If you’ve been on the world wide web this year, it’s become a daily occurrence to trip over new AI startups, “revolutionary” content generators, and a partridge in a pear tree claiming it’s powered by GPT-9. Half promise to revolutionize content, the other half are trying to turn your grocery list into a TED Talk. revolutionize content It’s a firehose of demos and LinkedIn threads. And somewhere in that mess, a few companies are doing the unthinkable: making money. One of them is Creati: a real, working business pulling in more than $10m in ARR. And there’s a lot to learn from that success. The AI Hype Cycle (a.k.a. The Graveyard of Good Ideas) The AI Hype Cycle (a.k.a. The Graveyard of Good Ideas) Everyone’s got a hot take, a weekend project, and a landing page. But most of these tools collapse even quicker than they’re propped up because they never answer the one question that matters: “What’s the problem this actually solves?” Creati’s founder, Ella Zhang, learned this the hard way. Her first product, an AI image generator, clocked 7 million MAUs. Great on paper. But those users weren’t paying. And they weren’t staying. Turns out, “Look what AI can do!” isn’t a business model. It’s a science fair project. The Pivot: Talk to Your Users, Not Your VC Deck The Pivot: Talk to Your Users, Not Your VC Deck Here’s where Creati pulled a move that shouldn’t be radical but somehow manages to be here in 2025: they asked their users what they actually wanted. Wild stuff, I know. Wild stuff After interviewing over 300 people - 300 real conversations, not just SurveyMonkey hits - they learned that their audience wanted AI that helped them look good: marketers, e-comm sellers, creators, brand managers. them Not just flashy outputs, but branded, on-message, plug-and-play content. So Creati made a hard pivot: they became an element-based video generator, letting users upload their own brand assets (think products, faces, backgrounds) and drop them into slick, pre-built AI video templates. No prompt engineering. No rolling the dice with random generations. Just “put in your stuff, get something useful back.” Conversion rates went from 5% to 50%. That’s not an A/B test win—that’s a holy sht* moment. holy sh That’s, dare I say, product-market fit. product-market fit Don’t Be Afraid to Charge Don’t Be Afraid to Charge Here’s a lesson I learned the hard way: traffic is a vanity metric; revenue keeps the lights on. Creati stopped treating their product like a toy and started charging based on business value—save time, make content that converts, stop throwing money at agencies. Simple pitch. Better results. They dropped the freemium handcuffs and put their faith in actual utility. Fast forward within a year: over $10 Million in annualized revenue. Turns out, people will pay when you solve a real problem. Building Product-Led Virality Building Product-Led Virality A lot of startups talk about growth loops and network effects. Most of them are wishful thinking dressed up as diagrams. Creati actually executed. Every video their users created was ready-made for social media. Every share was a free ad. And because Creati also partnered with skilled creators to build their template library, they got a second loop: when a creator shared their work, they brought their audience with them. That’s a compounding flywheel between creator and brands. Most SaaS companies would kill for that kind of baked-in distribution. What AI Founders Can Actually Take Away from All This What AI Founders Can Actually Take Away from All This If you’re building in the AI space and hoping to do more than go viral on Product Hunt, take a beat and ask yourself: Are you solving a real problem, or just demoing magic tricks?
When’s the last time you actually talked to your users?
Are you afraid to charge? (You shouldn’t be. People taking out their wallets is the only way to determine if there’s a market.)
Is your product easy to share? (Word of mouth is still undefeated.)
Can you plug into existing workflows instead of asking people to learn something new? Are you solving a real problem, or just demoing magic tricks? When’s the last time you actually talked to your users? Are you afraid to charge? (You shouldn’t be. People taking out their wallets is the only way to determine if there’s a market.) You shouldn’t be. People taking out their wallets is the only way to determine if there’s a market. Is your product easy to share? (Word of mouth is still undefeated.) Word of mouth is still undefeated. Can you plug into existing workflows instead of asking people to learn something new? Creati didn’t get here because they built cooler tech: they won because they listened, adapted, and treated revenue like a feature rather than an afterthought. listened Everyone else? They’re still out there polishing their pitch deck, hoping to get lucky.

This story contains new, firsthand information uncovered by the writer.

Every

Pivot

Why I’m Jealous of Today’s Builders

If Your Products Aren’t AI-Searchable, You’re Already Losing

The Discoverability Company

More About Drew Chapin

Too Long; Didn't Read

CTA: Download the FREE AI Productivity Shift report

How Creati Quietly Hit $10M in AI Revenue

About Author

Comments

TOPICS

THIS ARTICLE WAS FEATURED IN

Related Stories

Untitled Story

Wikipedia Rules Everything Around Me

The Noonification: How Often Do NFTs Pass The Howey Test? (1/13/2023)

Darwin's Hybrid Intelligence to Align AI & Human Goals for Startups & VCs

The Noonification: White Man (11/26/2022)

The Noonification: The Metaverse is a Sh*tshow (11/2/2022)

Wikipedia Rules Everything Around Me

The Noonification: How Often Do NFTs Pass The Howey Test? (1/13/2023)

Darwin's Hybrid Intelligence to Align AI & Human Goals for Startups & VCs

The Noonification: White Man (11/26/2022)

The Noonification: The Metaverse is a Sh*tshow (11/2/2022)

Light-Mode

Classic

Newspaper

Dark-Mode

Neon Noir

Minty

HN StartUps