Weekly writing about how technology and people intersect. By day, I’m building Daybreak to partner with early-stage founders. By night, I’m writing Digital Native about market trends and startup opportunities.
If you haven’t subscribed, join 65,000+ weekly readers by subscribing here:
The ChatGPT Prompts That Can Be $1B+ Companies
My friend Chris Paik has an armchair theory that any subreddit with >1M subscribers is large enough to support a standalone, venture-scale business.
In some ways, this a modern-day version of the unbundling of Craigslist or the unbundling of Ebay—two frameworks that underpinned large-scale startup creation in the 2000s and 2010s.
Here’s the unbundling of Ebay:
And here’s the unbundling of Craigslist:
Of course, not all these companies survived—the list has consolidated over time:
But the point is: big horizontal early-movers were eventually broken off into valuable standalone pieces.
Verticalization = better products, and better products = value getting siphoned off of horizontal players over time. I think we’re going something similar with ChatGPT.
I don’t necessarily mean that ChatGPT will splinter into vertical search and vertical chatbots. That will happen, but it’s easier to predict. Of course we’ll have different chat interfaces for therapy than for tutoring; more tailored AI applications are already emerging.
What I mean is: any well-crafted prompt can probably be its own standalone app.
The Productization of Prompts
Let’s take an example. I’ve become interested in the sub-culture of TikTok that recommends ChatGPT prompts. Prompt-writing has become something of an art form, and many enterprising creators have turned this art form into TikTok content.
Here’s a TikTok that recommends a ChatGPT prompt to “make 2025 your best year yet.” To save you time, the prompt here is:
I want to create my goal and vision for 2025. Please guide me through this process by asking me 10 questions one by one, and when I'm done answering the 10th question, please write my dream life story for 2025 what my 2025 affirmation will be and then my top 5 goals broken down by family, money, body and life.
When you finish answering the questions, you’re instructed to input another prompt:
Can you put this into a daily schedule and also create a goal tracker? Put this into a table that I can copy or download.
This is a well-written prompt. Not many people would come up with writing it without a little nudge. So why should someone have to write it? Shouldn’t this prompt exist in its own intuitive app? We’re about to see the productization of prompts.
Let’s say we turn the above prompt into it’s own app. Let’s call the app “Dream Life.”
Dream Life is pretty simple:
You download the app and you’re met with 10 successive questions, each nicely designed on their own screen.
You tap through and answer, with text or voice or both.
When you’re done, Dream Life generates your daily schedule and goal tracker.
You turn on notifications and you get reminders (“Good morning! Don’t forget to drink your 8oz of water this morning.”) and kudos (“Nice work! That’s 163rd days in a row of hitting your 10,000 steps!) a few times a day.
Not long ago, this app would take a decent amount of time and expertise to build. Now, with tools like Cursor, you can spin this up in no time. The barriers to software creation are falling—see: last month’s The Age of Fragmentation: AI’s Impact on Content and Code—which means that we’re going to see a lot more software get built. This means apps, apps, apps. Most apps will have zero users (about half of App Store and Google Play Store apps today have zero downloads). But others will have millions.
The App-ocalpyse 2.0
When the App Store launched, there was an app for everything. I can’t find the tweet, but someone on Twitter posted this week about what funding looked like back then, linking this TechCrunch funding announcement.
Here’s the first paragraph:
Yo, the simple app that just sends a “yo” to your friends, has closed $1.5 million in seed funding with a $10 million valuation and is finally ready to talk about its investors. They include Betaworks, Mashable’s Pete Cashmore, and the founders of China’s Tencent, among others.
What a time.
That was the first App-ocalpyse, when we all waded through 17 flashlight apps. We’re now heading into the second: the AI-generated apps are about to hit us.
The title of this piece hints at $1B+ businesses from ChatGPT queries. That’s a bit sensationalist. We’ll probably have some $1B+ businesses—to revisit the categories above, maybe it’s an AI matchmaker (a Tinder killer?) or an AI personal stylist (new category creation) or an AI health coach (replacing your trainer or doctor or both?). At Daybreak we have companies in #1 and #2 and would love to find one in #3.
We’re already seeing large venture-backed companies built as “ChatGPT prompts on steroids”—for instance, Function Health and Superpower are basically more robust productized versions of this TikTok commenter’s favorite ChatGPT prompt:
Of course, most of the companies that successfully unbundle ChatGPT usage will be smaller apps. They might not be $1B+ in size, but many can generate $10-20M in revenue with very little investment or upkeep. We’ll see a long tail of AI-powered apps emerge, each productizing well-written but little-known ChatGPT prompts.
Here are three prompts I’ve used in ChatGPT in the past:
My goal is to put on 15 pounds of muscle in six months. Write me a workout regiment that incorporates: chest and tricep day, back and bicep day, leg day, and at least two days of cardio per week. Make each workout 60 minutes and provide alternative options with bodyweight exercises for when I don’t have access to a gym.
I don’t have time to read all the classic books of American literature. Pick the 50 most essential books in the American literary canon. For each, create a two-minute summary with key takeaways, morals, and cultural impact. (Don’t judge me for this one!)
I want to host more dinner parties for friends. Give me creative ideas for how to throw a good dinner party, including: recipes, tips for seating arrangements and guest invites (a mix of people I know and asking people to bring guests), and conversation topics to guide fun, structured conversations without being cheesy.
Each prompt could be its own app, which would actually be a better user experience: seamless onboarding, elegant design, app notifications.
ChatGPT prompt-writing is a popular topic on Reddit—this question here attracted 330 responses in the ChatGPT subreddit:
These prompts will all get productized, as software creation costs (with regard to both time and money) drop exponentially. We’re about to be inundated with AI-generated apps, just like we’ve been inundated with AI-generated content (looking at you, Buzzfeed).
Most apps will fail to break through, but some apps will be savvy with acquisition and build stable, enduring businesses. It won’t be long until 50%+ of App Store apps are made mostly by AI.
2025’s clever ChatGPT prompt will be 2026’s popular, retentive, highly-profitable app.
Thanks for reading! Subscribe here to receive Digital Native in your inbox each week:
