My side project has no MRR after 10 years. Here's why I'm going all-in.
For over a decade, I've been working on a side project with no meaningful monthly recurring revenue (MRR). Most people would have quit by now. So why did I just commit to working on it full-time?
The idea sparked in 2011 from a simple frustration: I wanted to read "Journey to the Centre of the Earth" in its original French, but my skills weren't good enough. The English translation was right there.
Why couldn't I just "layer" the two languages on top and toggle between them when I got stuck?
That question led to a ten-year journey of building, learning, and wrestling with a problem I couldn't let go of. It even helped me land jobs, learn new programming languages and frameworks along the way.
The content bottleneck
One of the earliest versions was a native iOS app that received encouraging reviews. It wasn't much, but in-app purchases brought in some revenue. Requests poured in for an Android version and more stories, but every new reader highlighted the same bottleneck: fresh, high-quality bilingual content.
I built a CMS for bilingual authors to publish their work, with the majority of the revenue going to them. However, creating high-quality bilingual stories was not easy. Each author could realistically only cover one or two languages, and topics are taste-dependent.
Quality also required policing. While some authors were amazing, others were less diligent. I constantly worried that if sales didn't ramp up, authors would feel cheated.
Signals that kept me going
- Reader pull: Language learners requested more stories and new language pairs, even when I had nothing new to sell them.
- Competitive gap: Other implementations of the same idea have surfaced, but none have truly cracked distribution.
- Personal stake: My own French and German skills are dwindling, and I’m adding Portuguese to the mix. I need this tool as much as the audience does.
AI removes the content bottleneck
Large language models changed the equation. Instead of churning out auto-generated sludge, I’m using them to curate classics. Think fables and public-domain stories adapted to specific proficiency levels. Each piece is rewritten with careful prompts, tagged with reliable metadata, and reviewed before shipping.
The result is hundreds of bilingual stories tuned for beginner and intermediate readers. Tap once and the French paragraph becomes English (or vice versa). The reading flow stays intact, and that little “aha” moment, the whole reason Mytoori exists, finally scales.
Content curation over AI creation. That’s the guardrail that keeps the library both growing and trustworthy.
The bet I’m placing
I’ve given myself a 3-5 month runway with some possibility for extension. Two concrete milestones will decide whether I keep pushing:
- The free version of the app should see over 500 real Daily Active Users (DAU)
- Positive customer feedback and intent to purchase subscription once ready
If I miss either target when the runway closes, I’ll revisit the plan with cold eyes. Until then, it’s full focus on hitting both.
How to monetise
Self-learners are the primary target audience. The core reading experience stays free, but a subscription (~€10/month) unlocks premium features:
- Audio features, inline vocab notes, spaced-review flashcards
- Progress dashboards that show reading streaks and vocabulary absorption.
On the side, I’m exploring collaborations with language programs. A lightweight LMS integration could allow single sign-on (SSO) for class assignments and shared analytics. Schools would be able to layer Mytoori into their curriculum without replacing their existing tools.
What I need from you
If you’ve acquired self-serve subscribers in language learning or education, I’d love to hear how you cracked (channels, pricing experiments, landing pages that actually converted). And for anyone who has sold into schools: what should I watch out for when integrating with LMS platforms?
About Mytoori.com
It's a bilingual language learning platform. You read a story in the language you're learning. The translation is readily available.
Take a look at the current library here: mytoori.com/library. After a decade in the wilderness, I’m finally riding this thing somewhere. Now I just have to prove it can carry the weight.
Ten years is serious commitment. The AI curation angle makes sense — using it to adapt classics rather than generate generic content keeps the quality bar high while solving your bottleneck.
Your milestones are smart and concrete. 500 DAU is ambitious but doable if you nail distribution. One thought: have you considered partnering with language learning communities (Reddit, Discord servers, etc.) early? They could give you fast feedback and your first power users before you hit the monetization phase.
Rooting for you on this push.
Inspiring journey, Giwan! Love how you tackled the content bottleneck with AI while keeping quality high. Persistence really pays off. This resonates with me—at Claremont Roof Care, we also focus on long-term reliability and serving real needs rather than quick wins. Excited to see Mytoori grow!
Ten years is remarkable persistence. The content bottleneck was a real problem, but using AI to curate classics rather than generate slop is a smart approach. Your milestones are clear and realistic - 500 DAU is ambitious but achievable with the right distribution strategy. Best of luck with the 3-5 month push.