When Rebecca Yu got tired of her friend group arguing over where to eat every Friday night, she did something most people wouldn’t think of. She opened an AI coding tool and described the app she wanted. Seven days later, Where2Eat was live, recommending restaurants based on shared preferences. Yu had never written a line of code in her life.
She’s not alone. A growing number of founders and non-technical professionals are building their own software tools from scratch, not to sell them, but to use them. They’re called micro apps, and they’re quietly replacing the SaaS subscriptions that used to cost businesses thousands of dollars a year.
What Micro Apps Actually Are and Why They Matter
Micro apps are small, purpose-built tools that one person creates to solve one specific problem. Unlike polished software products on the App Store, these apps are often used by just the creator and maybe a few friends or teammates. TechCrunch calls them “fleeting apps” because some only exist for days or weeks before the creator shuts them down.
The concept gained traction after AI coding tools like vibe coding platforms made it possible for anyone to describe what they want in plain English and get working software back in minutes. The vibe coding market hit $4.7 billion in 2025 and is projected to reach $12.3 billion by 2027, growing at 38% annually.
Real Founders Who Stopped Subscribing and Started Building
Nick Simpson built an app that scans parking tickets and automatically pays them. It started as a personal fix for his own forgetfulness, but after putting it on Apple’s TestFlight, a dozen of his friends wanted it too. He never planned to sell it. He just needed it to work.
Jordi Amat built a web gaming app for his family over the holidays. Once the vacation ended, he shut it down. Total cost: a few hours and an AI subscription he was already paying for. No Slack thread, no vendor demo, no annual contract.
Then there are the founders building micro apps to replace specific SaaS tools they were paying for. One software engineer built a heart palpitation logger for a friend who needed to show her doctor a timeline of symptoms. Instead of hunting for a health tracking app that did exactly that one thing, she just described what she needed and had it running the same day.
The Numbers Behind the Shift
This isn’t a fringe experiment. Lovable, one of the leading AI app-building platforms, hit $100 million in annual recurring revenue in just eight months, potentially the fastest-growing startup in history. Cursor crossed $2 billion in ARR. Replit’s revenue jumped from $10 million to $100 million in nine months after launching its AI Agent feature.
Mobile-focused startups are chasing the same opportunity. Anything raised $11 million led by Footwork to help people build mobile apps without code. VibeCode raised a $9.4 million seed round from Alexis Ohanian’s Seven Seven Six fund. The addressable market for micro apps is estimated at over $50 billion in displaced SaaS spending.
Meanwhile, the average small business spends between $20,000 and $100,000 per year on SaaS subscriptions, according to industry research. Much of that goes to tools where teams use maybe 20% of the features.
Why Founders Are Choosing “Build” Over “Buy”
The logic is simple. When AI can generate a working prototype in under 30 minutes, the calculus between building and buying flips entirely. Bolt.new produces working prototypes in 28 minutes on average. Lovable takes about 35 minutes. Even Replit, which is more powerful, gets you there in 45 minutes.
For founders running lean teams, this changes everything. Instead of paying $200 a month for a project management tool with 50 features they don’t need, they can build exactly the tool they want for the cost of their existing AI subscription. The micro-SaaS segment is growing at roughly 30% annually, from $15.7 billion in 2024 to a projected $59.6 billion by 2030.
Collins Dictionary named “vibe coding” its Word of the Year for 2026. That’s not a tech industry award. That’s a signal that building software has crossed into mainstream culture.
The Risks Nobody Talks About
Micro apps aren’t without problems. Research shows 45% of AI-generated code fails security tests, and projected technical debt from vibe-coded software could reach $1.5 trillion by 2027. A micro app tracking your parking tickets is low-stakes. A micro app handling customer data or financial transactions is a different conversation entirely.
There’s also the maintenance question. SaaS products come with teams of engineers fixing bugs, patching security holes, and keeping things running. A micro app built in an afternoon doesn’t come with any of that. When something breaks at 2 a.m., you’re the entire engineering team.
What This Means for Founders Right Now
The smartest founders aren’t treating micro apps as a replacement for all software. They’re using them to fill gaps. The essential tools for early-stage startups still matter, but the space between those core tools is where micro apps shine. Need a custom dashboard that pulls data from three different sources? Build it. Need an internal tool that does one thing your CRM can’t? Build it.
Y Combinator’s Winter 2025 batch revealed that 21% of accepted startups had codebases that were 91% or more AI-generated. The barrier between “I have an idea” and “I have a working product” has never been thinner. For founders who’ve been paying for software they barely use, that’s not just interesting. It’s a signal that the $50 billion question, build or buy, finally has a new answer.
