Most people looking for a SaaS idea start in the same place. They scroll Product Hunt. They browse Reddit threads about "underserved niches." They look at what is already working and try to build a slightly better version. The result is predictable: another scheduling tool, another AI writing assistant, another project management app aimed at remote teams who already have seventeen options.
Jordan did not find his idea on Product Hunt. He found it through a paper letter from a friend in prison.
What he built, Parakeet Chat, became a $1.5 million business. It serves a customer that practically nobody in tech was thinking about. And the way it grew, entirely through word of mouth inside a closed system with no app store, no social media, no traditional marketing, contains a lesson about niche markets that most startup advice completely misses.
This is that story. And more importantly, this is the framework underneath it.
Table of Contents
- The Problem: A Market Everyone Had Written Off
- The Discovery: What a Paper Letter Taught Him
- The Build: One Month, No App Store, 200 Paying Users
- The Growth Engine Nobody Talks About
- The Numbers: What $300K Per Year Actually Looks Like
- The Lesson: User and Customer Are Not the Same Person
- The Framework: How to Find the Customer Nobody Is Building For
- What I Learned From This Startup Story
- Key Takeaways
- FAQ
The Problem: A Market Everyone Had Written Off
There are roughly 2 million people incarcerated in the United States. That is a number most people know in the abstract. What most people do not know is what the communication infrastructure around that population looks like.
Phone calls from prison can cost a dollar a minute. Sending a photo to an incarcerated family member involves mailing a physical print through a system that can take weeks. Accessing the internet is not possible. And the services that do exist inside the prison ecosystem, the approved communication platforms, are widely considered overpriced and low quality. Not because nobody noticed. Because nobody with the technical skills to fix it was paying attention.
This is not a niche that appears in startup databases. You will not find it in a list of "20 underserved SaaS markets for 2026." It is invisible to most founders because the customer, people inside correctional facilities, has no social media presence, no way to leave a review, no way to post in a Reddit thread asking for a better tool. They are, almost by definition, unreachable through the channels most founders use to discover problems.
Jordan found it anyway. Not through research. Through a relationship.
The Discovery: What a Paper Letter Taught Him
Jordan was a freelance software engineer in San Francisco when a client he had been working with disappeared from a Zoom call. His girlfriend explained what had happened: he had been sent to prison. Jordan stayed in touch for a couple of years, the way you do when someone goes inside, through paper letters. Very high-tech, as Jordan himself described it.
Over those years of correspondence, his contact told him something that stuck: the services available inside were a massive scam. High prices, poor quality, no real alternatives. Jordan thought he could do something better.
The idea that became Parakeet Chat was straightforward. Incarcerated people in most US facilities have access to an internal email system, a monitored digital messaging platform that allows limited communication with the outside world. Jordan's insight was to connect that system to AI. Specifically, to build a bot that processed incoming emails from inmates, queried ChatGPT or other AI services on their behalf, and sent responses back through the same internal email infrastructure.
No app to download. No interface to learn. No smartphone required. From the inmate's perspective, it looked exactly like sending an email and receiving a reply. From Jordan's perspective, it was a fairly clean engineering problem with a TypeScript backend, a Postgres database, and a queue system built around Redis. The hard part was not the code. The hard part was understanding what people actually needed it for.
The biggest use case turned out to be legal research. People studying their own cases, understanding their rights, trying to figure out what options they had. The second was communication with family. And the third, which Jordan found genuinely surprising, was entrepreneurship. A lot of entrepreneurial questions came through the platform. Some of the sharpest ones he had seen anywhere.
The Build: One Month, No App Store, 200 Paying Users
Jordan had about a decade of software experience by the time he started Parakeet Chat. The prototype took him a month. The payment system took another month. By the end of month two, he had 200 paying users and the business was profitable.
That speed is worth pausing on. Most SaaS founders spend months building before they have a single paying customer. Jordan had 200 within 30 days of the MVP going live. The reason is not that he was a faster developer. It is that validation and building were the same step.
Prison is a closed ecosystem. You cannot send an inmate a link to your landing page. You cannot run a Google ad that reaches someone inside a correctional facility. The standard startup validation playbook, collect emails, run a pilot, gather feedback from a waiting list, simply does not apply. Jordan had to put an actual working product in front of his contacts inside and let them use it. Their reaction was the validation. Within a month, 200 people were paying $15 to $20 per month for it.
The tech stack he used was solid but nothing exotic. TypeScript throughout, React on the front end, Postgres as the primary database, Redis for in-memory queuing, Auth0 for authentication, Prisma for database calls, and Zod for schema validation on incoming data. He has not opened his code editor in three to six months at the time of recording. The AI system writes the code now. The infrastructure that took him two months to build in the early days would take a fraction of that time today.
One design decision was particularly important: the product has no user interface that inmates interact with. There is no app to download, no onboarding flow to complete, no password to remember. They use the prison email system they already know. That constraint, forced by the closed ecosystem, turned out to be a feature. The friction of adoption was near zero because the interaction model was already familiar.
The Growth Engine Nobody Talks About
Word of mouth. That is it.
Jordan is the first to admit that people want a more exciting answer. But when you build something genuinely useful for a community that has almost nothing useful built for it, and when that community is physically contained in a shared space where people talk to each other every day, word of mouth spreads faster than most growth channels a startup could buy.
He did build one formal referral mechanism: if an existing user recruited a new paying user, the original user received a month of free credits. Simple. Easy to explain. Works within the constraints of a closed communication system. But even that was secondary to organic spread.
The user-customer split amplified this. In most SaaS businesses, the person using the product and the person paying for it are the same. In Parakeet Chat, they are not. Inmates use the product. Their families on the outside pay for it. This means there are two communities talking about the product: people inside who tell other inmates about it, and families outside who tell other families about it. Two parallel referral networks, neither of which requires Jordan to do anything.
This structure is worth noting carefully. It is not unique to prison tech. Any product where the beneficiary and the payer are different people has this same dual-network potential. Parents and children. Patients and insurance. Employees and HR departments. The question is whether the beneficiary's experience is good enough to generate advocacy on both sides of that divide.
In Parakeet Chat's case, it clearly was. Jordan's team has received messages from users describing conversations with daughters they had not spoken to in years. That kind of outcome generates a different quality of word of mouth than a productivity tool that saves someone twenty minutes a week.
The Numbers: What $300K Per Year Actually Looks Like
Parakeet Chat made just over $300,000 in 2025. Over its lifetime it has generated $1.5 million in total revenue. The pricing is $15 to $20 per month depending on the plan, with a discount for annual subscriptions. Roughly 30,000 people have used the product since launch, which represents approximately 20% of the entire US federal prison population. The platform has facilitated over 9 million messages.
These numbers are publicly verifiable on TrustMRR, a revenue transparency platform, which Jordan mentioned specifically for anyone who does not want to take his word for it.
For context: $300K per year from a solo-bootstrapped product, with no investors, no team, and no marketing spend, puts Parakeet Chat comfortably in the top tier of micro-SaaS businesses by revenue. Most solo SaaS products never reach $10,000 in monthly recurring revenue. Jordan's product has been running profitably since month two.
He also has not written code in six months. That is not laziness. That is what a well-built, low-churn, high-retention product looks like when it is serving a community that has consistent, ongoing need and no better alternative.
The Lesson: User and Customer Are Not the Same Person
Most startup advice treats these two things as synonymous. Find your user, understand their pain, build for them, charge them. That is the standard model and it works most of the time.
Parakeet Chat runs on a different structure, and that structure is one of the most interesting parts of the business. The user is the incarcerated person. The customer, the one paying $15 to $20 a month, is their family on the outside. Jordan is not selling to the person with the problem. He is selling to the person who loves the person with the problem.
This separation is what made growth possible in a closed ecosystem. Families outside can access the internet. They can find the product, sign up, pay for it, and activate it for their loved one inside. The incarcerated person does not need to do anything except use the email system they already have access to. The payment flow and the usage flow run through entirely separate channels.
When looking at other founder stories where growth happened without traditional marketing, this split between user and payer comes up more often than most people expect. The businesses that spread fastest inside tight communities are almost always ones where the value is felt by someone who cannot easily discover the product themselves, and paid for by someone who can.
The Framework: How to Find the Customer Nobody Is Building For
Jordan's story points to a repeatable way of thinking about market discovery. Not as a research exercise, but as a byproduct of genuine human connection with people who are normally invisible to founders.
The framework has four parts.
First: look for closed ecosystems. A closed ecosystem is any community, institution, or situation where people cannot easily reach the outside world to discover or evaluate products themselves. Prisons are the most extreme version. But nursing homes, rural areas with poor connectivity, specific professional environments with strict compliance rules, and tight ethnic or cultural communities with limited English-language product access are all versions of the same thing. These markets are invisible on Product Hunt not because the need does not exist, but because the people with the need cannot post about it.
Second: follow relationships, not research. Jordan did not discover this market by analyzing TAM data. He discovered it through correspondence with a friend. The most underserved markets in the world rarely show up in keyword research because the people inside them are not searching for solutions online. They are talking to people they trust. The only way to access that information is to be one of those people, or to know one of them well enough that they tell you honestly what is broken.
Third: separate the user from the customer early. Ask yourself: who benefits from this product, and who has the financial access and motivation to pay for it? In many underserved markets, these are different people. Building for the beneficiary and selling to the payer is not a workaround. It is often the only distribution model that works.
Fourth: let the constraint become the product. The prison ecosystem's constraints, no smartphones, no app downloads, no internet access, forced Jordan to build something that runs entirely through email. That turned out to be an advantage, not a limitation. Whenever a market constraint makes building harder, it also makes copying harder. The constraint that slowed Jordan down is the same constraint that protects him from competitors who would otherwise replicate the product in a weekend.
This is the same pattern that shows up in farming and agriculture startups built for underserved rural operators: the markets that look hardest to enter from the outside are often the ones with the most loyal customers once you are in.
What I Learned From This Startup Story
The detail that stays with me is not the revenue number. It is this: 20% of the entire US federal prison population has used Parakeet Chat. That is not a product metric. That is a penetration rate that most consumer apps would consider a wild success in a mainstream market. Jordan hit it in a market that does not have a single tech conference dedicated to it, no investor thesis written about it, and no "founder community" where people swap notes on what is working.
The uncomfortable truth underneath Jordan's story is that the best niche markets are often the ones that make people uncomfortable to talk about publicly. Prison is not a topic that gets startup founders excited at networking events. Nursing home residents are not a demographic that gets covered in TechCrunch. Undocumented immigrants, people with chronic illness, communities with low digital literacy. These are groups with real, expensive, persistent problems and almost no one building for them. The reason is not that the opportunity is too small. It is that the founder community largely does not have personal connections to those groups, and it is uncomfortable to build for people whose lives look very different from your own.
Jordan built for this market not because he read a report about it. He built for it because he kept in touch with a friend who had been sent away. That personal connection gave him information that no amount of market research could have surfaced. The willingness to maintain that connection across a difficult circumstance, paper letters, years of correspondence, is what made the business possible. Most founders are not willing to do that. That is the actual competitive advantage.
The honest caveat I want to leave here: building for a captive, underserved population comes with a weight that building a B2B productivity tool does not. The people using Parakeet Chat are not choosing between two good options. For many of them, it is the only meaningful tool they have access to. Jordan seems to understand this. The messages his team receives from users describing reconnected family relationships suggest the product is doing real good. But it is worth naming clearly: the most defensible markets in the world are often the ones where people have the fewest alternatives. Building in those spaces requires a level of care and honesty about your own motivations that the standard startup playbook does not address.
Key Takeaways
- Parakeet Chat hit $1.5M lifetime revenue by building for incarcerated people, a market almost no tech founder was paying attention to.
- Jordan's validation and MVP were the same step. In a closed ecosystem, you cannot validate without building first.
- The user (inmates) and the customer (families paying outside) are different people. This split created two parallel word-of-mouth networks and made growth possible without any traditional marketing.
- 200 paying users in month one. Profitable from month two. No investors, no marketing budget, no team.
- The constraints of the ecosystem, no app store, no UI, email-only, became the product's protection against copycats.
- The best underserved markets are often invisible to founders because the people inside them cannot post on Reddit or Product Hunt about what they need.
- Building for a captive, vulnerable population requires more ethical clarity than building a B2B tool. That does not mean avoiding it. It means going in with your eyes open.
Frequently Asked Questions
What is Parakeet Chat and how does it work?
Parakeet Chat is an AI communication and learning app for incarcerated people in the United States. It works through the internal prison email system: an inmate sends an email to a specific address, the bot processes the message, queries AI services like ChatGPT, and sends a reply back through the same system. There is no app to download and no interface to learn. From the user's perspective, it looks like any other email exchange. Families on the outside pay for the service at $15 to $20 per month.
How did Jordan get his first users?
Through direct contact with people he already knew inside the prison system. Because the ecosystem is closed, traditional digital marketing does not reach incarcerated people. Jordan spoke with his contacts, showed them what he was building, and got feedback before launch. Within the first month, the product had 200 paying users, almost entirely through word of mouth spreading inside the facility.
Who pays for Parakeet Chat if inmates have limited money?
Families on the outside pay for it. This is a key structural feature of the business: the user (the incarcerated person) and the customer (the paying family member) are different people. Families who want their loved ones to have access to legal research tools, AI assistance, and communication support pay the monthly subscription. This separation meant Jordan could build a payment system accessible to people with internet access and bank accounts, while the product itself works for people who have neither.
What does Parakeet Chat's revenue look like?
The business made just over $300,000 in revenue in 2025 and has generated $1.5 million in total lifetime revenue since launch. These figures are publicly listed on TrustMRR. Approximately 30,000 individuals have used the product in total, representing around 20% of the entire US federal prison population. The platform has facilitated over 9 million messages.
How do you find a niche like this as a founder?
Jordan did not find it through research. He found it through a sustained personal connection with someone inside the system who told him honestly what was broken. The practical lesson is to look for communities you already have some genuine connection to that are underserved by technology, particularly closed or semi-closed communities where people cannot easily discover or evaluate new products themselves. That friction is exactly what protects a product once it is built. You can also read about other digital product categories where underserved audiences are driving consistent revenue for context on the broader pattern.
Is Parakeet Chat still active?
Yes. As of Jordan's most recent interview, the product is live, profitable, and growing. Jordan himself has not written code for the product in several months. The AI tooling now handles development. The business continues to run largely on the momentum of word-of-mouth referrals and the referral credit system Jordan built into the product early on.
The markets worth building for are not always the ones being talked about. Sometimes they are the ones that nobody can talk about because the people inside them have no platform to speak from. That is not a warning. It is a map.