Garden Metal Art. When High Clicks Don't Mean Sales
High CTR, Low Sales in a Beautiful Niche
Context
Small ecommerce brand turning vintage silverware into handcrafted garden art. Products were visually striking and sold in small quantities on Etsy. The founder wanted to see if Facebook ads could scale it into a meaningful direct-to-consumer business.
Problem
Budget was tight, roughly $10K total across three structured test rounds. The site was weak and conversion-unfriendly. Ads delivered some of the highest click-through rates I had ever seen (3% to 8%+), but purchases simply did not follow.
Approach
We treated each around $3K test round as a distinct experiment: new account structures, different audiences, fresh creative angles, and clear hypotheses about who might actually buy. We watched the full-funnel metrics closely (impressions, clicks, add-to-carts, and checkout starts) to separate ad problems from site and product problems.
Three-test-round timeline with each round annotated for hypothesis, result, learning.
The before-to-after, silent-legible: where it started, what the data showed, the decision that turned it, where it landed. Past performance guarantees nothing. The diagnosis method is the product.
Stack
- Facebook Ads
- Basic ecommerce site
- Etsy as existing but limited sales channel
Result
Despite multiple high-engagement campaigns and heavy site traffic, we never unlocked a profitable, repeatable path. After three rounds, we called it: the combination of product, price, audience, and site simply was not adding up for cold-traffic scale.
"We called it" outcome card. The experiment closed with the loss named.
Impact
The founder got a clear answer instead of a slow, vague bleed. I walked away with a deeper respect for how hard it is to manufacture demand where real purchase intent is not already flowing.
Lessons
Even when a niche has worked for you before, not every adjacent product is scalable. High CTR is not product-market fit. Without enough budget, a forgiving site, and clear buyer motivation, paid traffic is just an expensive mirror.
Why this matters to you
For founders hoping a single $2K to $5K test will prove their entire model, or for anyone who wants a clear picture of what disciplined testing can and cannot do.
Sounds like you? Get a quote