I'm not a developer. I want to say that upfront, because it matters for everything that follows.
This isn't a polished "here's my SaaS" post. This is the real version — the version where I almost gave up twice, broke my own security without knowing it, got rejected by every directory I tried to use before I built my own, and still ended up shipping something real. If you've ever stared at a blank editor with zero idea what you're doing and still hit publish anyway — this one's for you.
A few months ago I was doing something completely ordinary — looking for a WhatsApp group to join, something niche, nothing dramatic. I typed the usual search, clicked the first "directory" result, found a group that looked right, tapped the invite link...
And landed on an expired invite. Group full, or deleted, or the link just dead.
Annoying, but not unusual. So I tried the next result. Same thing. And the one after that. Every single "directory" I clicked through was a graveyard of broken links — pages that had clearly been thrown together once, stuffed with invite URLs, and never touched again. Nobody was maintaining any of it. The internet was full of these lists, and every one of them had quietly rotted.
At some point, sitting there frustrated for the fifth time in a row, the thought that actually stuck wasn't "this is annoying." It was: why does every single one of these have the exact same flaw?
The decision that scared me a little
Here's the part I want to be honest about: I had no coding background. Not "a little rusty," not "did a bootcamp once" — genuinely nothing. The idea of "I'll just build the thing myself" felt almost embarrassing to say out loud. Like, who was I to think I could build a working website, with a database, with search and filtering, when I'd never written a line of real code in my life?
But the problem kept nagging at me. It was so simple to describe — a directory where dead links actually get removed — and yet apparently hard enough that nobody had bothered to solve it properly. So I decided I'd at least try. Worst case, I'd learn something and quietly abandon it like a hundred other "side projects" people start and never finish.
That's basically how Groupverse started. Not as a startup idea. As stubbornness.
The moment that actually tipped me over
Before I even started building, I tried something small — I had a WhatsApp Channel (the newer, one-way broadcast format WhatsApp introduced) and I wanted to list it somewhere, just to see how these directories actually worked from the submitter's side.
I couldn't find a single directory that would take it.
Not "wouldn't approve it" — literally couldn't submit it. The submission forms were all built years earlier, before WhatsApp Channels even existed as a feature. Every form only had fields for "WhatsApp Group" or "Telegram." No option for a Channel link at all. I tried half a dozen directories and got the same wall every time — forms frozen in time, never updated for a format WhatsApp had already rolled out to everyone.
That was the moment the annoyance turned into something more specific than "these directories have dead links." It became: these directories aren't just poorly maintained, they're structurally outdated. Nobody had gone back and touched the actual data model behind these sites in years. Adding a new platform type wasn't a five-minute fix for them — it probably meant touching a database schema, a backend, an admin panel, all things whoever built the original site had likely walked away from long ago.
That's the exact gap I decided to build into Groupverse from day one — WhatsApp Groups, WhatsApp Channels, Telegram Groups, and Telegram Channels, all as first-class types from the start, not bolted on as an afterthought.
The early days: awkward, slow, humbling
I leaned on Supabase for the backend — Postgres database, auth, all the pieces I didn't understand yet but that seemed to be the path of least resistance for someone starting from zero. Every small thing took forever. Things that I now know are trivial — setting up a table, writing a basic query, hooking up a form to actually save data — ate entire evenings.
I remember the specific feeling of getting the first group listing to actually save into the database and show up on the page. It was such a small thing. A tiny CRUD operation that any junior dev would call boring. But I sat there staring at that one listing on the screen like I'd built something enormous. It genuinely felt like a small miracle, because two weeks earlier I couldn't have told you what a database even really was.
The mistakes that taught me the most
I want to talk about the two moments that actually hurt, because I think that's the part people skip when they tell these stories.
The sitemap issue. For weeks, I couldn't figure out why the site wasn't showing up anywhere in search. I'd built this thing, it worked, people could technically use it — but nobody could find it, because nobody was searching for a URL they'd never heard of, and Google wasn't indexing it properly either. I didn't even know what a sitemap was supposed to look like at first. I just knew something was wrong and I couldn't see it. That was a quietly demoralizing stretch — building something and having it be functionally invisible.
The security hole. This one scared me more. I was deep into learning Supabase's Row Level Security — the rules that control who can read or write what data — and at some point I realized my policies weren't actually locked down the way I assumed they were. In plain terms: there was a window where data that should've been private wasn't properly protected. Nothing catastrophic happened, but the moment I understood what I'd missed, my stomach dropped. It's one thing to ship a broken feature. It's another to realize you might have exposed people's data because you didn't fully understand the tool you were using. I went back, rebuilt the policies properly, tested them like a paranoid person, and it became the single biggest lesson of the whole project: working and secure are not the same thing, and as a non-coder, that gap is exactly where you're most vulnerable, because you don't know what you don't know.
I won't pretend that was a fun day. It genuinely made me question whether I was in over my head.
Funnily enough, that fear connected back to the exact thing that got me started. The old directories that couldn't accept WhatsApp Channels weren't just outdated — they were probably also sitting on years of unpatched, unreviewed backend code nobody wanted to touch anymore. Once you've felt what it's like to find a real hole in your own security setup, you start to understand why those sites never got updated. Going back into old infrastructure is scary even when you built it. I didn't want Groupverse to become one of those frozen-in-time sites in two years, so I made a point of designing the data model to be flexible from the start — new platform types, new fields, new categories should be additions, not rewrites.
Building back up
But I kept going, mostly because at that point I'd already learned more than I expected to, and stopping felt worse than pushing through.
Piece by piece, the actual product started coming together:
Filtering by category — eventually growing to 45+ categories, because "group directory" turned out to mean wildly different things to different people
Country and language filters, including regional Indian languages like Telugu, Marathi, and Konkani that I noticed literally no other directory bothered to support
A submission review flow, so new listings don't just dump straight onto the site unchecked
And the feature the whole thing was actually built around: a report button on every single listing, so when a link dies, it doesn't just sit there rotting for years like every directory that started this whole frustration in the first place
Then there's the part I didn't plan for at all: games. I added one small in-browser game as a throwaway experiment — something to click while a page loaded, nothing serious. Except building it broke my brain a little, in a good way. Everything else in the app up to that point was forms and lists and database reads. Games meant real interactive state, live on the page, no page reload, no backend round-trip. It was the first time I built something on the frontend that felt alive instead of just functional. I got hooked. One game became a small collection — a proper games section, sitting right on the homepage, not hidden away. It has nothing to do with group links directly, and I kept it that way on purpose. It's there because building it taught me more about real frontend engineering in a few days than months of CRUD screens did, and because I wanted one part of the site that was just fun, for me and for anyone browsing.
And then came the part that made this stop feeling like a hobby project and start feeling like a real product: premium. I added a subscription layer — verified badges, priority review, sponsored channel placement, bulk submissions — not because I had some grand monetization plan on day one, but because once real users started submitting real groups, some of them started asking for more: faster review, more visibility, a badge that signals trust. Building that meant learning an entirely new category of problem — payment flows, plan tiers, access control tied to a paying user instead of just any user. It's still early and small, but the first time someone actually paid for a plan on something I built with my own hands, alone, with no coding background eight months earlier — that hit differently than any line of code ever had.
And through all of it, the report button stayed the actual center of gravity. Games are fun. Premium is validating. But the report button is why this exists at all — it's the one feature directly answering the exact frustration that started this whole thing. Every dead link that gets reported and pulled is the loop closing on the problem I set out to fix. Everything else got built around that core, not instead of it.
Where it stands now
Groupverse (groupverse.co.in) is live, free, no sign-up required. It's still small. It's still rough in places. I'm still fixing things I find, and I still catch myself unsure about something that a real developer would consider obvious. That hasn't gone away, and I don't think it will for a while.
But it works. People are submitting groups. Dead links are getting reported and pulled instead of piling up. The thing I was annoyed about, months ago, clicking through dead invite after dead invite — that specific problem, for this specific product, is actually solved now. I built the thing I wished existed.
What's next
This isn't going to be the last thing I build. If anything, going from "I don't know what a sitemap is" to "I found and patched my own security vulnerability" in one project has changed how I think about what's possible for someone starting with literally nothing. The plan now is to keep improving Groupverse — better moderation tooling, more categories, probably a proper API at some point — and to keep building other things once this is in a place I'm proud of.
If you're sitting on an idea and the only thing stopping you is "I don't know how to code" — I'm not going to pretend it's easy. It wasn't. There were real moments of feeling out of my depth, a real security mistake that scared me, weeks of being invisible on search with no idea why, directories that flat-out couldn't accept what I was trying to submit because their whole foundation had been left to rot. But none of it was actually unsolvable. It just required sitting with the discomfort of not knowing, and fixing one broken thing — a sitemap, a security policy, a missing feature — at a time, until one day the thing you were scared to start is the thing people are actually using and paying for.
If you want to see what came out of all that: groupverse.co.in — filters, live listings, a report button that actually does something, a small games corner that has nothing to do with any of it, and a premium layer that exists because real people asked for it.
If you're a dev reading this and thinking "I could've built this in a weekend" — you're probably right, and that's exactly the point. I didn't have that weekend of experience. I had stubbornness and a lot of broken evenings instead. Genuinely open to feedback, questions about the stack, security holes you can already spot that I haven't found yet, or just hearing from anyone else building their first thing with the same "I have no idea what I'm doing" energy I started with.
Top comments (3)
The launch story is interesting. The part I would watch after this is moderation cost. Directories look simple until stale groups, spam, and duplicate listings become the actual product work. Did you set a rule for when a group gets removed?
The practical question for agent workflows is where the boundary lives: what the agent may decide, what it may only propose, and what evidence it leaves for review.
Nice article, I appreciate the concrete angle. The small changes section is what I will actually take away. What would you try next with this?