Overview
A brand with a Shopify store, some of whose deals also close outside the website. The client's name and industry are confidential at its request. Here we measured one thing only: actual store sales against the same period a year earlier. Not platform-reported revenue.
The challenge
A high-ticket product (average order above ₪3,000) and a long buying decision. Many customers consult on WhatsApp or come to a meeting before buying, so the ad platforms miss a large share of sales and don't know whom to learn from.
What we did
- Measured budget growth - we scaled Google and Meta, and checked every step against store sales rather than platform reports.
- Feeding off-site sales back to the algorithm - WhatsApp inquiries and purchases closed outside the website are reported back to Google and Meta as conversions, so campaigns learn who actually buys.
- A new channel - we started testing ads inside ChatGPT, for people who search and compare products in chat.
Results (May-Oct 2026 vs May-Oct 2025, actual sales from Shopify)
- Store sales: ₪1,399,173 vs ₪366,254 - 282% growth
- Orders: 430 vs 81
- Ad spend: ₪311,208 vs ₪66,928
- MER: 4.50 (vs 5.47 a year earlier) - ₪4.5 in actual sales per shekel, even after the budget grew 4.6x
These figures include Shopify store sales only. Sales closed outside the website are not counted here.
Why it's different
Growing a budget 4.6x usually erodes efficiency. Here efficiency dipped slightly while actual sales nearly quadrupled. When you measure against the checkout and not against platform reports, you know exactly how far it pays to scale.
