# AI dried up a lot of my client work, so I started building small sites and selling them. $7,405 sold so far. — retained record

> Retained reddit record from reddit:EntrepreneurRideAlong.

Canonical: https://highsignal.app/data/records/2053c58758507c53

[← sources](https://highsignal.app/data)


source: [reddit](https://highsignal.app/data/reddit) (reddit:EntrepreneurRideAlong)  published 2026-10-08 16:45 ingested 2026-10-09 02:36

[original source ↗](https://www.reddit.com/r/EntrepreneurRideAlong/comments/1x0vyjo/ai_dried_up_a_lot_of_my_client_work_so_i_started/)

## retained text



```
I've spent about twenty years working as a freelance consultant and developer. This year the work has really dried up.

Clients started coming back with AI-generated answers, and then I suspect they just started getting AI to do the work themselves. Hard to blame them really.

To generate some cash I started selling digital assets from my portfolio.

For a few years I'd been building local service directories, niche job boards and launch platforms off a repeatable codebase — the same engine pointed at a new niche rather than a fresh build each time.

I listed some of them, mostly on Flippa, not really expecting much.

They sold.

* MobileDetailing.io — $2,300
* WildlifeControl.io — $1,000
* HVAC USA — $1,000
* WindowCleaners.io — $605
* AppLauncher.io — $2,500

That's **$7,405 in sales**, and most of them had zero revenue.

The funny thing is that selling them actually opened the door to more client work and projects, which was great. So it wasn't just the money from the sites themselves.

It also made me realise I'd accidentally built something quite useful.

The apps are basically plug and play. I can take one of the codebases, point it at a new niche, configure it and have a new site launched and populated in roughly a day.

They're designed to be heavily automated, too. There are proper admin panels for everything, but I also built Telegram controls for the things I need to do regularly.

That makes a surprisingly big difference when you're running a bunch of sites. I don't want to log into six different admin panels just to approve a listing or check what's happening.

It's not going to replace my main work, but it's turned into a pretty nice side income and, more importantly, a way of building things I actually own rather than always building for clients.

Once I had a few other people asking about the code, I decided to package it properly rather than keep sending people bits and pieces.

So I've turned the codebases into **DirectoryLab -** a small community around building and running these kinds of niche sites.

There are currently three:

**Local Lab** — local service directories  
**Job Lab** — niche job boards  
**Launch Lab** — niche launch platforms

They're the same basic engines I've used to build the sites above.

I'm still figuring out where this goes. Maybe it stays a side project, maybe it turns into something bigger.

But after twenty years of selling my time, there's something quite satisfying about building an engine once and then using it to create multiple things.

  
If anyone's interested in learning more you can see the live sites and docs on directorylab.io 
```



document fetched 2026-10-09 00:26 · canonical https://reddit.com/r/EntrepreneurRideAlong/comments/1x0vyjo/ai_dried_up_a_lot_of_my_client_work_so_i_started
