I built a free Telegram bot for personal finance. Here's what 6 months of building taught me
Hey IH! Wanted to share a milestone and honest lessons from building NotiFin - a personal finance bot inside Telegram.
The origin story
I've been into personal finance for 15 years - no formal education, just genuine obsession. I use credit cards strategically: pay for everything with credit, keep my cash working in investments, pay off cards at end of grace period. It works great... until you forget a due date and the bank charges you a fee they secretly love charging.
After one late fee too many, I built a bot that reminds me properly - not one weak notification like banks send, but an escalating system that actually gets the job done.
What I learned
• Lesson 1: Solve YOUR problem first. I built this for myself. Turns out a lot of people share the same pain.
• Lesson 2: Telegram is underrated. A billion users. Built-in notifications. Zero friction. No app store review process.
• Lesson 3: "Boring" features win. The Friday 3 PM weekend warning is what people mention most. Not the AI, not the charts. A well-timed reminder.
• Lesson 4: Trust is everything in fintech. People are rightfully skeptical. So we don't connect to banks, don't collect card numbers, don't store sensitive data. Manual entry only. And it's completely free - no premium tier.
• Lesson 5: Localization matters early. 4 languages from day one (EN, RU, ES, TR) opened up audiences I wouldn't have reached.
Honest numbers
• Users who get reminders complete 73% of planned transactions (vs 41% without)
• Friday warnings increased same-day completion by 2.3x
• 4 supported languages
• Zero paid marketing - all organic
• Revenue: $0 (it's free, deliberately)
t.me/notifinofficialbot - would love your feedback.
Happy to answer questions about building Telegram bots or the fintech space!
The localization insight hit close to home. I support Telegram as one of seven channels on my platform and the user behavior differences across regions are surprising — what counts as "intuitive" UX in one market is friction in another.
Your point about Telegram being underrated is underappreciated in the Western builder community. A billion users, push notifications that actually work, and no App Store gatekeeping — it's genuinely the fastest channel to ship on. The bot ecosystem is mature enough that you can go from idea to deployed in a day.
The 73% vs 41% completion stat is the kind of before/after number that makes a case study. Have you considered writing up the behavioral design decisions (like the Friday 3pm timing) as a standalone post? That's the part of this story that could get real traction.
Thanks for the thoughtful feedback! The completion rate difference (73% vs 41%) was one of those "wait, this actually works" moments. You're right about the behavioral design angle - the Friday 3pm timing, the three-tier escalation, timezone-first thinking - these weren't obvious decisions, and the reasoning behind them could be valuable.
I'll dig into writing that up as a standalone post. Appreciate the push to focus on the "why" behind the numbers, not just the "what."
Also curious about your seven-channel platform - how do you handle localization at scale across different messaging ecosystems?
That Friday 3pm weekend warning is such smart behavioral design. You're catching people right at that mental transition to weekend mode when they still have the bandwidth to actually do something about it. Perfect timing beats perfect technology.
The trust-first approach in fintech resonates hard. I've seen so many finance apps die because they asked for too much data too early. Manual entry feels like friction, but it's actually the feature that builds confidence.
Telegram really is underrated for this kind of utility. No app store gatekeepers, built-in notifications that actually work, and the bot ecosystem is surprisingly mature. The 4 languages from day one thing is interesting too - most people think localization is a 'later' problem, but sounds like it opened doors you wouldn't have found otherwise. Which market surprised you most?
Friday 3pm was genuinely just "when would I want this?" - sometimes being your own user is the best research.
Most surprising market? Russian speakers outside Russia - expats, Kazakhstan, Georgia, Turkiye, Europe. Way more engaged than I expected. The diaspora use case wasn't on my radar at all.
Love this — messaging apps are such underrated distribution channels. "Boring" features that solve real problems always win over flashy tech.
The Friday 3 PM weekend warning is brilliant behavioral design. You're not just reminding people about bills — you're catching them at the moment they're mentally transitioning to "weekend mode" but still have the capacity to act. That's the difference between a notification and a nudge that actually works.
I'm building an AI secretary on WhatsApp and the lesson about trust in fintech hits hard. We're also taking a "no data storage" approach — people shouldn't have to trade privacy for convenience.
Curious: have you thought about monetization, or are you keeping it free as a deliberate trust-building strategy? I've seen some bots do "pay what you want" for optional features and it works surprisingly well for goodwill.
"Boring features that solve real problems" - stealing this :D
No-storage approach is the right call. People don't read privacy policies but they feel trust.
Monetization is intentionally on pause for now. Pay-what-you-want only works when people feel grateful, not just dependent - still building toward that.
Have you tried the bot? Any feedback is highly appreciated!