MarlaJewelry×VisionAgency
Buying jewellery takes time. We rebuilt Marla Jewelry's Meta account around a long decision cycle and took over creative production: click-through rate rose 140%.
About the brand
Marla Jewelry sells gold and diamond jewellery through its own e-commerce site. Order values are high, so most buyers do not purchase on the first visit; they compare for a few weeks and come back. Meta ads are the brand's main sales channel.
When we took over there was a single sales campaign; cold and warm audiences sat in the same ad set. The retargeting window was set to 7 days, while the brand's own order data showed more than half of purchases arriving 14 days after the first visit. That gap broke the whole structure.
Why they chose us
The brand asked for two things: show the ad to the right person at the right moment in a product with a long decision cycle, and build a creative flow that goes beyond the product photo. The goal was more qualified clicks and orders from the same budget, without increasing spend. The ad budget stayed flat for the first three months.
- Build retargeting windows that match the long decision cycle
- Separate cold and warm audiences into measurable campaigns
- Raise click-through rate and order count on the same budget
The challenges we faced
In jewellery the problem is not finding traffic, it is keeping an interested user in view until they decide. Four issues had to be solved in the account:
Decision Cycle Exceeded 14 Days
Order data showed more than half of buyers purchasing two weeks after their first visit. A 7-day retargeting window lost these users, so the audience closest to purchase never saw an ad.
Audiences Mixed in One Campaign
Cold traffic, site visitors and previous customers all ran in the same campaign. It was invisible which audience produced the revenue, budget drifted to whatever gave the cheapest click, and the warm audience got too few impressions.
Creatives Gave No Reason to Click
Every ad was a product photo on a white background. The user could not see the stone size, the chain length or how the piece sits when worn, so click-through rate stayed below 1%. The ad carried less information than the product page itself.
Repeat Purchase Was Not Measured
There was no separate audience for previous buyers. In jewellery the second purchase arrives around a special occasion; while we were not reaching that audience, the cheapest source of revenue sat unused.
4-phase structure
We rebuilt the account in four stages: first we separated audiences and windows, then we tested creative, we funded the structure that won, and finally we matched spend against the store's real order revenue.
Audience and Window Split
We split the campaign in three: cold traffic, visitors within 30 days, and previous customers. Retargeting windows were built separately at 7, 30 and 60 days, and each window received its own message.
Creative Testing and Production
We produced three new creatives every week: close-up video showing the piece on the wrist and neck, a visual with size comparison, and a static carrying a customer review. We measured the winning format within two weeks and kept it running.
Budget to the Winning Structure
Sixty percent of budget went to cold traffic, the rest to retargeting and the customer audience. We tracked cost per order daily and switched off ad sets above the threshold; spend concentrated on the best converting product groups.
Measuring Real Revenue Across the Long Window
Because the decision cycle is long, the ad panel's day-by-day report was misleading. We matched spend against the site's order revenue each month and read every order back against its first-visit date, which stopped us from switching off the campaign that brought in the late-returning buyer. Budget decisions followed the money that reached the till.
What we achieved together
15 million impressions turned into 55,200 clicks at a cost per click of €0.032. The period closed with 178 orders. Click-through rate rose 140% against the previous period.
CTR Increase
+140%
vs previous period
Impressions
15M
Cost per Click
€0.032
What Ran
- Cold traffic took sixty percent of budget at €0.032 per click.
- The 60-day retargeting window produced thirty-eight percent of orders.
- Close-up video creatives earned twice the click-through rate of statics.
- The previous-customer campaign delivered 41 of 178 orders at the lowest cost.
Lessons We Took
- The retargeting window must match the product's decision cycle.
- Without splitting cold and warm, no one sees which layer earns.
- A product photo on white gives no reason to click in jewellery.
- Previous buyers are the cheapest revenue source and deserve their own campaign.
In jewellery what wins is not more traffic; it is recognising the buyer's decision cycle and building the ad structure around it. That was the difference at Marla.