RizGPT History Timeline

• Jul 2023 - RizGPT launch
• Aug 2023 - $80k MRR
• RizGPT rebranded to Plug AI
• Sept 2023 - The app reached $200k MRR / $2,500,000 ARR
• Oct 2023 - Plug AI reached 1.5 million downloads 

Beginning

Blake’s college roommate kept nagging him for help with dating app chats. Blake wondered, "Could ChatGPT handle this?". He decided to find out.

Building the app took a month. Blake didn't know how to code. ChatGPT did 99% of the code. It walked Blake through the process step by step.

1. Start with design: Blake sketched the layout in Figma.
2. Choose your platform: He picked React Native, Swift UI and Kotlin.
3. Use ChatGPT to build: Blake made sure to give ChatGPT clear instructions. Something like:
   • gradient background, a logo, two buttons
   • adjust spacing between buttons
   • when Blake needed to add a feature, like accessing the camera roll to upload an image, he simply told ChatGPT. It explained how to implement it. Every time he hit a bug, he’d go back to ChatGPT, ask why it wasn’t working, and get the fix.

The key, Blake found, was that it was easier than expected. It was like having a professional tutor right there, 24/7.

MVP

There were no notifications. No requests for reviews. It was the simplest app you could imagine—press a button, upload a screenshot, get your answer. Nothing more.

After that, they were catching up, learning what should have been learned before the launch. 

It started free, then after a few credits, the cost was $7 a week. Now, it’s just $7 a week flat.

And that’s how RizzGPT, now Plug AI, came to be.

Why rebranding?
Blake didn't do research before getting to market his app. There was already a RizzGPT app. It had done less than 200k downloads at this point (Blake did that in their first week after tiktok promo). To prevent legal issues, Blake changed the app's name to Plug AI.

Distribution

Blake launched the app on TikTok, and at first, it did okay. A few hundred downloads trickled in during the first few weeks, but they hadn’t cracked the code yet. Then he found a niche: Riz content—guys showing off how smooth they were with girls. Untapped potential. He hit 2 faceless accounts, offering $50 for promos. These accounts didn’t have huge followings (50k - 100k followers), but they were getting millions of views.

These influencers didn’t even change their content. They just worked the app right into the middle of it. Normally, they'd post screenshots of their Tinder conversations, showing off their game. 10-image slideshows with a meme tossed in. Blake reached out and said, “We love your content. Want to use our app to show off your Riz? Put it in the middle. Say this is where you get your Riz from.”

It worked like a charm. The audience was hooked, thinking, "If I download this app, I’ll be just like him".

And the results? It blew up - 200k downloads in 6 days, 80k MRR right off the bat.

Umax History Timeline

• Dec 2023 - Umax launch
• End of Dec 2023 - $100,000 revenue
• Mid Jan 2024 - $200k MRR - a copycat showed up
  • 5 days later - Umax reached $500k MRR
• Feb 2024 - $700k revenue made in a single month
• Apr 4th 2024 - Umax reached 4,000,000 downloads
• Sept 2024 - $6,000,000 ARR

Piggybacking on a trend 

The "look-maxing" trend had taken hold. People, especially young men, became obsessed with improving their looks. It was the next evolution in social media, much like fitness content. Most men worked out not just for health, but to appeal to women. Look-maxing stripped away all pretense. It wasn’t about self-improvement anymore; it was about becoming attractive. That was what caught Blake’s attention.

There were 2 problems to solve in this new space.

1. Many were desperate to know how attractive they were. Whole subreddits appeared, filled with people asking for ratings, seeking validation.
    
2. They wanted advice—personalized advice. But what they found online was general, one-size-fits-all.

Around that time, GPT Vision was released. Blake saw an opportunity. He didn’t need to build an expensive machine learning model to analyze faces. With a little tweaking, GPT could do the job.

The not-so-modest beginnings

And so Umax was born. Blake, once again used ChatGPT to create an app (85%~ of Umax's code was generated by ChatGPT) then he brought his brother on for the technical support and partnered with Sam Zia (TT/IG), an up-and-coming the main look-maxing influencer with a modest following but huge engagement. 

In one month, Umax was ready. Built with Swift UI and an AWS backend. When it launched, Sam’s video pulled in over 100k views, and downloads began stacking up. 

In the first month, Umax brought in $100k.

Copycats and rapid growth

By January 2024, a competitor appeared, and they were ruthless. They copied everything—marketing, metrics, even the jawline scoring. They surpassed Umax in downloads within two weeks, something that had taken Blake’s team a month and a half. The threat was real. The competitor didn’t care about refining their product; they wanted to win. Blake faced a choice: keep steady and let the competition overtake them, or fight back.

The plan had always been to grow slowly, make improvements, optimize retention. Blake believed that once consumers experienced a product, their first impression stuck, even if the app improved later. But faced with a competitor willing to sacrifice quality just to beat them, he had no choice. They had to burn through the forest while it was still theirs.

In 3 days, he spent more than $200,000 on influencer campaigns, every dollar they had earned up to that point. He reached out to influencers like K Shami and Dillon Latham, locking them into exclusive deals. The competitors had been circling them too, but Blake moved fast, offering $50,000 upfront. It was a gamble. But it worked. Umax’s numbers soared—revenue climbed from $200,000 MRR to over $500,000 MRR in just five days. Downloads surged by the tens of thousands daily.

Blake hardly slept through that period. He was on the phone constantly—dealing with influencers, coordinating with engineers, keeping the social media machine running. 

In the end, Blake’s team won. The competitor had a strong start, but Umax outpaced them in the long run, quadrupling their downloads and revenue. Umax had become the dominant force in the space.

Cal AI History Timeline

• April 10 2024 - Cal AI launch
• July 20, 2024 - $300k MRR
• Aug 31, 2024 - $500k MRR
• Sep 5, 2024 - Cal AI reached $1,000,000 in revenue
• Sep 28, 2024 - $700k MRR
• Oct 14, 2024 - $1,000,000 MRR
• Mar 2nd, 2026 - $50,000,000 ARR and sold to MyFitnessPal for approximately $100,000,000

How it started

In March 2024, Zach, who was just 16 at the time, messaged Blake giving some advice about Umax. It was sharp, straight to the point, and made sense. Soon enough, they were on calls together, and before long, Zach and Blake became partners, equals in the venture.

When Umax was riding high, Zach suggested a collaboration. "I've got a buddy I've been working with. We're open to working on anything you’ve got. We’ll handle operations, and you just give us the playbook, maybe put in some capital."

That’s when Blake came up with the idea for Cal AI. It clicked. Both Zach and Blake were heavy users of MyFitnessPal (calories, fitness & fasting tracker), but the app didn’t cut it. It was clunky, outdated.

They knew calorie tracking was one of the biggest categories on the App Store. It had potential, but it was stagnant, stuck in the past. Their idea was simple: bring in AI, make calorie tracking smarter, more precise, and easier than ever before. That’s what led to the birth of Cal AI.


Blake took the lead in laying out the strategies, the keys to success that had made RizzGPT and Umax thrive. He provided the playbooks, and budget that set the foundation for their success.

Influencers

Cal AI relied on a network of influencers in the fitness, wellness, and self-improvement space. Besides paying influencers a fixed amount, they also offer referral codes to motivate them further. 

They are running targeted influencer marketing campaigns aiming for about $5 CPM.

Once you can figure out a profitable cost per acquisition (CPA) and a conversion rate per thousand views, having a CPM target helps refine the influencer strategy. With regular influencer marketing—whether using small or big influencers—it’s usually easy to estimate how many views you’ll get, as it often matches the influencer’s average performance

More about their strategy in the playbook section

* (It wasn't Zach's first success) - Gaming platform for students

During the pandemic, high schools handed out Chromebooks to keep students apart. But instead of focusing on lessons, they turned to games, scattered across different websites. Multiple browser tabs were opened all the time.

Zach saw it clearly. A single hub for all those games—it just made sense. So he built one, called it Totally Science. It let students play almost any game they wanted. Knock-off versions of Minecraft, and others like it. The idea wasn’t just practical. It made Zach feel like he’d be the cool kid if everyone at school used his site.

He shared it with a few friends. Word spread fast, faster than he expected. It caught fire.

But Zach didn’t stop there. He wanted more. He made TikToks during class. The first one pulled in 100k views, the next, a million.

The site kept growing to over 4.5 million unique users, pulling in $5,000 a month from ads. Two years later, Zach sold it for $100,000.

1. Seizing opportunity: the new era of bold ideas

The opportunity staring us in the face right now.
In mobile apps, since the App Store opened its gates in the late 2000s, we’ve seen no real game-changer.

Sure, there have been improvements in machine learning, vision, or optical character recognition, but nothing that shakes the core of how we interact with machines. But now, with AI and these large language models, the possibilities have exploded.

The cost to create them has never been lower and it's also easier than ever. Blade built RizzGPT and Umax using nothing more than ChatGPT. Anybody can do it now.

Developers don’t grasp social media virality.
And the social media virality people don’t understand building software.
Take best from the both worlds and you'll win.

2. Simplicity wins: building products that solve real problems

Find a simple product. Don’t overthink it.
With RizzGPT, it was clear. Men already faced a challenge—talking to women and getting responses on dating apps. The solution had to be simple, direct. Something that could connect with many people. That was the driving idea behind the app.

Umax was the same. Men wanted to know how to be more attractive, get personalized advice. The goal was to package this into a simple app. People often create complex solutions for basic problems. But Umax kept it simple.

It's all about clarity—building apps that tackle one clear problem without overengineering.

The first version of Cal AI? It was strikingly simple. The approach? Release a basic version, gather feedback, and grow from there. This approach, more than anything, is the key to success.

Before starting the app make a quick test:

• The Friend test—whenever an idea comes, it’s easy to run it by friends. When they say, “Yeah, that’s cool, you should definitely do that,” they mean well. They’re trying to be supportive. But beneath it, there's something else. It’s as if they’re saying, “That’s a damn stupid idea. I’d never use it.”
  
  But when they look at it and their eyes light up, when they say, “Holy hell, this is great,” or they laugh because they genuinely find it funny—then I know it’s different. That’s real. That’s something they’d use.

And before releasing the app, do another one:

• The Mom test—hand the app to your mom. If she can navigate it easily, it’s probably solid. The average user isn’t as tech-savvy as you, or even your friends. 

3. The circle of writing code for your app

There's a meme about a developer’s journey (the one above). Unless you’re building something massive, something that requires scalability for millions of users, your code doesn't need to be perfect.

For the single-user experience, complexity is a waste. Let there be delays if they come, because in the early stages of any startup, what’s crucial is getting something out there. What’s important is releasing something, seeing if it works, and improving later.

As for tools, Blake relies on ChatGPT or Claude 3.5 Sonnet for coding. These language models power the backend of RizzGPT and Umax, and they get the job done.

4. The quickest way to test an idea

To validate your product fast, try using micro-influencers. They can help you reach the right people.

Choose your influencer with care. They should fit your niche and speak the right way for your product. Take your time. There’s no rush—just make sure it’s a good match.

Validation happens when you find your audience. Aim for at least a 100k views to see if there’s real interest. This will help you decide if your idea is worth pursuing.

5. Scaling strategy: the lean path

Start with what you have
If your capital is limited, focus on creating your own content—it’s cost-free. Utilize every available platform to share your content and aim for visibility. While internal user-generated content (UGC) can be highly effective, it requires a significant investment of your time, which can be costly for a founder.

Leverage micro-influencers
Identify micro-influencers within your niche. For example, the initial promotions for RizzGPT were priced at $50 each and resulted in nearly 200,000 downloads. While this success involved an element of luck, it also provided valuable insights.

Calculate your ROI
When you invest $200 in a micro-influencer, and that promotion generates $400 in profit, you’ve netted $200. This is the pivotal moment where you can start to strategize for scaling your efforts. Focus on identifying and replicating successful tactics.

It’s not easy, but it’s the way.

6. The Influencer Strategy

Scaling through influencers is the biggest play.

Avoid one-off deals
Many brands make the mistake of striking one-off deals, paying for a single post and then returning later for another. Instead, the focus should be on securing retainer agreements, where influencers commit to posting multiple times each month. This approach allows you to gradually build a roster of influencers and steadily increase your viewership.

The risk of audience fatigue
However, there’s a potential downside. Over time, an influencer’s audience may experience fatigue, leading to diminishing returns. This issue can be difficult to anticipate, as tracking results and attributing success to influencer partnerships can be complex. When fatigue sets in, what once proved effective may quickly become unprofitable.

7. Finding the right influencer

Finding the right partners is tough. Many companies cast a wide net, sending countless messages and settling for anyone who responds. Others rely on agencies, but the approach remains haphazard.

Success requires a sharp instinct for spotting influencers who can convert. It's not just about follower counts; it’s about engagement. Look at the comments. If their audience listens to recommendations, you’ve found value. Many influencers lead to losses, but the few that deliver a 10x return are worth pursuing.

Know your audience
When assessing an influencer, remember that numbers tell only part of the tale. The best partnerships come from understanding your audience. Immerse yourself in their world to find influencers who truly resonate.

Qualities of a good partner
Finally, consider the influencer's quality. Are they just creators? A solid partner should be hardworking and entrepreneurial. Some excel as content creators, but they may not help you build a business. Seek true partners who can grow with you.

8. Breaking through: the art of unstoppable networking

Getting an influencer’s attention is tough. Out of every 1,000 attempts, maybe 1 will respond. But it’s not just about that one reply. It’s about persistence and hard work. Once you get a response, don’t stop. Ask for introductions and expand your network.

Don’t waste time on small fish, though. Working with a big player changes everything. But breaking through is hard. With only a few hundred followers, your DMs might go ignored. Connections can help you find the right people. Prioritize. Friends, family of the influencer, anyone who might help. Tell them you’ve got money for a promotion and you need the right contact. Connections smooth the way, especially once you know which DMs actually get answers.

Finding the right hook
Use a direct approach, like starting with “Paid promo?” It tells the influencer exactly what you want. If they’re interested, they’ll respond.

Be impossible to ignore
Blake Anderson faced the same wall with unresponsive creators while building Rizz GPT. He didn’t wait; he messaged everywhere—TikTok, Instagram, Discord. He became impossible to ignore. Make enough noise, and they’ll respond, even if just to quiet you down.

Why big brands fail
Big brands often miss the mark in influencer marketing because they lack grit. Startups need to reach out daily. Many companies think they’ve struck gold with one reply from a million-follower influencer. But that account might only get 10,000 views. Numbers can be deceiving

9. Navigating influencer partnerships

When structuring deals, there are generally two types of groups:

• The first-time earners: these individuals are trying to secure their first success, aiming to generate between $0 and $10k a month. Often, they may not have significant capital readily available.
  
  
• The scaling group: these people are looking to scale up from their initial success, having already achieved some level of consistent earnings.

Upfront costs with Influencers
Larger influencers tend to demand high upfront payments, often reaching four figures or more. This can be a significant challenge for those in the first group who may not have the cash to invest.

Partnering for quick scaling
If you want to scale fast and reduce financial risk, partnering with an influencer can help. Offering them 40% equity lets them handle distribution, requiring little cash from you. This can push profits from $0 to $50k a month. But while it’s effective for quick growth, it might not be the best for long-term stability.

Although influencers provide valuable exposure, they may not be the most reliable business partners for long-term growth. Many lack the business acumen needed to sustain growth, making it risky to rely entirely on their involvement for long-term success.

Cash vs. reach
Most influencers prefer cash payments over equity. Their value should be assessed based on their reach:

An influencer receiving 100k views per video could potentially help a viral product reach 500k views. For niche products, those same influencers might only generate 20k to 50k views.

Knowing your gross profit per thousand views is essential when determining how much you can afford to pay an influencer. The better your understanding of your market, the easier it will be to judge an influencer’s worth.

Performance vs followers
When paying influencers, focus on views, not follower count. A simple deal structure: “I’ll pay X for Y views, with bonuses if we hit Z views.” That way, you only pay for results.

10. Influencer marketing spend margins

Businesses chase influencer marketing, but most don’t turn a profit. Many, like dropshippers, run at a loss for online fame. They build shaky businesses, then sell courses claiming million-dollar success. What they hide are the losses. Followers don’t know, and the myth of wealth grows as courses sell.

Even big brands show high revenue, but profits often lag. The key isn’t follower count; it’s views and engagement. Keeping this focus can push margins to 20-50%.

Tight margins in apps
In the app space, Apple’s 30% cut squeezes profits. But in web-based B2B SaaS, margins widen, giving more breathing room.

11. Creative freedom in influencer campaigns

When working with influencers, the big question is: how much control do you need? Do you trust their instincts or give strict instructions? The answer lies in between.

Trust their skill, but ensure conversions
Influencers know how to keep their audience engaged, but they don’t always drive sales. You need both creative freedom and key instructions. For instance, make sure your product appears in the first 15 seconds. Viewers won’t wait till the end. Ads buried too late never convert.

Seamless ads that don’t feel like ads
Audiences, especially younger ones, are skilled at tuning out ads. Your product must blend into the content, making viewers curious without feeling sold to. But remember, always follow FTC guidelines—be clear it’s an ad.

Virality over conversions
Many brands make the mistake of focusing on conversions when creating organic content. The goal should be virality. A million-view video with a 1% conversion rate beats a 100k-view video with a 5% rate. Grab attention first; conversions come later.

Scale with trends
In niche markets like Umax, creators see what works and replicate it. Give influencers examples, not just instructions. Trust them to take control, but guide them with clear expectations.

Always approve before going live
To succeed, certain rules must be followed: feature the product early, and approve content before it goes live. Beyond that, let creators experiment.

Subtle conversions
When viewers hit a paywall, they won’t need much convincing if they’ve already seen the product in action. Real-life usage in videos beats hard-selling every time.

12. App pricing strategy

Launch strategy: free vs. paid
Some developers believe the app store promotes newly launched apps, which is why offering it for free initially can boost early traction. However, Blake’s strategy avoids free versions, opting for paid apps from day one.

Pricing models: Umax vs. Cal AI
Umax is priced at $5 per week, while Cal AI costs $4 per month. Pricing decisions should be based on target market analysis and competitor research.

UMax started at $4 per week, but as competitors quickly adopted similar pricing, it became the industry standard, despite limited testing.

Cal AI is priced lower at $4 per month to compete with higher-priced apps like MyFitnessPal, which charges $20 a month. Initially, Cal AI's price was set at $10, and some split testing was done. However, the focus was less on maximizing lifetime value (LTV) and more on increasing user adoption.

Split testing for optimal pricing
Tools like Superwall enable real-time split testing to determine the best pricing model. Early testing with lower prices can attract users, with adjustments made later based on data.

13. The simplicity of marketing: the game of numbers

RPM vs CPM
Marketing boils down to two key metrics: RPM (Revenue per Thousand Views) and CPM (Cost per Thousand Views). Success is simple—if you’re paying less to get attention than you’re earning from it, you win. Methods may vary, but the principle remains the same. While not always scalable, this approach works.

A winning campaign happens when RPM exceeds CPM. For example, if your RPM is $5, you target influencers with a CPM of $3, offering $3,000 for a million views. Calculating costs is key—knowing the influencer’s average views lets you set fair terms.

Costs matter
Beyond marketing, product costs like production, shipping, and platform fees must be factored in. Though Apple takes a 30% cut, its users convert better than android ones ("Android converts 1/10 the amount of IOS users"), and Apple may even promote your app, helping boost downloads.

Marketing isn’t complicated—keep costs lower than returns, and you’ll succeed.

14. Viral loops

Their viral loops were sharp, the K-factor high. Every user brought more than one, sometimes two. They incentivized heavily, weaving it into the product itself.

With Umax, they gave users a way out of payment. Share the product with others, and your cost disappeared. It was necessary because their competitor was doing the same, and the fight for market share had become a bloodbath. 

Additional Sources

1. Podcast with Omar Waseem and Blake 
2. SquidTalk Podcast with Blake
3. Brett Malinowski podcast with Blake
4. Twitter Space with Zach & Blake
5. Q&A sessions on Arib Khan's discord channel

Authority Bias

Blake knew people trusted experts. RizzGPT worked because men saw influencers use it and credibility was established. Umax had Sam Zia, whose advice on looks held weight. With Cal AI, fitness influencers promoted it, and their followers believed the AI made calorie tracking more accurate and easy to trust.

Bandwagon Effect

RizzGPT started slow, but once influencers got involved, people followed quickly. Umax spread just as fast—once one person downloaded, others didn’t want to be left out. Cal AI caught on in the fitness community as everyone started using it to stay ahead.

The Mere Exposure Effect

All three apps used the mere exposure effect by getting influencers to mention them repeatedly in their content. The more times these apps were mentioned, the more people recognized and trusted them they will solve their problem.

Solving very clear problem in a friction-free way

These 3 apps solved very clear problems (with huge demand) and provided very clear solutions to them in a friction-free way - that's way easier said than done.

RizzGPT helped guys with their smooth replies for dating apps. Upload a screenshot, get a response—quick and easy.

Umax gave fast, personal feedback on looks. Just upload a photo, get your score. No waiting, no fluff.

Cal AI fixed calorie tracking for fitness guys. The AI provided exact numbers without the hassle. Simple, personal, and straight to the point.