How I Started A $800/Month AI Automation And Website Studio For Small Businesses
Who are you and what business did you start?
Hey, I'm Saad. I run Agentyra AI, a small studio in Lahore that helps local businesses stop losing customers they've already paid to get.
In practice that means a few things. We set up AI assistants on WhatsApp and on the phone that reply to customers any time of day and book them in. We build automations in n8n that chase up leads nobody followed up on. And we do the websites and SEO that bring people in to begin with. Most of our clients are clinics, law firms, online stores and consultants.
I started in September 2025. Right now we're at about $800 a month, and there are seven of us including me.
The Agentyra AI homepage
How do you come up with the idea for Agentyra AI?
Honestly, it wasn't one big idea. Two things just kept happening.
I was freelancing, doing websites, SEO and ads, and clients started asking me if "the AI stuff" could take some work off their hands. Replying to messages, following up, booking appointments, that kind of thing.
The other thing was a pattern I couldn't unsee. A business would pay for ads, the ads would work, someone would call or send a WhatsApp, and then... nothing. Nobody replied for hours. By evening that person had already gone to someone else.
That's when it clicked for me that a lot of these businesses didn't need more leads. They needed someone to answer the ones they had. And by then tools like n8n, Vapi and the WhatsApp Cloud API had got good enough, and cheap enough, that a small clinic could actually afford something that replies at 2 am. I already knew how these businesses got their customers from the website side, so building the part that responds felt like the obvious next step.
How did you launch Agentyra AI and get initial traction?
There wasn't really a launch. I told people I knew what I was doing now and posted about it on LinkedIn.
My first paying client came through a referral, someone in my network, and it happened inside the first month. Funny thing is, it wasn't for AI at all. They wanted a website.
I learned a lot from that. People pay for what they already understand. The website was the thing they knew they needed, and once they trusted us with that, talking about automation was much easier. I'd do it the same way again.
What was the growth strategy for Agentyra AI and how did you scale?
Almost all of our clients so far have come from two places: referrals and LinkedIn (plus a bit of other social).
Referrals are the big one. Business owners here talk to each other a lot, and when someone's new website starts bringing in calls, or their WhatsApp bot books an appointment while they're asleep, they tell people. The person who gets referred usually shows up already halfway convinced, which makes my job a lot easier.
LinkedIn works for a similar reason. I post things I've actually built, like a bot replying to a customer at midnight, a booking flow, or a website before and after. Most small business owners have never seen an AI assistant in action, so a short clip of one working does more than any explanation I could write.
I'm also putting time into SEO and writing guides on our site, things like how to automate lead follow-up or what an AI assistant actually costs. Our domain is brand new, so that's slow and hasn't brought in much yet. But it's still useful. When a prospect asks me something, I can send them an article instead of a long voice note.
The newest thing I'm trying is cold email to businesses in Lahore. I don't send a generic pitch. I go through their website first, find two or three real problems (placeholder text left in, a booking link that goes nowhere, that kind of thing), point them out and offer to fix one for free. It takes ages to write each one, but people reply. A dental clinic said yes to the free fix within a few days.
If I had one tip on growth, it'd be to show people the thing working. A 30-second demo beats any pitch I've written.
What were the biggest lessons learned from building Agentyra AI?
The hardest part for me has been starting with a website nobody had heard of. No traffic, no reviews, no links. Google really doesn't care how good you are until other sites vouch for you, and I didn't realise how long that takes. If I was starting over, I'd start writing content and collecting reviews and listings from the first week, long before I actually needed the traffic.
The thing that saved me was not betting everything on SEO. Referrals and LinkedIn kept clients coming in while the site slowly builds up.
What I'd tell anyone running a small business: before you spend more on getting leads, look at how fast you reply to the ones you already get. I've seen it with almost every client, and it was true for my own business too.
Discover Similar Business Ideas Like Agentyra AI
More about Agentyra AI:
Who is the owner of Agentyra AI?
Saad ali is the founder of Agentyra AI.
When did Saad ali start Agentyra AI?
2025
How much money has Saad ali made from Agentyra AI?
Saad ali started the business in 2025, and currently makes an average of $9.6K/year.