How I Turned Telegram Posts Into an SEO-Friendly Blog (and Why SSR Mattered)

I wanted a simple blog on my personal website where I could automatically repost my Telegram updates.

The goal wasn’t philosophical—it was very practical:

• Build my personal brand
• Capture potential SEO traffic
• Get a domain under my name properly indexed
• Publish everything automatically, without extra manual work

What I built (automation + GPT + two languages)

I connected GPT to the pipeline so it could generate meta tags automatically. I also added bilingual publishing and a basic admin panel where I can import posts from Telegram and manually edit them if needed.

At first, everything looked perfect. I shipped quickly and pushed the project to production.

The nuance that changed everything: SEO needs indexable pages

Then I ran into one important nuance.

I originally built the first version in Lovable, and the frontend was essentially React. For many products that’s totally fine, but for this particular job it didn’t make much sense. I wasn’t building an app—I was building a blog that search engines should index reliably.

That meant I needed SSR (server-side rendering) so bots could consistently “see” the content without relying on client-side JavaScript execution.

Migration: from a React-style frontend to Next.js SSR

I moved the project into Cursor, rewrote it with Next.js, and then validated the result by checking how bots were seeing the pages. After that, everything started working the way a blog should.

In the end, my stack looked like this:

• Lovable
• Cursor
• Next.js
• Vercel

Lovable for “vibe coding”: my real pros and cons

Here’s my experience after actually shipping something.

Pros

• You can build a working project very quickly (within reasonable limits)
• It’s enjoyable to work with
• Everything is assembled conveniently in one place
• For simple services, it feels amazing
• Sometimes it honestly feels like magic

Cons

• It’s expensive
• Tokens disappear very quickly
• As the project becomes more complex, it starts thinking longer, gets confused, and holds logic worse
• Small bug fixes can end up costing more than the first 80–90% of the project

A concrete example from my case: building the core relatively quickly cost around $10, and then another $40 went into polishing small details and fixing bugs.

Even so, with WordPress it still would have taken me longer—and probably annoyed me more.

Conclusion

Vibe coding is excellent for quick launches, experiments, and small personal products.

Not painless. Not free. But very fast—and very addictive.

This is clearly becoming a hobby for me. While my team focuses on serious, large-scale projects, I genuinely enjoy experimenting with small builds like this on the side.

I’m pretty sure I’ll ship more than one mini project like this.