Case Study · Performance marketing & automation
Caffe Tamati
Meirav Peretz

"Before Knightica we had never spent a shekel on advertising. Our sales doubled in the first month and stayed there. The numbers they show me are the store's real numbers, not what the ad platform claims. And the WhatsApp reminders bring back orders we used to just lose. Sharp, honest, and a real pleasure to work with."

Meirav Peretz · Founder
★★★★★
01 / Impact by the Numbers
2x
Store revenue from the first month of campaigns, verified against the store's own books. Before: zero paid advertising
6x
Blended net return on ad spend, measured as MER, not platform claims
+53%
New customers window on window, on reporting the owner can trust
35%
Of abandoned carts recovered by the WhatsApp flow, median 5 hours from message to payment
02 / The Client

Client snapshot

Caffe Tamati is an Israeli coffee brand selling through its own online store. Loyal returning customers, new ones arriving by word of mouth, and until 2026 not one shekel spent on advertising.

Industry
Coffee & e-commerce
Channel
Online store
Platform
WooCommerce
Paid ads before Knightica
None
Market
Israel
Caffe Tamati coffee
03 / The Problem

A product people love, and no engine behind it

0
shekels ever spent on paid advertising. Every sale came from word of mouth and Instagram.
~14
checkouts a month started and never paid, thousands of shekels a month, with nobody following up.

No paid reach

Growth ran on word of mouth alone. A great product with no way to put it in front of new customers on purpose.

No measurement

No analytics worth the name. Before the first shekel went into ads, there was no way to know what it would bring back.

Carts left behind

Shoppers reached checkout, stalled, and nobody followed up. The recovery emails already installed went unread.

Platform numbers you can't trust

Once ads run, Meta grades its own homework: it later claimed 5x the purchases the books showed. The store needed its own source of truth first.

04 / Process & Build
Discovery & Build
Month 1

Measurement first

Wired GA4, GTM, the WooCommerce API and the Meta Marketing API together, so the store's own books were the source of truth before a shekel went out.

Month 1

First campaigns live

Meta account built from zero: retention, broad, lookalike and traffic. Store sales doubled in the first month.

Month 2

Account tune-up

One clean UTM scheme, monthly reporting on MER, and Meta's inflated claims set next to the books where the owner can compare them.

Rollout
Month 5

WhatsApp cart recovery built

Every 15 minutes the store is checked for checkouts that stalled. The shopper gets a WhatsApp with a one-tap pay link, then one follow-up.

Month 6

Recovery goes live

8 of 8 shoppers who had abandoned a cart were reached on day one.

Month 6

One report, everything in it

Ads, store revenue and recovered carts on one page, verified against the books, so the owner reads numbers that mean what they say.

05 / Selected Workflows
GA4 GTM Meta API
System 01 · Wave 1

Measurement & Ads Engine

  • GA4, GTM, the WooCommerce API and the Meta Marketing API wired into one source of truth
  • Full Meta account management: retention, broad, lookalike and traffic, built from zero
  • Monthly reporting on MER, store revenue over ad spend, not platform ROAS
  • One clean UTM scheme from day one, so every sale traces back to what drove it
WooCommerce ManyChat WhatsApp
System 02 · Wave 2

WhatsApp Cart Recovery

  • Checks the store every 15 minutes for checkouts that stalled, and waits 10 to 25 minutes so shoppers who pay on their own drop out first
  • Sends a WhatsApp message with a one-tap pay link, no login needed, then a single follow-up
  • 1 in 3 abandoned carts recovered, median 5 hours from message to payment
  • Every send and every recovery logged, and reported monthly next to the ad numbers
Back to home See other work