Garden-art ads drew clicks but failed to establish sales
Three structured campaign rounds exhausted about $10K before we stopped testing.
The ads drew 3% to more than 8% click-through and nobody bought, so I ended the test after three rounds.
Context
Small ecommerce brand turning vintage silverware into handcrafted garden art. Products were visually striking and sold well on Etsy, but even there they didn't scale. 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'd ever seen (3% to 8%+), but purchases didn't follow.
Approach
We ran three rounds of a little over $3,000 each, with a different campaign attempt in each: new account structures, different audiences, fresh creative angles. I watched the numbers closely to separate ad problems from site and product problems.
Stack
- Facebook Ads
- Basic ecommerce site
- Etsy as existing but limited sales channel
Result
Three campaign attempts drew heavy traffic to the site, and nobody bought. After three rounds, we called it.
Impact
I still don't know why nobody bought. The site was weak, which could explain it, and the people who loved the pieces on Facebook may not have been people who pay for garden art. Since then I look first for where people already spend money in a market before trying to start that spending from scratch.
Lessons
I had sold millions of dollars of product in this niche, and this one still didn't sell. The click-through showed people liked the pieces. Only purchases could show they'd pay, and nobody bought. A brand with no sales history should set aside about $10,000 for testing, enough for three to five campaigns.
Who this is for
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 can't do.
Get a quoteThe method behind this result
- Paid AcquisitionPaid acquisition tests need a purchase target, cost ceiling and stopping rule. Business owners can use delivery costs to judge whether another test is affordable.
- The Value Optimization Framework: How Annual Prepayment Funds a GuaranteeA customer who pays for a year of software gives the seller cash before the seller has finished the work. That cash can fund the company. The customer still needs the installation, the support and the result they bought, and a refund promise creates another obligation the seller has to fund. Those obligations determine whether the offer works.