I accidentally found a growth channel for my AI tattoo app, then turned it into an Etsy revenue stream
I built TattooAI.net with a pretty simple assumption:
Social media would be one of my main growth channels.
Instead of manually posting every day, I built an automation that picked tattoo designs I liked from the platform and automatically shared them to Pinterest, Instagram, Facebook and X.
I expected some traffic from all of them.
That didn’t happen.
Most platforms did almost nothing.
Pinterest was completely different.
I didn’t even pay much attention to it at first. The automation was running in the background and I was focused on building the product.
At some point, I noticed that the Pinterest account had accumulated a lot of notifications.
When I finally checked the analytics, I realized it had grown to around 2 million monthly impressions.
That was my first signal that something interesting was happening.
Then something much stranger happened.
Under one of the Pinterest posts, a woman shared a photo showing that she had actually gotten the tattoo on her leg.
That changed how I looked at the traffic.
These weren’t just impressions or random social media engagement.
Some people were clearly using Pinterest to actually discover tattoo ideas.
I later learned that Pinterest is a much bigger discovery channel in the tattoo space than I had realized.
At that point I had another problem:
How do I turn millions of impressions into revenue?
A friend suggested Etsy.
So I tried a very simple experiment.
Instead of creating a completely new product catalog, I looked at the tattoo designs already getting the most attention on Pinterest.
I selected roughly 30 of the strongest-performing designs and listed downloadable tattoo designs/stencils on Etsy.
Then, once again, I mostly left it alone.
No serious Etsy strategy.
No ads.
No daily optimization.
It just sat there.
And slowly, sales started coming in.
So far, the numbers are still small, but interesting:
- 19.86M impressions
- 51.96K saves
- 4.05K outbound clicks
Etsy
- 24.8K visits
- 27 orders
- 4,469 TL revenue
This is obviously not a life-changing revenue stream.
But I think the experiment is interesting for a different reason.
I originally thought I would decide which social platform to grow on.
Instead, the market decided for me.
I didn’t choose Pinterest as the distribution channel. The users did.
And Etsy came later as a way to monetize the signal Pinterest was already giving me.
This week I went back to the project and did something I probably should have done much earlier.
I reviewed the Etsy listings one by one.
I removed products that were getting views but no meaningful interest.
I improved titles and descriptions on some of the existing listings.
And instead of guessing which new products to add, I went back to the Pinterest data.
I looked at:
- impressions
- saves
- outbound clicks
- which individual designs were still getting attention
Then I added a few new Etsy listings based on those signals.
The store now has 28 active listings.
Now I want to see whether this improves conversion and revenue over the next few weeks.
There are a few lessons I’m taking away from this experiment.
Automation can create optionality.
The Pinterest automation took some time to build, but after that it kept running while I focused on other things.
Months later, it surfaced an opportunity I wasn’t actively looking for.
Distribution channels can reveal themselves.
I could have spent weeks trying to grow Instagram or X because they felt like the obvious choices.
The data said otherwise.
Engagement is useful, but intent is much more interesting.
Millions of impressions were nice.
Seeing someone actually tattoo one of the designs was a much stronger signal.
And eventually, seeing which Pinterest designs generated saves and clicks gave me a much better way to decide what to put on Etsy.
You don’t always need to build a new product.
Sometimes the opportunity is just sitting between two things you already have.
In my case:
TattooAI -> Pinterest -> Etsy
None of this was part of the original plan.
That’s probably why I like the experiment so much.
I’ll share another update once I have enough data to see whether the latest Etsy optimization actually changes the numbers.
If you want to see the project itself: TattooAI.net
And for anyone curious about the actual experiment, here are the receipts:
Pinterest: https://www.pinterest.com/TattooAIGenerator/
Etsy: https://www.etsy.com/shop/AITattooDesigns
The gap between 51.96K saves and 27 orders is the key signal. Have you compared conversion by design, or are you still optimizing from engagement?
Not rigorously yet, the first pass was mostly engagement-led.
One important caveat is that many of the Pinterest pins weren’t linking directly to Etsy at all. Some were pointing to Instagram or TattooAI, so there was a lot of funnel leakage between “interest” and “purchase.”
I’m fixing that now: for designs that already have an Etsy listing, I’m pointing the Pinterest pin directly to that exact product.
The next step is to compare each design across:
saves / impressions -> outbound clicks -> Etsy views -> orders.
With only 27 orders, the dataset is still too small to make strong claims, but that’s exactly the metric chain I want to track next.
That metric chain should give you much cleaner evidence than optimizing from saves alone. If you’re open to it, what’s the best email to reach you on?
Letting the market pick the channel instead of forcing Instagram/X is the real win here. The tattoo-on-a-leg photo is a much cleaner intent signal than impression vanity metrics. Ranking Etsy listings off save rate (and outbound clicks when volume allows) feels right — impressions tell you what got shown; saves tell you what someone wanted to keep.
Exactly. That tattoo-on-the-leg photo was probably the moment I stopped looking at Pinterest as “traffic” and started seeing it as intent.
I also like your framing: impressions tell me what Pinterest distributed, saves tell me what people wanted to keep.
I’m going to start ranking future Etsy candidates by save rate first, then use outbound clicks as the stronger commercial signal when there’s enough volume.
The interesting part will be seeing whether those Pinterest signals actually correlate with Etsy sales over time.
Really like the "the market chose the channel" framing. We're learning a smaller version of this with Noviora (a real-life-as-an-RPG app): nobody searches for "gamified habits", they search for concrete things like "what to cook with what's in my fridge", so we're now building around those intents instead of our own category name.
Curious about one thing in your data: when you picked the ~30 designs for Etsy, did saves predict sales better than impressions? With ~52K saves vs ~4K outbound clicks, saves look like the stronger "I actually want this" signal. I'd be tempted to rank new listings by saves per impression rather than raw reach.
That’s a really good point.
When I picked the original ~30 designs, I didn’t have a formal scoring model — I mostly selected from the designs getting the strongest overall attention on Pinterest.
Looking at the data now, I agree that saves are probably a much better intent signal than raw impressions. Outbound clicks may be even closer to commercial intent, but the volume is much smaller.
I’m actually starting to test this now: looking at save rate + outbound click rate, then comparing those designs against Etsy views and sales.
The Etsy dataset is still too small for me to claim that saves predict sales yet, but if it grows enough, that would be a really interesting follow-up analysis to share.