Use This Prompt Engineering Tool To Get 10x Better Responses on ChatGPT

Okay, the difference between your AI results and all of the experts out there is actually in the prompting. We’re going to turn you into an expert AI prompter in 30 minutes or less. But better yet, you won’t even have to do the prompting — because we’re going to give you an AI assistant to do it for you, so you can have expert-level promptings without ever having to write something yourself.

All of that on today’s show, Marketing Against the Grain.

So let’s dive straight into one prompt that I think maybe our listeners have seen. I think it’s one of the better templates. And then we’re going to give away some pretty interesting things for our subscribers to YouTube exclusively, or for our subscribers to the RSS exclusively.

This is a well-shared prompt I’m showing — for those not watching along on YouTube. And this is actually from Greg Brockman, the co-founder of OpenAI. He was talking about the fact that he still sees a lot of people cannot prompt, and that is why their results are much worse than other people when using these LLM models.

What he did is he provided this really easy-to-understand prompt, and then broke it apart into its core parts. I thought we could go through this. And then I want to show a quick hack that people can use to create their own prompt engineer from this template that not many people do.

This has a couple of core components, right? If you want to construct a really good prompt — and this says “for the o1 models,” but I think this is just good prompt hygiene in general.

First: you want to have a clear goal. In this case, he has a clear goal around hikes in San Francisco.

I will say, anytime I see “hikes” and Americans say “hikes,” I’ve realized in Europe this means going up mountains. I think in America this means going to the shops on the flat surface. It’s not the same thing. I would say a hike in America is walking on a trail of some type.

Yeah — a trail, but it’s not mountain climbing.

In some cases it is, in some cases it isn’t, right? It’s the very wide definition of hiking in America.

So he has a goal — a clear goal. You state your goal. You really want to understand what your outcome is.

Then he has a return format. Now I’m going to get into a much more detailed version of this, and then I’m going to get into a pretty awesome giveaway for a five-page prompt that our listeners can just go take and start doing some cool stuff with.

Return format: what is the format I want this to be returned in?

Warnings: the warnings that he provided here were things like be careful to make sure the name of the trail is correct, that it actually exists. I think these are things you want to point it towards — and that the time is correct.

So, examples would be: make sure, if you are doing any kind of prompting to research something, you would say, “Hey, make sure the name of that company is real. Make sure they have actual employees,” because you don’t want it to hallucinate — or whatever kind of warnings you want to create for the model.

Well, Kieran, I think this is a simple example of AI might find something in one source that may not be that accurate, right? And what it’s basically telling the AI is: check multiple times to make sure this thing’s real.

Right, right, right — exactly.

One of the things I do internally when I’m prompting — I use this prompt all the time — but again, I’m going to give you an easy way to use this so you never really have to remember the format.

Even though I’m giving you the format here, I’m going to show you a way where you never have to actually remember it.

One of the things I do when I use this internally for internal projects — and I give it a ton of context, which we’re going to go into now — I’ll always say: cite the documentation and the exact place where you got the information to make this statement.

So when I get the reply from the LLM, it always cites all of its sources and reasons why it made the assumptions, or gave the output that it did. So I can see exactly it’s getting it from documents that actually exist.

And then the most important thing that people still don’t get: context.

I don’t think people think context is like a sentence.

Even here, the context that Greg gives in the example — it’s like a couple of paragraphs, right? What we did on the landing pages — if you recall, in our landing page episode (and if you haven’t watched that, you should go watch that) — the context we provided to create the world’s best landing page was actually a 30-page style guide, right?

This is the beauty of these LLMs: they have a huge ability for you to add context. You can add a ton of things to the context window and really tailor those results.

So to kick off this episode: if you’re stuck with prompting in general, this is a great first place to start, right?

Do I have a goal? Have I stated my return format? Have I given some sort of warnings to say these are the things I really need you to do? And have I given it good amounts of context?

That’s a good starting point for this episode that will make people much, much better at prompting if you haven’t been doing this to date.

So. We don’t remember all of that because we’re lazy.

One of the things I do — and I’m going to share my screen here — is I’m constantly taking prompts and turning them into templates, and then creating custom GPTs to replicate that template for me.

What do I mean by this?

If you’re following along on YouTube, what you’ll see is I’ve gone into “My GPTs.” I’ll actually go into the whole process.

I go to My GPTs, I go in here, and we’re going to get into a couple of more advanced prompts in this episode.

I really love this one that I’ve created, which is taking a deep research prompt and creating a YouTube video.

But for now, we’re going to get into this one.

This is a prompting… get asked for all of these, just so we know — yeah — because I love our viewers, I am going to give away the deep research to YouTube video, which I think is actually incredible.

I’m going to show you something even better: deep research marketing — taking a marketing tactic and being able to templatize it into something where you can have an AI replicate that marketing tactic for you.

I am not yet going to give away this prompt, but I’m going to show you the results and what we may do in the future.

If you’re a listener, I think we’re going to figure out how to give our listeners some exclusive stuff at some point. But you want to hit subscribe because we’re going to try to do some subscriber exclusives.

This YouTube one is dope. I was like, maybe I should keep this one for myself, because it’s got a style guide for YouTube video that is actually insane.

All right — so I’ll go.

What I’ve done here is I’ll go into this prompt engineer o1. So this is my prompt engineer for creating o1 prompts. I can just go in here, I’ll show you what it looks like.

It’s pretty simplistic, right? Basically what it does is it provides a little bit of context for what I want, and then it tells it to follow the exact template.

Anytime I need a prompt, what I find really easy is I can come in here and I can say:

“Okay, I’m looking to create an incredible Twitter thread on the latest Google… you don’t have to worry about spelling… announcements. Create me a prompt to research those announcements that marketers would love and create an actionable thread on how they can use the AI to be way more productive.”

Obviously, you have an actual, real conversation here, go back and forth. Then you say, “Cool — that’s a cool conversation. Create me a prompt for this.”

So you’re not even asking it to write a prompt, right? Because it has all the instructions.

What is it going to disrupt?

Give me something that’s going to disrupt.

Anything software?

Yeah — it’s going to disrupt the sales industry.

Okay.

“I’m going to provide both internal and external data to analyze as part of the memo.”

All right — so that’s two sentences, very basic, very straightforward.

It’s taking the template that you just showed everybody — with the goal, the format, the warnings — and it is giving you a prompt.

It’s basically making sure you follow that template every single time.

The outcome is a strategic memo that pitches how we need to reformat our sales team. The return format should be a two-page memo detailing the core changes. You should only make it applicable to software sales. Use your own context.

I would say: you would actually provide a ton of context. It should give you the prompt exactly what you want it to do.

Mine is formal. I never… I always provide a lot more information. I give it a ton of information.

“Since you mentioned you’d be writing to new brand data. Do you want to model contextual grounding?”

Yes — assume AI has access to the data.

All right — so it’s going to do it now.

And it’s like a prompt engineer: it will make sure that you give it all the things it needs, which I think is a good checkpoint for people.

I use this all the time. I use this to create all of my prompts for o1 models. I provide it with a lot of information up front, and then it will craft a pretty good prompt.

And you know what’s actually pretty interesting: you can make this prompt a lot more detailed. You can say, “I want it to be very advanced,” so you can actually play around and it will improve upon that.

Now, the other thing: one of the things I would get into the habit of doing is — I believe this about all AI — I would create a first version yourself to get used to prompting. I don’t think you want to rely solely on AIs.

But if you are a novice at prompting, doing what I just showed you would take you up to above good pretty rapidly. Your results are just way, way better.

If you ask it to see how much more advanced it is — if you actually play around with it and ask it to continue to edit it — I’ll show you one of the wild things.

What we’ve done so far is we’ve taken a single template, created a custom GPT to be your prompt engineer, to replicate that template whenever you want.

But the interesting thing is, if you go back and forth between models…

You did a whole video on Gemini 2.5. I realized — I was doing this this morning in prep for the show — if you go between the models and say: “For an advanced reasoning model, how can I improve this?”

Even if you don’t say “o1,” you can just say “an advanced reasoning model,” if you want to be inclusive for any of the reasoning models, which is a good tip.

Then you can put in your prompt, and it will give you a much more detailed version of that.

What I would do is go through the prompt and say: cool, I like these parts, I don’t like these parts, I want these improvements.

But if you flip-flop back and forth in the models, they can edit on each other’s work.

So that’s our first standard o1 prompt. You can enhance — wow — way, way more detailed. It’s giving you… Gemini 2.5 Pro is really good. It’s really good.

I’m going to get into all of the other prompts I’m going to show for deep research. They’re done this way. It’s a combination of two models.

Let’s quickly show the same format — that template Greg created — but we’ll do it for deep research.

If you want to maximize your results on deep research, one of the things is: context really, really matters.

An example would be for HubSpot: whenever I’m doing something for HubSpot, I give it full details on my role, the company size, the industry we’re in. I give it information about who our customers are. I give it information about our tech stack, if that’s applicable. I give it a lot of context: what are the key goals and challenges.

What I’ve found is if you set the context about who you are, the company you’re in, and all of the things that matter to the research you’re about to do, the results are way, way better.

The blurb here is obviously very simplistic. It would be more detailed, but: “I’m a content manager in a 200-person SaaS company. We use HubSpot, Jasper, all day. We’re evaluating new tools to scale our content and grow our audience faster.”

So: you set context.

You give a clear assignment. You want to clearly state what research you want done and what kind of answer would be most useful.

You want to go through what your objectives and key metrics are. I’ll go through this because we’re going to give you this one that we can just give away — it’s out there already. I give it away already.

Scope and priorities: you want to set your scope and priorities. After we set objectives and key metrics, then: okay, this is the scope of the research, this is what I’m trying to focus on.

Then deliverable format — and this is what I want to spend the rest of this episode on.

I want to riff on this part because deliverable format is pretty interesting.

You can create an internal report. You can create a template. What I’m going to show you is go from the research into a YouTube episode.

The five core components of a deep research prompt: set the context, make sure you have clarity on the assignment, set objectives and key metrics, set scope and priorities, and then set deliverable format.

What I want to help people understand is that last step — the deliverable format — is so important.

Because I think what most people I talk to do is: they do deep research and say, “Hey, I want a bunch of information on this thing.” They get it. Then they do something with it. Often they put it in an LLM with a different prompt.

You’re way better off telling deep research: “I need this formatted to go with this prompt that I’m planning to use in this LLM.” It will package everything up so you get a far better result, which is what you’re going to show.

Yeah — that’s why that’s so important.

So what we have here is another GPT trained to reproduce that prompt, right?

This is again a simplistic way. I’m going to say:

“I want to create a report on what AI SDR tools can improve my sales outreach open rates…”

It’s probably going to do the same thing.

Okay — so we’ve given it a little bit of information. Obviously you would give it a lot more. Again, you’re going to get the guide here.

All you need to do is create your own custom GPT, give it the prompt template we’re going to give you. Anytime you want a deep research prompt to create a detailed report on anything, you could do that.

So we’ve given it some context. We’ve said what industry we’re in, who we’re targeting — pretty basic. It should create a deep research prompt here to create your entire…

One of the things I love: all the emojis.

The emoji — yeah.

I know it’s really fake. I don’t know why. It doesn’t do anything. But man, I like the emojis. Brings a bit of color and life.

It does. It makes it feel like less — just staring at black and white all day.

So it’s going to craft you the whole…

Wow — again, the size of the prompt.

You can go from beginner-level deep research prompter to pretty expert-level pretty quickly by creating these custom GPTs that are trained on a prompt that really works.

You can copy and paste this, kick off your report, and it will bring you back a detailed report on AI tools that can help increase your cold email open rates by 10%.

We’ve given it some scope and priorities. We use — it has to be a HubSpot integration. It includes case studies or performance benchmarks where available. It would do a lot of cool stuff.

Honestly, the level of report you can get — in most companies, you can probably use this and send it internally and no one would know it was not from you.

That’s basically what this is showing you.

We’ll kick this off to run in the background so we can show this, because it won’t take that long.

There’s a concept here: going from research to output, right? You’re doing detailed research on something, and it’s creating something for you.

The next one I want to show is this YouTube one.

You haven’t seen this one yet.

No, you haven’t seen this one yet.

There’s an important concept I want to…

All right — so that was a report, but I suspect marketers are like: okay, what marketing-type things can we do?

This one — I’m going to start with the different phases.

It does the same sort of thing. It says context: you’re a research, analysis, and narrative strategist — especially a creative video script writer. You’re expert at a lot. You’re like five different people.

In this case, you’re right.

Your mission is to perform deep, multifaceted research on a topic provided, and then transform that research into a world-class YouTube video outline designed for viewer engagement, retention, shareability, and impact.

I wouldn’t actually put retention in there, if I’m being honest. The model doesn’t care about retention. I care. But viewer engagement…

Then it says: the user will provide a topic, an audience, a video length. Then it will go run research.

It will do the research scope, look at the core narrative, define key takeaways, pinpoint engagement elements.

This is really cool: it looks for surprising statistics, counter-intuitive findings, relatable analogies, ethical dilemmas, quotes.

It’s trying to find all of the things in this topic that would make it a good YouTube video.

Then it says: okay, I have those things. Now I’m going to convert it into this outline.

This outline is a very detailed YouTube outline.

Kieran — this was done using the thing I usually do: taking videos that have worked well, taking the transcripts, adding them into context, then working with AI to distill it into a style.

So this is like a YouTube kind of style outline. I did it in Gemini and ChatGPT. I’ve been switching between both models.

It works way better between two or three models. That’s why you kind of need — if you’re just a one-model shop — you kind of need more than one.

Yeah — and I haven’t been using Claude as much. I feel bad about it. I was the number one super fan of the world. I don’t know what’s happened to me. The pace of innovation there is a little slow right now.

Yeah. I think Google has come out with some incredible things. ChatGPT continues to come out with cool things. Google has come out with so much stuff recently that nobody knows about.

We’re going to do a series of shows coming up with some of the cool stuff. The AI in Google Sheets is incredible now.

We’d have to do one show per release.

We’ll do an overview and then a couple of big ones. We’ll do a quick run-through and marketing use case.

So it’s broken into a hook and an intro. It says you have to set up the premise and the promise, clearly state the topic and the question.

It goes through how you create engagement right up front.

It sets the stage. Context is key. Unpacks the research. The so-what impact and implications. Critical perspective.

Then it has these discoverability and polish elements: compelling title suggestions, the video description, keywords and tags. It goes into sound and production, and then sources.

So that one’s running.

We actually want to create a video for YouTube around Google. So let’s do it.

“I want a prompt to create a YouTube video for Google’s latest AI launches from Next 25.”

It will ask me those three questions.

So I’m in my custom GPT — trained to reproduce prompts in the style I showed you.

We go in and say: this is a real video Kipp and I want to do.

Topic: Google’s latest AI launches. Who’s the target audience? Marketers, startup founders. Ideal video length: 30 minutes.

Now it should give me the detailed prompt.

Okay — we might have cut this — because it’s searching the web.

No, fine.

Now it’s going to give me the detail prompt, and we’re going to run this one in Gemini deep research.

You can see it’s given the exact time. It gives the hook with the exact time length. It gives the exact time length of the intro. It gives the exact time to set the context.

This is actually… start using this.

It gives the exact time to do all the core insights.

It ends with five minutes at the end where you talk about the impact and implications — why does this matter — and you provide the nuance and credible critical perspective.

Then you have a call to action, an engagement loop, and an end screen.

It’s given an exact breakdown mapped to the 30 minutes, which is pretty incredible.

Let’s run this prompt in Google, and Kipp and I can probably use this and do a show and see if it’s good.

This is a little bit of… I always wanted — if I was doing a cooking program — this is kind of how it feels. This is the one I did earlier.

So this is what it produced: the Google Next 25 using the deep research prompt.

It provided an entire… it started with a pretty good hook:

“50,000 images a month. Eight weeks of work done in eight hours. Google’s Next 25 conference unleashed a tsunami of AI innovations that promise to supercharge marketing…”

Did Google just hand marketers a creative superpower? Are we looking at the beginning of the end for traditional marketing as we know it?

Sound dramatic — but I mean, I love the… it’s a pretty great opener.

Then it goes into premise and promise: a tsunami of AI hits marketing. It gives an overview.

Hooked yet? Buckle up, marketers.

Then it pulls out quotes: “AI will drive a transformation of advertising in 2025 even more significant than the mobile revolution.”

It’s pulled out from Twitter, actually.

Did you know it could pull stuff from Twitter?

I didn’t know.

That’s pretty wild.

It’s pulling quotes from Twitter as well as pulling quotes from all over the place — news, search, Twitter.

There’s a part of me that thinks we shouldn’t give this away. This is actually incredible.

Oh, you’re changing your mind on us.

Google’s AI superpower. Pretty amazing.

We could condense this down and make a show out of this. This is some of the best research I’ve seen.

My one note is it’s even got… it’s included things like sound, keyboard type, and “ask user to taste a task.”

It’s thought of all the ways to integrate sound into this episode.

It’s got tons of citations for quotes we can use.

I’m really curious about the Twitter quote.

The only thing I would want to play around with is: remember when we give it warnings to say explicitly what we don’t want — the thing I would definitely ask it to do is map to the outline more. And map a little bit more to marketing use cases.

Make sure it includes the section title so we can follow along. Because it’s using things called “chapter,” which I never said to use.

“Can you reframe this into the following outline format?”

Okay — so that’s basically…

Again, don’t think this is: “Oh cool, I’ve got a prompt now that I can create any type of outline for a pretty incredible YouTube video.”

Don’t stop there. You can use deep research and swap out the YouTube outline for any outline: a blog post outline, a Twitter thread outline.

The thing to learn here is going from deep research analysis to deep research combined with a template — any output you want.

Okay.

The last one I’ll show — just to get people’s thoughts racing — is this one takes any marketing tactic and creates a detailed style or template execution guide for an AI assistant to replicate that.

What it does is distill exactly how to recreate this, provide prompts for the AI to use, and the AI can replicate any marketing tactic you give it.

I think this is a great use of deep research, because people are thinking of deep research as a thing to create a report analysis.

I think of it as a way to get very specific on things you can execute on — take action on.

That’s one of the takeaways from this episode.

So if we run through what we give you really quickly:

We give you this o1 prompt. We put that o1 prompt into a prompt engineer so you can recreate that anytime you want.

We give you a much more detailed deep research prompt. Then we put that deep research prompt into your own custom GPT so you can replicate that whenever you want.

Then we give you a version that you can build upon. I happened to do the YouTube video — a deep research prompt that creates a very detailed YouTube video.

We showcase that by running it for Google.

We did this by using it for Google’s Next 25 AI launch that we’re going to actually use the results from — we will create a show on.

But I think about today’s show: we, in literally 30 minutes, taught somebody how to be an expert prompt engineer.

We gave them all the tips, all the tricks.

We gave this bonus that you should never do dumb AI anymore. You can always have deep research provide powerful context to any problem you’re trying to solve with AI.

That is one of the power hacks that you and I use. It is very transformative. Nobody does it.

It’s something you’re going to continue to see on the show and something we’ve talked about in the past.

But the use of custom GPTs, the anatomy of an amazing prompt, and how to use those prompts to get really powerful deep research is a game changer for how everybody’s going to be using AI.

Yes — go forth and prompt.

Go forth and prompt. I love that.

We’ll see everybody real soon on Marketing Against the Grain.