Ran AI video ads for my dad's dental clinic for free, accidentally turned it into a ~$3k/month side thing. AMA
Started this about 3 months ago. My dad runs a dental clinic and his usual ad guy was charging him $1500 for a single Instagram video. I'd been messing with AI video tools and figured I'd try to make him something for free.
The first one was rough but it got more bookings than the $1500 video had. So I made him a couple more. Then his friend who owns a med spa asked if I'd do one for her. Then it kind of kept going.
Here's one of the recent ones (this one ran on Meta last month for a local pizza place):
https://reddit.com/link/1tj8251/video/gnahdfl5re2h1/player
Where I am now:
- About to cross $3k MRR this month
- 7 ongoing clients on monthly retainers, one in trial
- Pricing: $600/mo for 5 ads, $1000/mo for 10 ads. One-off ads are $250-300 but I push everyone to monthly.
- Costs are maybe $50-80/mo in tools, the rest is mine
- Honestly I think I should be at $5-6k by now and I'm leaving money on the table by being slow on outreach. Sales is the bottleneck, not production.
I'll get this out of the way because Reddit will bring it up otherwise. Yes, some of these ads look obviously AI. That's actually fine for the businesses I work with. Their previous ads were stock footage of a different dental office in a different country. AI video that's specifically about their location, their service, their offer outperforms generic stock every time, even when you can tell it's AI. I'm not pretending it's the next Spielberg. It's a $30 ad that books appointments.
What I've learned:
The hard part is sales, not production. I can make an ad in like 90 minutes. Finding a business that will say yes takes way longer. I cold-message local businesses I find on Google Maps. About 1 in 15 of those replies, maybe 1 in 30 closes.
Lead with a free spec ad. I make the ad first, then send it to them. "Hey, made this for your business, here's the link." Conversion on this is way higher than pitching the service abstractly. Most of my closed clients said yes because they already had the ad in their hand.
Go for businesses that already advertise. Don't try to convince a small business to start running paid ads. Find the ones who are already running mediocre ads and offer to replace them. The pitch writes itself.
Tools I use:
- Google Maps for finding local businesses and grabbing images of their storefront, food, interior, whatever I need as a reference
- For the ads themselves I use bonzi studio because it has all the video models in one place and lets me try different ones for the same scene. Was on Runway before that. Honestly any of them work, the tool isn't the moat.
- Capcut for final edits
That's the whole stack. People assume there's a complicated workflow behind it but it's three tools.
The actual workflow inside Bonzi for one of these ads looks like this:
Each ad is basically: pick a hook scene, lock the location/product reference, generate 3-4 cuts that build on each other, then assemble in Capcut and a CTA card at the end. Sounds more complicated than it is. Once you've done a few you're mostly just swapping the hook and the offer.
Cons / what I'd do differently:
- I undercharged for the first month. Should have been at $600/mo from the start.
- I didn't get contracts signed for the first few clients. One ghosted after I delivered. Now everything is invoiced upfront.
- I tried scaling by hiring a friend to help. Quality dropped. Went back to doing it myself.
- I'm still bad at outbound volume. If I sent 50 cold messages a day instead of 10 I'd probably be at $6k MRR by now. Working on it.
Not passive income. It's a service business and you have to do client work. But the production speed of AI lets one person handle way more clients than a traditional video freelancer, which is where the margin lives.
Happy to answer specifics on outreach scripts, pricing, the ad style itself, anything.