If you work in any field, everyone eventually dreams of owning their own business. If you’re a lawyer, you want your own law firm; if you’re a dentist, your own clinic. As software engineers, we’re no different—we’ll all want to build our own SaaS (Software as a Service) at some point. Believe me, I’ve done it twice already. I’ll build something, code for months… and then realize marketing is the less-colorful side of the party.

By now you’re probably wondering: What am I building next? You know me as the “Flutter guy” who builds full-stack apps, but this time I’m going all-in on web tech—and boy, do I have stories for you.

Buckle up, nerds… this is going to be a wild ride. I’ll be posting my journey every single day as I add features, squash bugs, and push updates to production. If you’re impatient, though, go ahead and check out v0.2 of the app—it’s live now at https://www.leadfinder.store/.

Before you grab your keyboard to drop comments, let me explain the name and the core idea behind the product.

Lead Finder

As the name suggests, Lead Finder is a platform that helps you discover business leads for any line of work. Imagine you just graduated from university and you’re hunting for an internship. You scroll through social media, sometimes spotting opportunities, sometimes hitting dead ends. It gets rough.

Maybe you’re a freelancer, or a doctor looking for patients. Gathering all that data and dumping it into a spreadsheet feels outdated and tedious. And spreadsheets go stale fast—leads change, companies pivot, contact info shifts.

You need a service that keeps things moving and surfaces the best leads for your field. Whether you’re a software developer, a law firm, a hospital—or anything in between—Lead Finder delivers fresh, actionable contacts right to your fingertips with a single click. That’s the power of this SaaS.

Back Story

The back story is very simple, I am a free lancer most of the time and I want to find lead using the internet, it was hard, going to different ites, search the thing, wrote down all the email, phone numbers and other into a excel sheet and then few time later all the data is out data so I though why not just I buid this stuff. it won’t tak that long, so I made quick react project, install the depiedences and then make basic UI and then I intergrated the api, I added the export all the data into a simple excel sheet and then bam, i have all the data, it was so simple and clean but then a thought hit my mind, If i am having this issue then other might be facing the same thing so I asked around in my community and I found out yeah, this is a problem and where their a problem, I have a solution and I asked chatgpt about this and he helped me push thing much better as well.

The Beginning

When I started coding this project, it was just a super simple one-page app using React and Vite. I integrated an API into a form, displayed the results, and then exported them into a basic CSV file. Straightforward stuff.

But then a thought hit me—what if I turn this into a SaaS? If I was facing this issue, chances are others were too. So I started thinking: what would the core features look like? I asked ChatGPT (of course), and after a long discussion, we agreed—yep, this is a real problem, and yep, it could become a real product.

After reflecting on everything, I came up with 7 key things I needed to build:

1. Get leads
2. Export leads into an Excel sheet
3. Payment gateway (Free & Pro tiers)
4. Auth system
5. Backend to manage everything
6. Simple admin panel (for me)
7. Set API usage limits

And just like that, the race was on.

I started coding and ticking off features one by one. I also added feature flags so I could update things easily and keep everything flexible and future-proof.

Tech Stack

Once the features were locked in, it was time to decide the tech stack. And if you’re expecting some wild fancy stack—nah, it’s pretty simple:

1. Vite
2. React
3. Tailwind CSS
4. Clerk
5. Convex
6. UI built with Redux + Shadcn

When I was building the project, I thought the stack was solid—and honestly, it still is. Sure, there were moments when I considered switching to a full framework like Next.js. But even without it, this stack is doing wonders. It’s fast, I have full control, and I can do a lot. It’s not perfect, but some things are just great—like blazing fast build times, everything is dynamic (no static content), and I can manage feature flags, blogs, and edge cases all in one flow.

Why I Didn’t Go with a Framework

When I first started building LeadFinder, it was just a single API call and some simple integrations. But as I added more features—routes, a blog, a more complex admin panel, feature flags, Convex data migration—I realized maybe I should’ve gone with a proper framework.

But by that time, I had already written over 20,000 lines of TypeScript and JavaScript. Rewriting the whole thing, swapping out Clerk, reworking Convex setup for dev and prod… nah, not worth it.

So I stuck with what I had.

And honestly? It’s working perfectly. With Vercel, the whole thing runs smooth. If I ever want to expand or refactor later, I will—but for now, this single-page app is getting the job done.

Pricing & Thought Process

After a long thinking session (and way too many coffee-fueled nights), I couldn’t decide on the price point. Initially, I thought $8/month for the Pro plan and $0 for the Free plan. But even then, I wasn’t fully convinced the value was there. So, I decided to add more features—stuff that would actually make it feel like a full platform, not just another lead scraper.

The idea? A one-tab solution. Search leads? Same tab. Save them? Same tab. Need help? AI’s right there. Want to reach out to the lead? Yup, same tab. No more hopping between 3 services or 10 open browser tabs.

Decided Features

With that in mind, here are the features I committed to:

1. AI integration (Gemini & OpenAI)
2. Email integration
3. In-app communication with leads
4. Full CRUD operations for both leads and emails

These felt like solid features that actually brought value—and yeah, now I was confident: $8/month Pro, $0 Free. Done.

The Real Cost

What really costs me isn’t just time—it’s API usage.

• Google Maps API for finding leads? Not cheap.
• AI integration using Gemini + OpenAI? Prices are through the roof.

So to balance it out, I stuck with the $8/month Pro plan. Right now, there are no API call limits for Pro users. Go wild—scrape and search all you want.

In the future, I might introduce more tiers or set some soft limits. But for now, this setup works.

The Future

Oh yeah—I’ve got plans. Big ones.

Here’s what’s on the roadmap:

1. Add more lead-finding APIs
2. Expand the admin panel (more control, more clarity)
3. Add more AI models and prompts
4. New lead-finder services like:
   • LinkedIn lead search
   • Facebook lead lookup
   • A Google scraping service to extract leads from search results
5. Clean up the codebase—make it more readable, scalable, and properly documented
6. Monitoring: I want to build an all-in-one logging & error handling setup inside the admin panel, so I’m not flying blind when stuff breaks

UI and Design

Alright—with all the features in place, let’s finally take a look at the UI and see how it’s all shaping up.

❮❯

1 / 13Fullscreen

As you can see, the UI runs in dark mode by default—but yeah, there’s a light mode too, with multiple accent color choices. You’ve got:

• A clean landing page
• A functional dashboard where users can connect Gmail
• An email module to send messages directly to leads
• A “Leads & AI” section where users can save leads and let the AI generate emails for them

Don’t know what to write? AI’s got your back—generate a cold email, tweak it if needed, and send it straight from the app. No switching tabs, no guessing what to say. It’s all there.

Coming Blog Posts

In the upcoming blog posts, I’ll break down everything—code structure, how I organized things, and even the parts where I hardcoded stuff that later came back to haunt me. 😅

To deal with those issues, I had to write a bunch of migration logic that works across both the dev and production environments. It wasn’t clean, but it worked—and I’ll share the exact thought process and trade-offs I made along the way.

Stay tuned if you’re interested in the messy reality behind building and shipping something solo.

Conclusion

This project has taught me a lot. From managing APIs to building out a scalable system, to realizing just how far I can push a simple Vite + React app—it’s been a ride.

And honestly, I’m super excited to see where this goes next.

This is just the beginning. 🚀