How to build SaaS App in a weekend using AI (Steal This)

Transcript

00:00

Today we're going to talk about how to use AI to build a SAS in a weekend. I came across this Reddit post by this guy or girl, Lord 007 TN.

It had 27 upvotes

00:15

and I was reading it and I was like, "This is absolutely gold." He outlines exactly how to build SAS in a weekend. Now, there's some things I would do differently, some things um that I really like about this.

Um, and if you stick around to the whole episode, you're going to learn a playbook for how

00:31

to build a SAS in a weekend using AI. We're going to go through a bunch of the steps.

I'm going to show you how to do this. Now, I don't know if that SAS does a million a month, $100,000 a month, $1,000 a month.

I don't know. But I do know that uh there's a huge opportunity right now to be building software um SAS

00:49

businesses. Uh AI makes it a lot easier.

And um what we want to do is basically have the highest probability of success when we launch a product. So let's talk about that.

[Music]

01:09

So uh he says the first you got to come up with something to build. Now I kind of disagree with him on on step one.

I feel like there's step zero which is you got to figure out an audience. You got to figure out which community you're going after.

You got to figure out a

01:25

niche. So, I actually think there's a step zero around thinking about what niche makes most sense.

Now, how do you do that? One is you can say, well, what is an unfair advantage that I have?

You know, maybe I was in nursing for 20 years and understand nurses better than

01:41

anyone. Um, maybe I'm a developer and understand, you know, a specific programming language better than uh someone else.

Or maybe there's just a trend that I'm seeing. Um, and I see this trend.

I I I you know, I I I know

01:56

where it's going to go from here. I have a good and I think that there's a business idea here.

So, you go and start a social account. X, LinkedIn, Instagram, Tik Tok.

Um, you go and you go do that. That's step zero.

You start actually figuring out two things actually. You got to figure out what is

02:12

the format of the type of content I want to create. Meaning, am I telling stories?

Am I showing images? Is it videos?

What is the format? And what is the system so that I'm consistently putting out content every day?

Step step zero, build an audience. Then figure out

02:28

what something to build is. And actually, how you figure out uh how you how you build something is you start teasing things with your audience and they're going to tell you what to build.

Step two is I get to Gemini. I get Gemini to do some deep dives on who else is doing similar stuff.

I got to know

02:44

the competition, right? So, let's go and do this live.

And I'll be honest with you, I don't use Gemini a lot. Actually, you can see I literally have never used it on this account.

Um, so I've used it

03:00

on my personal Gmail, but I haven't used it on my on my business. But I I I don't know about you, but I'm more of a chat chat GBT code um Grock type of guy.

So let's go come up with an idea and uh see what Gemini says. So say

03:18

I have an audience that maybe is people who like books and through that I realize that goodreads.com which is now owned by Amazon is you know this community of people you know who are reviewing books

03:35

and you know when you look at this website it just feels old school. So maybe I'm like maybe there's an opportunity to reinvent Goodreads.

And I'm pretty sure Goodreads got acquired back in the day for a bunch of money.

03:52

Um yeah, Amazon acquired them, I think. Yeah, for $150 million.

So I'm like, okay, I want to make $150 million. And that was a long time ago.

Let's go and rebuild it. So let's go and ask Gemini who else is competing with

04:09

Goodreads. So, we're going to say, I want to build a startup that competes with Goodreads.

What is the competitive landscape look

04:25

like? And aka who else beyond Goodreads am I going to be competing with?

and let's go see. So, now I'm

04:41

learning about a bunch of I don't know about you, but I've never heard of any of these uh a bunch of competitors that I can go look into. So, I see the story graph.

Um this platform has gained significant traction as a popular alternative to Goodreads. It's datadriven personalized recommendations

04:57

based on your mood, themes you enjoy, library thing, bookworm. It doesn't make a difference what these things are.

The the bottom line is if you're building a SAS in a weekend, um you basically go and check and write notes as to who these

05:14

people are and what is interesting about them and what what maybe isn't interesting about them. Um and then it says other platforms with overlapping features.

Uh they're not direct Goodreads competitors, but they offer features your startup might consider. So like a book discovery platform like

05:30

BookBub, social reading apps like Book Ship. Again, I've never heard of any of these things.

Or Bookstagram, IG book community. That makes sense.

So, I'm getting a lot of this is like competitive analysis, Mackenzie analyst in, you know, in your pocket

05:45

type thing. So, this is really cool.

Then it says key considerations for your startup. Um, given the existing players, what unique value proposition will your startup offer?

will be specific niche genre focused academic reading a unique feature AI powered

06:01

recommendations or more nuanced asking the right questions right like I feel like this is probably the most important question and then you know user experience look many goodreads users express frustration with this dated interface I'm not alone uh community building how will you foster

06:18

uh an engaged community you get the point okay so that's step quick break in the pod to tell you a little bit about Startup Empire. So, Startup Empire is my private membership where it's a bunch of people like me, like you who want to

06:36

build out their startup ideas. Now, they're looking for content to help accelerate that.

They're looking for potential co-founders. They're looking for uh tutorials from people like me to come in and tell them, "How do you do email marketing?

How do you build an

06:51

audience? How do you go viral on Twitter?" all these different things.

That's exactly what Startup Empire is. And it's for people who want to start a startup but are looking for ideas or it's for people who have a startup but just they're not seeing the traction uh

07:07

that they need. So you can check out the link to startupmpire.co in the description.

That's uh step two and I think that makes a lot of sense. Step three, I check out what those competitors are offering and what makes them tick.

helps figure out my angle.

07:23

Yeah. So, that makes that makes a lot of sense.

You basically go through each and every one of these uh competitors. So, let's just say let's just go to the story graph for fun.

I even saw that in like the preview

07:40

text on Google that um they do challenges which is really cool. Challenge directory.

Um that's really cool. So, story graph because life's too short for a book you're not in the mood for.

Simple tracking and insightful

07:56

stats. This feels extremely nerdy.

Discover books by mood. I think that's really interesting, but like I don't know.

Are people looking for like a checkbox? Like I feel like there's so many more beautiful ways to show

08:11

showcase that. Um I saw this uh really beautiful um product design for this app called Q which is a way for people to um discover

08:27

videos like you know movies and TV shows to watch. Um and you know the co-founder says I'm proud of our visuals and designs in Q.

It's contributed to why millions of folks have downloaded and used our apps. Here's what three of my favorites.

So look at this just feels so

08:42

much nicer than what I just saw. um swiping with a partner to pick a movie and getting alerted with when you watch um our 3D trailer feed and the poster flying down to the bottom right when you add it to Q.

That's really cool. But I think the coolest part is the spinner.

08:59

Like having a spinner like this for books, I think would be way cooler than what the story graph is doing doing. So, um, yeah, I guess I guess like you just go through all the competitors, you write notes about what you like, you

09:15

don't like, you do some research, and then then I hit up uh code with my idea and the basics. I make it grill me with like 20 questions to see if the idea even holds water.

So, let's go to code now and ask it like 20 questions as it

09:33

says here. uh with my idea uh with my with with our idea here and see what they say, which is basically like a Q, you know, Q the Q app but focused on

09:50

like young people, Gen Z and something super beautiful. So I open up Clo here.

I use the um Mac app. So I say uh I guess the prompt would be something

10:05

like I want to build Q but for Goodreads. basically a more beautifully designed version of Goodreads with cool

10:23

features like this spinner from Q. I know that I probably need a niche to focus on.

So that's what Gemini said before, like you

10:39

need a niche. But I'm thinking Gen Z.

Is that too broad? I want you to grill me with 20 questions.

Let's see how they worded it

10:55

originally in that Reddit post. Grill me with 20 questions to see if the idea holds water.

To see if this idea even holds water. This is crazy.

11:11

I've actually never used AI like this. It feels so dumb, but it actually makes so much sense.

So now what's happening is it's coming up with a bunch of key questions to evaluate if the idea is any good or not. What specific pain points

11:26

with good reads are you solving beyond aesthetics? Have you researched how Gen Z specifically engages with reading and book discovery compared to other demographics?

What unique features beyond the spinner would differentiate? It's literally asking all the tough questions that actually make sense.

Wow,

11:41

that's really really good. Let's see.

Let's see what else uh Lord 007 says. Um it if it survives the interrogation, I tell Clo to write up a super basic onepage plan.

So, you know, if I was actually doing this idea, I I would have

11:58

to literally research all these things. So, this is the hard work about building a startup, building a SAS that does take time.

Um, and this might take me a full day just to actually go and figure out all this stuff out, but let's just pretend like this is

12:15

a good idea and pretend like I've gone and researched this. Um, and move on to the next step.

I'll say uh I'll ask it to to write up a super basic one-page plan. It's called a PRD.

Just get an outline, an outline. and they'll say,

12:31

"Can you write a super basic PRD based on what you think is the best answers to these questions and will result in a $150

12:47

million exit like Goodreads did to Amazon." And let's see, novel spin. Okay, so it's really leaning into this spinning idea.

And it's just writing out the executive

13:02

summary, the market opportunity, the target audience. It went with Gen Z with a secondary focus on millennial key points.

Okay. Discovery, the core features are the novel spin wheel, moodbased recommendations, and micro

13:17

genre exploration. I actually think that's pretty good.

Beyond basic genres to ultra specific niches that Gen Z identifies with with social features like reading challenges, book clubs, and social sharing. It gives you an uh some idea around what should the user

13:33

experience look like? Should it be minimalist?

Customizable shells. I actually think that's really cool and also that's really screenshot worthy, right?

Like I want to share that on other platforms. Um I think that's brilliant.

Um because that's going to generate that's going to basically get

13:48

uh my K factor aka uh you know my virality up. People are going to be sharing these like crazy.

Think of like Strava the you know the running app. People love sharing that stuff.

Wow. Okay.

So, it gives you launch

14:05

strategy, the success metrics, competitive advantage, and the path to $150 million exit. Okay.

I feel really good about that. I don't know if you do, but I do.

So, boom. Step six.

Now, I've got a simple outline. Back to code, but this time we're talking looks.

I get to

14:22

break the whole thing into small shippable chunks focusing only on the UI for each CL chunk. It tells me what each page will show, what you can do it, and even draws little user flow diagrams.

It's surprisingly fast. Now, I didn't

14:37

even know that you can use code for this. So, um let's just see.

Okay, focusing only on the UI. So, let's let's go and do this and see what happens.

So, we'll say, "Okay,

14:54

great. I want you to break the whole thing into shippable chunks, focusing only on the UI.

For each chunk, tell me what each

15:10

page will show, what you can do on it, and even draw little user flow diagrams. It's surprisingly fast.

Um, all right. I'm I'm I I'm not sure

15:28

this is going to work, but if it does, okay, shippable UI chunks. It's doing it.

It is doing it. So, what it's doing is making these shippable chunks.

And then we're going to take this. Wow.

The user flow doc. I

15:45

mean, user flow diagram like pretty basic, honestly. But like maybe this works.

Maybe this will be good enough for Vzero. Like it's super basic.

This is basically like a uh you know I you know a pen on a napkin, right? An idea

16:00

on a napkin. Okay.

Wow. Okay.

So then let's go back here. I check clothes work.

I almost said cloud there. Made make any tweaks and get then get it to turn each UI chunk into a prompt for v 0ero dev.

Okay. So make each UI chunk

16:19

into a prompt for v 0ero dev because and this is this is a little thing I like to explain to the LLM why I'm doing stuff sometimes cuz or else it gets lost because I am going to get Vzero to make

16:35

this a beautiful and living thing. Let's see if this actually works.

Oh man, crazy. See, you don't even need to know how to prompt really well here, right?

Like there's so many guides around how

16:52

do you how do you get the most out of VZ, for example, but it's like just ask code or ask chat GBT how to do the best prompting. Like it will tell you how to do the best prompting.

It's incredible.

17:07

Okay, so now we have all these incredible prompts and I'm just going I'm I'm reading through it. I'll read through one.

The novel spin wheel design an interactive spinning wheel feature showcasing book covers around a circular interface. When spun, the wheel should animate smoothly and highlight a

17:24

selected book with subtle hapt haptic feedback and visual emphasis. Include filter controls to customize wheel contents by genre or mood.

When book is selected, it should expand to show basic details with option to view full details

17:40

or add directed to library. I couldn't have said it better myself.

That's pretty awesome. So now let's go over to vzero.dev and start generating the UI and we could tweak the prompts uh as we go until it looks right.

So then

17:55

let's pull let's just do the no you know the novel spin actually core reading experience. Let's start with this chunk one the core re reading experience and see what happens.

Just going to literally just paste it yolo press

18:15

uh press go and then see what happens. So it's making a next uh js app here.

uh code is coming. The reason I started with uh the first chunk, by the way, was

18:31

because I felt that if I went with chunk two, it wouldn't I felt like it needed some of like the overall aesthetic uh prompting in order to get that right. So, I feel like you do need to start

18:47

with your first chunk. Um, as a as a little tip there, um, I gotta say, just like going through this, I I knew that that Reddit post, like I knew that it would be directionally right.

Um, it just felt directionally right because I'm I'm constantly building my own ideas, but

19:03

I'm I'm pretty shocked with how well everything is going so far. like as someone who's built um has advised multi-billion dollar companies, you know, as an adviser to Reddit, adviser to Tik Tok, I've I've built ventureback companies, consumer back companies uh

19:19

before, and I'm just impressed with how quickly that, you know, it gets to the to the to the substance, right? We're we're right to the substance.

I was I was really impressed with uh just like the questions actually like here's the

19:35

20 questions that you should ask yourselves. So many times we have ideas for stuff but we shouldn't even we shouldn't build them.

Um and we should just like kind of hash it out. And the way to hash it out, you know, it doesn't it doesn't necessarily need to take time.

Like we're talking a few hours to

19:51

like figure out if this is a good idea or not. But um yeah, cuz I could have gone through that process answering those 20 questions and be like, you know what, there's just actually just too much competition.

I'm not sure that focusing on Gen Z is the right approach. So happy

20:07

I went through that. Okay, novel spin.

Uh here we go. It's It seems like it's a mobile.

It looks like a mobile web app. So, wow.

It's gone and created a home, the library, the search.

20:22

Um, and it's, you know, it looks fine. I would say it doesn't nothing to write home about.

I don't know what you think, but I don't think there's anything to write home about. Um, I assume, you know, it's still loading the the the photos, so maybe

20:39

it'll look nicer with photos. But I'm just going to ask it like, is this a mobile app or is this a mobile web app?

Could you design a native mobile

20:58

app? I'm not sure it could.

I mean, maybe it could. Let's see.

I've never tried it. Right.

So, it's it's gone and it's told me that it's design a mobile optimized web app built on X.js, not a native mobile app,

21:13

but it could create a mobile app design using React Native. That would work.

Now, that's not a React Native isn't, you know, Swift, right? So it's it's not uh it might might not be as nice as for

21:28

example the Q app that I showed which is probably uh developed using Swift uh which is you know known as that's actually a native mobile app but uh let you know React Native would be a great

21:44

place to start if we're just if we're just testing to see and see if we can get product market fit with this idea. So, while that's going, let's go back to the Reddit post and see what we have to do next.

So, he says, uh, he generates

22:01

UI piece by piece, uh, tweaking the prompts as he goes until it looks right. And then once the whole UI is done in vzero.dev, he literally just downloads the co code.

He gets code to write in a simple readme file that explains what

22:17

we're building. Then he uses cursor or VS or code pil code code pilot you know pick a tool of your choice to start adding the database um backend logic and all the stuff that makes it actually work and that's pretty much how he goes

22:32

from a random idea to hopefully a working SAS product. It's kind of crazy how much AI speeds things up these days.

Yes sir. So, let's go and just see what Vzero ends up

22:48

uh po, you know, creating. And and by the way, like this is why I wanted to do this.

I you know, the the reality is we know that Vzero is going to create something that's like probably pretty nice. And you know, the point of this episode is not to show you

23:04

how to take it from, you know, 80% beautiful to 90% beautiful. um because I don't actually I actually don't think that matters um you know realistically.

I think what's going to make this product stand out with good

23:21

readads is it just needs to be clean. It needs to be aesthetic.

The copy needs to be high quality. Um and it needs to be actually serving a real painoint.

And this process of actually how to build a SAS app uh within in a weekend using AI. This is this is the process that I would

23:39

take. It makes a lot of sense.

So let's go and just see what it looks like. Um see it's already written 240 lines of code just on the profile screen alone.

So this is it's kind of crazy. So and

23:54

you can't preview it yet. It's just loading.

So you can see on the left hand side um look at all the different pages that it's it's creating all the how much lines of codes here um okay

24:11

so it's it seems like it's it's completed I've created a native mobile app using React Native okay but I can't preview it. So basically, wow, I can't preview this, which is kind of

24:26

lame. And then let me just ask this.

Why can't I preview this? Right.

So if you wanted to preview this, you'd have to download it. Download the code and and just throw it

24:42

into Xcode, which is, you know, you should watch. We we actually did a you know a video an episode recently about how we actually do that and and what are the best practices.

I'll link it in the show notes. Um but friends, we did it,

24:58

you know, we didn't we we did it. We figured out how to actually build a SAS app in a weekend.

This is the process I would use. Uh and uh hope you enjoyed hope you learned something and uh you know stay

25:17

sipping enjoy happy building. Don't just share this uh video or episode with a friend.

Don't just like and comment. Although like and comment and subscribe but most importantly go and ship your ideas.

Go build something. And I can't

25:33

wait to see what it is you built.

Summary

Transcript