Transcription
So, you have a business idea and some insight into the type of product you're trying to build. You think you can sell it to people interested in this thing, but the challenge is always getting that first version of the product out the door. This is essential for starting to run traffic to it, which leads to initial downloads and success.
My personal strategy is to spin up landing pages and create fake mockups of the product. I send traffic to these pages to get people to sign up. If I can't get sign-ups, I sunset the product, and we never actually build it. However, this is quickly changing with all the AI tools available that can create the first versions of the product based on the information you provide.
Today, we're going to discuss how to uncover product ideas and share how you envision that product should function and look in a recorded format. This allows you to create a transcript and transform it into a PRD, which stands for Product Requirements Document. You can then take this document to a product like Vzer to spin up the first version of the product. This is powerful because you can go from idea to a visual representation of that idea in literally under an hour.
When you combine these two elements—sending traffic to validate the idea and having a simple version of the product—you can quickly validate the idea or feature you're trying to build. If you want more content like this, subscribe to this channel and the email newsletter linked below in the description. Also, follow me on LinkedIn and Twitter; the links are down below.
This episode is brought to you by Talent Fiber. Talent Fiber helps you hire the best offshore talent for your company. It works differently than a traditional headhunting agency where you pay them 10 grand, and then they disappear. Talent Fiber is a business professional outsourcing company. They only get paid when that employee stays at your company. You just provide them a job description, and they're super easy to get started with. They handle all the HR and payments for those people, and they can attract the best talent because they offer the equivalent of a 401k or healthcare in the specific geography they're hiring from. This makes them a magnet for the best talent in that country.
The best part is, if that employee doesn't work out or quits, they'll replace that employee for you automatically and handle the whole process. We hire their offshore talent exclusively for all the companies I own. They are some of our best team members. I could not find people better than them in the U.S., and we're paying one-third of the cost. Go to talentfiber.com to learn more.
Before we get started with today's show, I'm hosting my friend Jay, who runs an AI product development studio. He's going to share his entire workflow for building out the first version of these products. His company's name is Casper Studios; go check it out. There's a link in the description.
With that context, let's get started with today's show.
Jay, what's good, man? How are you?
Cody, how's it going?
I'm good, man. How's your week been?
It's ending in the best way; I get to hang out with you.
I don't know about that; I'm probably way too geeked out for this. Anyway, today, I'm pretty hyped about this. We're going to do a session from an idea to what the first version of a product could look like, covering the whole process.
For context, Jay runs a product studio that builds AI products all day long. They're unbelievable at it and can take an idea to a concept in about 30 days. It just depends on the client and what they're doing, but their whole thing is AI-first with all the development.
Today, we're going to go through how to find this idea, identify a seam, and develop a business concept. The idea we're discussing today is super half-baked, and Jay's going to help flesh that out and make it a reality.
Before we jump in, I'm curious—why is this becoming the new standard? You've been in product for a while; that's your career. Why is there such a big opportunity with these tools like Vzer?
Excellent question. I'm grateful to be here and for you, Cody, for always inspiring a lot of founders and builders to think outside the box and try different things. I'm definitely a student of your work, and it's impacted my business and life. Thank you for doing what you're doing; it helps a lot of people out.
I love it, man. This is why we're here.
So, I'm curious to hear from your side as a founder. If we can get something out the door after a customer conversation, take that recording, and turn it into a visualization we can show the client within an hour, I see the value in that. But I want to hear from the product side—why is this such a pivotal moment in product history?
Most of my product background and learning came from my time at LinkedIn, where I was for five years. I remember when a product manager or a non-technical person would write up a document to make a business case for something, like launching a new feature or entering a new market. Half of the job was just getting people on the same page and communicating ideas effectively.
The struggle was that once we had a nice, beautiful document laying out what would go into a product requirements document, there was still something missing—a beautiful visual to tie it all together. Visuals and designs really make people move and act.
I've experienced this myself. One idea was a product we shipped—a Chrome extension called Casper. I remember thinking about it with a couple of friends and putting it into a one-pager that became a product requirements document. I sent it to a designer friend, and four days later, he had come back with a beautiful prototype showing exactly how everything would work. It went from text to actual Figma, and I was like, "Holy crap, I need this in the world now."
There's a magic moment when you can interact with something visually, especially if it feels like it will solve a problem you have. That makes you want to act. That's how we've designed our process at Casper Studios. We design and build AI products, take those learnings, and share them on social media to attract more clients. Design is how we win, and I'm always keeping up with the best design tools to reduce the time it takes to go from a conversation to that magic moment. I think Vzer is definitely helping with that.
That's fascinating, man. I love it. With that context, let's dive into what we're doing today.
Let me screen share, and I can walk through the opportunity and what we'll be working on.
I remember sending this to you early on when I first saw it because I thought it was ridiculous. I'm in a bunch of group chats with other founder friends, and I'm always tossing these ideas around.
I found this search term, "AI infographic generator," which is going bananas. I ended up buying the exact match domain, aiinfographicgenerator.com, for about 10 bucks.
This is the whole concept: the high-level product and how it fits together. Marketers use infographics to enhance blog posts. Typically, someone writes a blog post, then creates an infographic to embed in it. This serves as a link magnet, attracting other websites to scrape the infographic and include it in their content. It also increases dwell time on the blog post when there's something visual at the top of the page to hook people.
Traditionally, creating infographics involves going to something like Canva or Figma, pulling content from the blog post, and designing it in a way that relates to the brand.
Looking at this, I immediately saw a massive opportunity. The high-level idea is to take a blog post, upload it into this tool, and have it output graphics. It could generate five different graphics based on insights from the blog post, using a flat design language or specific color scheme related to the brand.
I envision the first version having five to ten templates in different design styles. You upload the blog post, and it generates each of those templates in those styles. A person could then download them as a PDF or PNG to embed on their site. That's the MVP and where I see the initial wedge.
I'm happy to answer any questions about the opportunity and curious to hear your thoughts or discovery questions.
I love it. Do you mind stopping the screen share? I think that might be beautiful.
I was in Cody Inception, which is usually awesome, but in this case, it was distracting me.
This is great. A lot of founders come to us with maybe not as much validation as you have with all the search terms. I get texts from Cody every other day about another one of these charts going up. I'm glad you come in with a lot of research.
A lot of times, people are just saying they are the user wanting the problem solved. Maybe we can have a quick mock discovery session. We try to go deep in terms of asking the right questions to understand the user's problems. This becomes a product requirements document, which we then send to a designer or an AI design tool to create the first version of the mockup or designs.
I'll pretend you're a customer of mine, and we'll go through the discovery session quickly, which could help founders in the future.
That was great; I appreciate the background. Do you mind highlighting a bit more about who you are, what you do, and why an infographic is important to you?
Totally. I would classify myself as a digital marketer. Below that, I might be an SEO specialist. Their job is to create the best, thickest content possible, providing as much value as they can within that piece of content. An infographic is just one of those tools in their belt, along with videos, photos, and well-structured content.
Typically, anyone doing some type of marketing is why they create infographics. It's an easy way to take complex information and make it digestible, hitting key insights and bullet points in a visually appealing way.
When someone searches a query and lands on a page, if they see a wall of text, it looks like a job to go through. They would include the infographic higher up on the page to break up that wall of text, providing something more interesting to the eye than just written words.
The user is likely to be an SEO or digital marketing specialist working in a larger organization or someone running a niche site or content blog trying to rank for specific keywords.
Why is it important for the content to be visually appealing? What's the value of that?
People's dwell time increases when you have visuals. A simple way to test this is to take a blog post without an image in the top fold and do an A/B test with an image versus no image. You'll see that people have a higher dwell time when there's something visual at the top.
Infographics are one of the oldest types of content on the internet, and people are used to seeing them. From a visualization standpoint, the visual introduces people to the brand style.
If you think about a wall of text versus images and graphics scattered throughout, it creates variability that the reader enjoys more. We see this in the dwell time on the page, and the higher dwell time increases the likelihood of ranking for the target keyword.
If it ranks for that keyword, it drives more conversions to the website.
So, if I'm hearing correctly, having visually appealing content in the form of an infographic helps people stay on the website longer. If they do that, they have a higher probability of converting, which drives revenue. Is that accurate?
Exactly. It depends on the type of blog they're running. If it's a niche site, they're trying to get you to click on ads. If it's a company selling something, they're trying to get you to sign up for their product.
I see this mistake by founders often. They say they get a ton of traffic but no conversions. I immediately think they have zero calls to action throughout the page. We'll implement that for them, and their posts go from 0.1% sign-ups to 5-10% overnight.
Infographics serve the purpose of getting people to engage more deeply with the content. They can also be used as a backlink tool. People build infographics, email them out, and ask others to include them in their content, which can lead to backlinks.
I'm going to ask questions quickly to understand your current workflow before using AI, and then I'll come back to understand why you want to build this product.
What are you doing today to create an infographic?
Typically, we have a template. The process starts with writing a blog post, copying the content, and putting it into something like ChatGPT to generate five insights. We then take those insights and create a graphic for each one in Canva. This process takes about 15-20 minutes to get everything into Canva and out the door.
The dream is to drop in the blog post URL and have it automatically generate the infographic. I could save 20 minutes of time if I can do three of those in an hour.
When you think about that at scale, if you write a thousand blog posts and spend 15 minutes per post, that's 15,000 minutes or 250 hours to create infographics for all of them. Even if I pay an offshore person $10 an hour, that's still $2,500. If I can do that for $250, that's the value.
You're saying this problem of taking too much time to create an infographic will only increase as AI-generated content increases. More blog posts will be created, and the cost of production will go to zero.
Exactly. People probably wouldn't do this for all their posts because it's a massive time suck. They might only do it for their top-ranking posts. If you can do it for all of them, it becomes a no-brainer.
There are deeper workflows where you could upload a list of blog post URLs in a CSV, generate an infographic for each, and automatically publish them to each blog post. The MVP could be as simple as uploading one URL to generate an infographic, saving 20 minutes of time.
A lot of this will likely fall into niche site owners or those running content blogs. I've seen people use this to create Pinterest pins to drive traffic back to their sites.
I have a good sense of the problem statement and the core jobs you're trying to accomplish as a digital marketer or small website owner. I can start to think about ideas for using AI to solve this problem.
Before we move on, why do you want to build a product like this? What about this idea excites you? What does success look like for you?
I like money, Jay. But seriously, there's a common problem that people face, and if we can flip this into a product, it could be a massive time saver.
Since I own the domain aiinfographicgenerator.com, it's an exact match keyword domain. Those domains typically rank easily for those keywords, even if they're competitive.
We can throw not a lot of backlinks at that site and get it to page one with just ten backlinks. I've done this in the past with long-tail products, and we can get as much inbound signup flow as we want.
This solves the customer acquisition problem, and then we just need to figure out the product. This is another tool to add to our portfolio.
For those running consulting businesses or speaking to clients, it's important to do the product requirements early in the discovery phase. If we had more time, we would go deeper to understand everything, user flows, and ideate on what could exist.
To make a sale, you want to get to the personal pain as fast as possible. A lot of people start with intellectual pain, but through curiosity and listening, you want to get to the personal pain.
For example, a client might want to disrupt an industry because they missed the last wave of technology and want to take advantage of this one.
For me, my personal pain is wanting to partner with Cody to see what we can build together. There's a revenue opportunity because it's one of the up-and-right search terms Cody keeps sending me.
It's important for a founder, salesperson, or therapist to listen and ask questions. This grounds the relationship and helps you build a better product.
I'll pause there, and we can keep moving.
I think that's great. For context, the ideas can be dumb. The power of these new AI tools is that you can get a dumb idea out the door, and it can evolve into something valuable.
Every company I've built started as a dumb idea that evolved into something people pay for.
I want to emphasize that many people have an idea of what they want people to buy rather than what people actually want to buy. This is crucial to think about.
A business is about having something people want to buy and being able to build it. If you can do those two things, that's a company.
The best companies build and sell things that people want. The market decides; we can't influence it.
Coca-Cola can't get people to buy more Coca-Cola, even though they want to. They had to acquire other companies because they couldn't convince people to change their behavior.
Don't think you can influence the market. The AI infographic generator is still making infographics, but it's doing it in an easier way.
This should be the filter when you build products. Focus on the discovery phase and understanding the problem.
Cody, you've built the right experiences to know if a search term is increasing, then we should build this. That's research too.
You can do search term research, speak with people, or do a competitive landscape. Just make sure you do research, however you choose to do it.
Let's move on to the next step of the workflow to get to Vzer.
I'll walk through what that looks like.
For the audience, we just recorded this whole thing. Jay took the transcript and is now going to use it to build out the initial information for the product.
Exactly. After a conversation like this, I would record it with Fireflies or Otter. I would then copy the transcript and put it into ChatGPT. Some of the notes I take live will prompt me to know what to ask ChatGPT to pull out the right insights from the transcript.
I can share a mock PRD to show you the right structure.
Can you explain what a PRD is for those who don't know?
Sure! At LinkedIn, I was part of leading the verification team. We launched a verification feature that now has over 50 million members verified.
That whole interaction started from a document called a PRD, or Product Requirements Document. It could be used interchangeably with a strategy document.
Think of it as having the context of the problem you're trying to solve. It walks through the key jobs to be done for a user.
For you, Cody, with the infographic use case, the job to be done is to create high-quality images from text faster.
A PRD is a collection of your thinking and research in a document. Without it, it's hard to get people on the same page.
You need one source of truth that everyone can review and comment on before moving on to the next step. Writing it out, even with AI, helps clarify your thinking.
For the verification workstream, we spent weeks in the discovery phase, understanding the problem, what competitors were doing, and filtering that into the document.
ChatGPT can help create something that leads to Vzer. You could even put that into Cursor or any other no-code AI platform. It helps the model know what you want to build and why.
That makes total sense. Thanks for the context for the audience to understand where this fits in the workflow. It's the written version of what the product will do before any code is written.
I didn't think about it as a buy-in tool between all the parties internally.
For you guys, it's like, "Here's this document. Does this look right to you from a client standpoint?"
Exactly. The distinct parts of our workflow are deliverables we share with clients: a product requirements document, our designs, and the actual build.
The first one ensures everyone is on the same page before we keep going. If that one's off, you're going in the wrong direction.
If you have that source of truth, you can keep pointing back to it. If you need to change something, you can, but that's why we're building this.
I'll have you screen share, and we'll jump into it.
Sure! I want to show you what a PRD could look like. Before the call, I went through and created a mock PRD.
Here's the objective: design and develop an AI-powered infographic generator tailored for marketing professionals and content creators.
It's important to capture the audience's attention, convey complex information, and signal key insights.
Here's a user persona. I don't know if you're actually 30, but I made you 27 for the vibes.
Then, I laid out the background and some of the problems and jobs to be done. For example, as a marketer, I want to quickly produce high-quality infographics to meet campaign deadlines.
As a non-designer, I need it to be easy. You can walk through what the jobs to be done are and what their life is like.
I forgot to put in that you're somewhat of a "dgen" living in Colorado, but that's okay.
You want to lay out your product principles. These principles are helpful because they will be the foundation of what kind of product is developed.
A good product principle should be so opinionated that the inverse could also be a great strategy.
For example, if we're building a voice AI bot, one principle could be "voice for everything." The alternative could be "text first," and that principle could also work.
These principles help when building because if there are disagreements, it's likely because one of the principles needs to be changed.
If we launched the voice AI app and no one is using it, the problem might be that we need to include text as an option.
This is helpful for thinking about product development and design.
Then, you go into proposed solutions based on ideation. You can also include a competitive analysis of the infographic market.
Some competitors include Canva and Pictochart, which are doing interesting things from an AI perspective.
You don't need to reinvent the wheel; look at what's working, model after that, and avoid what isn't.
I'll pause there. Any questions based on this?
I think it's brilliant. You built this out, and now you're ready to go to Vzer. What does that look like?
Before Vzer, I would send this to my design team, who are great. However, I would have to wait about a week to get something in front of the client again.
I'm not validating if those designs align with the client's expectations, and I'm investing resources without a clear ROI.
When I found Vzer, I was excited because I could take this document, put it into Vzer, and get initial mockups to show the client immediately.
This allows me to get feedback and update it live before sending it to my designer for the final design.
I would go to Vzer, copy and paste the PRD, and say, "Make this into a mockup. I want a clickable prototype of this PRD."
Sometimes, I don't have to make a deeper prompt than that, and it just works.
It's crazy how much heavy lifting can be done on the front end.
Have you heard of CodeGuide?
No, show me after.
You can see how fast this operates.
I just met the founder, and what they do is prep the information you need for whatever programming IDE you're using.
If you wanted to take this and put it into Cursor, it makes all the documents necessary for whatever you're using.
It reduces the time and credits needed to build the product later.
This is super related to what we're discussing, and I thought it was fascinating.
You can see here, this is what the actual PRD ends up being based on the idea.
We take the PRD, Vzer does ideation, and we take the best ideas or concepts into design.
From there, we go to our developers and engineers to build out all the functionality and components.
Exactly. The world of design is changing rapidly. If you're not using these tools, we don't really work with you.
We're advocating for it aggressively because the velocity you can move at is so much faster now.
I just saw that Gumloop raised 17 million in Series A with just three people.
There are a couple of these stories coming out, and I think that's the new standard.
How are they doing that? They're leaning on tools like this.
Here's an example of a Vzer version and what happened afterward.
I was building a voice journal application and created a PRD for it.
This is the original PRD, and I took it to Vzer, which created these designs.
I had voice, text, and photo options, and I was able to prompt it to get to a good spot.
I sent this to my designer, and he came back with the final design.
It shows that if you want to stand out in design, Vzer is not there yet.
It helps you get started, but you need great designers to take it to the next level.
The functionality is similar, but the final product is different.
If you just ship designs created by Vzer, it might work, but it won't stand out.
Vzer is the first version, but you have to expand on it.
There are tools where you can take the Vzer version, screenshot it, and ask for a design in a specific aesthetic.
You can go back to Vzer and improve it.
Once the designs are done, you can go back to Vzer and edit it.
The world of design is changing rapidly. You should definitely be using these tools to spin up the first versions of your product.
Over time, the capabilities of these models will only increase. A non-technical person should be able to use these tools to spin up everything.
We're actively thinking about client needs and what they require.
The tools are coming out every other day, so we have to change our service offerings and packaging.
What a time to be alive!
It's crazy, man. I appreciate your time as always. If people want to find you, where's the best place to reach out?
It's actually going to be LinkedIn. That's the spot for me.
I'll put it in the show notes.
Thank you again, brother. We'll talk to you soon!