is not the place for every temporary task instruction or isolated detail.
The source recommends a deliberate discovery conversation. Ask the assistant to review what it already knows, then interview you about the values, motivations, preferences, and goals that would help it support you across different contexts. Answer a few questions at a time, preferably in your own natural language, and explicitly ask the system to remember the durable points.
This exercise also helps you clarify your own operating principles. When you articulate how you want to be experienced in high-stakes moments or what tradeoffs matter most, you create guidance that can inform both the assistant and your own decisions.
Use Projects to Contain Context
Projects organize conversations around a sustained area of work or life. A project may hold a podcast, a client engagement, a campaign, a learning goal, or a personal topic. Grouping those conversations gives the assistant a clearer local context and makes it easier for you to find prior decisions.
Choose the memory boundary intentionally at the beginning. Some work benefits from access to your broader account history. Other work, especially client work or sensitive personal topics, should remain isolated so that its context does not mix with unrelated conversations. The source describes this as project-only context: a separate space for related work.
Organization is more than tidiness. A project lets ongoing work compound. Earlier decisions, source documents, and useful conversations remain together, which reduces the need to rebuild the background in every new chat.
Review the Boundary Between Convenience and Privacy
More context can improve personalization, but it also calls for better judgment about what you share. Follow your organization’s data rules and avoid placing sensitive client or personal information into a system unless you are authorized to do so. Turn on memory or reuse context because it serves a real purpose, not merely because the option exists.
Start with one upgrade. Write a short set of account instructions based on recurring friction, create a focused memory conversation, or set up a project for work you return to every week. The value comes from using the account deliberately enough that its context supports your thinking instead of forcing you to repeat yourself.
In this AI marketing podcast episode, Dan Sanchez walks through three transformative upgrades that anyone can make to their free ChatGPT account to dramatically increase output quality.
My Favorite AI Tools
- Tella (AI screen recording app I record all my solo episodes with) - https://danchez.com/tella
- High Level (My all-in-one marketing platform) - https://danchez.com/highlevel
- Zencastr (Best AI podcast tool) - https://danchez.com/podtools
- n8n (My favorite AI automation tool) - https://danchez.com/n8n
Custom Instructions:
Use quick and clever humor when appropriate and adopt a skeptical, questioning approach. Challenge my ideas if they are weak or ask questions proactively to gain clarification if more details are needed.
More About You Section:
Danchez ChatGPT Instructions
- Always use “dive” instead of “delve.”
- Never place an emoji before a title.
- In email scripts, use {{contact.first_name}} for personalization.
- Don’t preface responses with explanations about following these instructions (just answer directly).
- When I ask for markdown, return it in a fenced code block.
- When writing a blog, article, social post, report, proposal, or presentation, create it using the canvas feature.
- Avoid using em dashes (—).
Blog Post Rules
- Use clear structure: headers, subheaders, bullets/numbers, bold for emphasis, and short paragraphs.
- Write to the reader, from the writer’s own perspective (e.g. “Here’s what I found works for me…”).
- Maintain a casual but authoritative tone—like a neighbor sharing a recent solution.
- If adapting from other content, don’t reference or quote the source; make the post standalone as if the reader hasn’t seen the original.
Memory Map Prompt:
I want you to understand me on a foundational level, gathering information that would be helpful across a broad range of future interactions and topics.
Start by reviewing what you already know about me, if anything, and then build on that by asking questions that go beyond my current needs or immediate tasks. Focus on learning about my core values, preferences, long-term goals, and unique personal and professional characteristics. Your goal is to ask about things that, once known, will help you understand my perspective, decision-making style, and motivations in any context.
Prioritize questions that would help you provide relevant and personalized support across multiple areas of life, not just immediate requests. Think about who I am holistically and ask what would help you serve me well over time.
Timestamps
- 00:00 - Intro: Why most users miss out on ChatGPT's real potential
- 00:34 - Upgrade #1: Settings and custom instructions walkthrough
- 02:00 - Personality Customization: How to make ChatGPT push back and be more helpful
- 03:00 - About You Setup: Personalizing for better alignment with your work style
- 05:00 - Upgrade #2: How to activate and leverage ChatGPT's long-term memory
- 07:00 - Memory Map Prompt: The exact prompt to supercharge ChatGPT’s contextual understanding
- 10:00 - Upgrade #3: Using Projects to compartmentalize and focus ChatGPT conversations
Speaker: Most people never unlock 80% of ChatGPT's real power. And in this video, I wanted to show you three easy changes you could make to your account just one time. One time that I swear will multiply the amount of quality outputs that you're getting from your AI tool.
Speaker: So hang with me. i want to walk you through every single step. These are power user moves that hardly anybody ever uses. And the best thing is you don't even need a premium account. You don't need a pro account. You just need the free version of chat GPT. You do have to have an account.
Speaker: So create your free account and let's get started with these upgrades. So the first thing I want to show you is upgrading these settings across the account. Nobody ever does it. So go in and do this today. Go over and find your name in your account.
Speaker: Go to settings. Go to personalization. And these things will jump around from time to time, but it's usually under settings and personalization. And you can find it exactly here. Make sure to enable customizations. That's what this is all about. I'm just going to walk through step by step how I'm customizing it. I want you to steal everything.
Speaker: all of it. I will even copy and paste the exact settings that I'm using for my account to use as a starting point for you. If you're listening, just know that this is available via video on Apple, Spotify, or YouTube. So you're gonna have to come back and watch it another time, but I'll be walking you through the process so you know what you need to do later on.
Speaker: So I'm walking through enabling customization. I am not going to change the personality. actually like the default personality of ChatGPT, but you can change it from default to cynical, robot, listener, nerd.
Speaker: Whatever floats your boat. This one doesn't really matter. I prefer the cheerful and adaptive one because it's just nice to work with. But here's where it really starts to get different. Custom instructions for its personality. This is where you really want to hammer it down. I recommend copying exactly what I have until you have a very specific need.
Speaker: for its personality instructions, I recommend using this. I say, use quick and clever humor when appropriate and adopt a skeptical questioning approach. Challenge my ideas if they are weak or ask questions proactively to gain clarification if more details are needed.
Speaker: Because Chat2PT often just goes with you. I want to be skeptical. also want to be a little funny. I like it a little bit every once in a while if it does so in a funny way, if it's s skeptical, but with a twist of humor to it to lighten the mood.
Speaker: But we need ChatGPT to push back on us. The ChatGPT by default is a little bit too agreeable. And I find that adding just these few lines makes it a lot less agreeable, which is good.
Speaker: It needs to challenge your thinking in order to make you and the outputs that you're getting from it better. Let's move on. Tell it a little bit about you. I give it my nickname, Danchez. That's what people call me by.
Speaker: So I also tell it a little bit about my occupation. Don't just put in your current job title. Tell it who you think you are. i put in AI marketer, educator, creator. You could put something else.
Speaker: but make sure to be descriptive. The more descriptive you can be in all these things, the better ChatGPT can align to who you are, what you do, and what you want out of this tool. On the more about you section, this is where you can really load it down with some details.
Speaker: Let me share with you what's in mind. I have two sets of instruction. I have the Danchez ChatGPT instructions, And, and I have the blog post rules, which I'll tell you a little bit more about in a second.
Speaker: But every time you find that ChatGPT does something, you're like, ugh, I wish it would stop doing that. Well, guess what? It's your job to tell it. For example, almost all these Danchez ChatGPT instructions are things that I got tired of ChatGPT doing. So I said, stop doing it. and For example, some of them I have here are never place an emoji before a title because ugh. Who wants that?
Speaker: I like emojis sparingly sometimes, but I never liked the before title ever. Across no circumstances, emoji before title, a good thing when I'm having it right content.
Speaker: I even ask it to avoid using em dashes. You can put instructions in here. Most of the time it will follow it. There are a few like trying to avoid em dashes where it's just so hard coded in to use em dashes that... it's It's just going to use them anyway. But still, I try to curb some of it by saying, avoid using em dashes.
Speaker: Avoid using this. Use the Canvas feature whenever I'm asking for these things. And it does a lot better at remembering because it's in the instructions. You told it to. you How would you expect an employee to remember what you want? It's like trying to read your mind. We've all had the experience of trying to read people's minds and it being disappointing when they get mad at us for something we had no idea about, right?
Speaker: So stop trying to have Chai Chippity read your mind. If it's really important, you want it to remember every single time, this is where you put it. I actually put blog post rules in here because every time it wrote a blog post, I thought it sucked.
Speaker: So I wrote very specific instructions of what I wanted in a blog post every single time. It's five... about five rules for how to write a blog post according to Danchez. You might have different rules, whether you like my rules or you want your own rules.
Speaker: Again, I'm gonna put this whole thing in the description of this podcast or YouTube video. Go and check it out, copy and paste it, steal it, modify it, make your own. That's why this is so important. You need to make all of this your own.
Speaker: So also, Settings to change if you don't have these turned on turn them on memory reference saved memories You want chat GPT to remember and put to work all the different memories you've ever had in your account You might think this is a privacy problem It's not because chat GPT already knows it all or open AI has access to all your conversations You might as well put all that memory to work for you because it's already in the server If you didn't want it in the server, don't tell it to chat GPT in the first place By turning this on, ChatGPT can now leverage it for your sake. Reference chat history, you want that turned on. Reference record history, you want that turned on.
Speaker: All these advanced modes right here, you want them turned on because you're going to have to learn how to use all of them eventually. And some of them will come in future videos from this podcast. So, next thing is this memory. I skipped over this because this is the whole second part. In the first part of this video, I wanted to show you the instructions on how to make this thing sing, but in the second part, I want to show you how to leverage the memory.
Speaker: Now, ChatGPT can remember all of the different conversations you've had. It's not actually remembering every single word you've spoken to it before, but it creates summaries of every conversation. The farther back the conversation you've had with it, the shorter the summary gets.
Speaker: That way it can chunk it all and kind of remember it. Kind of like the human brain. You remember things more recently and have little summaries of it from years ago. ChatGPT works similarly, but, but there's this extra memory feature.
Speaker: And I want you to consider these to be like core memories. Things that are really important that you always want it to remember. See, I'm going to open up mine. You can see I have a bunch of different things loaded in mine. We want to load those up for you.
Speaker: Not with general things, not with blog post things. Shoot, you put the blog post things in this section right here yeah in the personalization tab. But we still want to load this memory section with things that are personal to you, personal.
Speaker: Personal as in things you don't want other people know but personal as in things ChatGPT needs to know in order to create better outputs for you. Let me show you one prompt I have one mega amazing prompt I have that you can put it in ChatGPT so that it knows all the right things and stores it in its memory.
Speaker: This is what I call the memory map prompt. Let me read it to you and then you'll understand why it's so good. It says, i want you to understand me on a foundational level gathering information that would be helpful across a broad range of future interactions and topics start by reviewing what you already know about me if anything then build on that by asking questions that go beyond my current needs or immediate Focus on learning about my core values, preferences, long-term goals, and unique personal professional characteristics.
Speaker: Your goal is to ask about things that, once known, will help you understand my perspective, decision-making style, and motivations in any context. Prioritize questions that would help you provide relevant and personalized support across multiple areas of life, not just immediate requests.
Speaker: Think about who I am holistically and ask what would help you serve me well. over time. Yes, it's a loaded prompt, but I promise you put this one prompt into chat GPT and it will ask you all the right questions.
Speaker: You then need to spend time answering all those questions. I prefer doing it in dictation format. Let me show you how it looks like. So going to go ahead and load up chat GPT. You're going to want to change this to thinking mode. It might also be under this plus button and over here, you want to change it to thinking mode. You really wanted to look through all the different conversations you've had and spend some time thinking about what questions it needs to ask you.
Speaker: Copy the prompt that I give you. Again, this prompt will be in the description for this podcast or YouTube video. And hit go. And you can see it's given me a number of questions. First, it summarized what it already knew about me, which it knows a lot. But then it went and asked me a number of different questions based on what it already knows.
Speaker: And if it doesn't know anything, then it'll still ask great general questions for you to answer on the first pass. What I want you to do is then hit the dictation button, this little microphone button, and it will actually start to record your answer. Here, I'll record an answer to one of my questions now.
Speaker: It's asking, what character traits do you most want people to experience from you in high stakes moments? Chad GPT, to answer question three, what I want people to remember about me in high stakes moments is what I call H2O. I want them to remember that I am helpful, that I am humble, and that I am always optimistic.
Speaker: Helpful, humble, optimistic. Those are the things that I want people to remember about me. Those are the core values that I want to bleed through in everything, but especially in high stakes moments. I recommend answering just a few questions at a time and then saying them here. And it'll see it went and created a whole dictation of what I just said.
Speaker: That way you can work through a number of different questions rapidly over 10 to 20 minutes. Yes, it does take some time. Create a time block to get this done sometime because once you answer these questions, ChatGPT will remember all of this.
Speaker: and be able to use it for you later. In fact, before each before you commit it each time, make sure to say, remember this, and it will store it in that memory feature. The third way you can upgrade ChatGPT instantly is just to start using projects.
Speaker: Yes, projects. It's over on the left-hand sidebar, and they are these little folders. If you don't have any, it'll just say new project. You want to create a project for every area of your work or different areas of your life, and there's one major setting when you set them up that you have to remember. in the very first part, because you only get the one chance to change this. If you don't do it in the beginning, you can't change this later.
Speaker: So if you want to create a project, give it a name. Maybe you call, I'm going make one for marriage, and I'm just going to have marriage conversations in here about how to improve my own. And you can say here, you could say by default, ChatGPT can reference other memories or other conversations within your ChatGPT account and vice versa. So if you're a conversation, a different project or in a vanilla ChatGPT conversation, it could reference conversations you've had in this folder or this project. One's about marriage.
Speaker: But... If you don't want it to reference things outside and you don't want other conversations to reference things in this project, you want to select project only. That way, all the conversations in this project, it's like you're having them within a new Chachaputi account that doesn't remember everything else you've talked about.
Speaker: That's golden, especially if you're an agency or you're freelancer working with clients, you can keep all your client content separate. But it's helpful for lots of different things like marriage, for example. If I want to just be able to talk about marriage, that's like maybe I don't want all my work life seeping into that folder, right? So I can create the project.
Speaker: And then I can start having conversations about this topic within just this one project. And once you get a project going, you can see all the different projects like I have for this podcast will just be located into this.
Speaker: It's great because it keeps your ChatGPT account way more organized, but it's even more important because ChatGPT now can group your conversations together and better reference the memory between them so that you can be more efficient and effective as you have more conversations trying to push an area of your life or an area of your work forward.
Speaker: I have more content coming out about projects soon, so stay tuned for the next video I drop on projects.