Chapter 01 of 16

← All work

E-commerce · Home furniture · Egypt

From day zero to 14x sales on ad spend

Homzo is an Egyptian home-furniture brand with an online store and a showroom in Zahraa El Maadi, selling with installment offers through valU. Before the first campaign the showroom was still being fitted out. I was involved from that stage: the presence people would see, the store and catalog the ads would sell from, and the data that shows what sells. After launch, each stage opened bigger work, up to a control room that checks spend against the store’s invoices.

Client
Homzo, home furniture
Market
Cairo, Egypt: showroom in Zahraa El Maadi + online store
Period
October 2025 to November 2026
Channels
Meta, Google, WhatsApp, the store’s sales system
My role
From the first meeting: store and system set-up, tracking, media buying

Showroom, Zahraa El Maadi

10 miles12 showroom invoices

All of Cairo and Giza4 showroom invoices

Wider reached more people but sold less, so the 10-mile radius came back.

The Homzo story, narrated · captions on screen, sound when you want it
Read the film

This is Homzo. A furniture showroom in Cairo, and an online store. In October 2025, I sat down with four owners in Cairo. Homzo was still an idea. Before the doors opened, I helped build the store, the sales system and the launch campaign, online and on the streets of Maadi. We spent about six months developing the store. Then we followed what happened after the click, from the product page to the sales desk. Shoppers reached a product and stopped. In a separate offer review, we found two prices for one bed, and flagged it for the developer. Then we widened the showroom campaign. More people messaged us. But Friday invoices fell from twelve to four. We took the campaign back to Maadi. The first six million pounds in sales took six months. The next six million took three. By then, Homzo had sold twelve million. Daily budgets grew from five hundred pounds to twenty to twenty-five thousand, across Meta and Google. Across Homzo’s lifetime, sales reached about twenty-eight million pounds, on about two million in ad spend. Fourteen times spend. Homzo. Built from day zero.

14xsales to ad spend, whole engagement
28M EGPsales across Meta, WhatsApp campaigns and Google, whole engagement
11.29M EGPshowroom invoice value in the store’s own system: 420 invoices, 3 Jan to 23 Oct 2026
All numbers
8.56M EGPof that invoice value already closed (346 invoices)
~2M EGPtotal ad spend behind the 28M
4.28xMeta account only, 1 Jan to 13 Nov 2026: 232,248 EGP to 993,694 EGP of tracked purchases

Each number has its own scope. The 14x, 28M EGP and ~2M EGP cover the whole engagement across Meta, WhatsApp campaigns, Google and earlier ad accounts: sales against ad spend, not sales a platform credited to one ad. The 11.29M EGP is invoice value in the store’s sales system, of which 8.56M is closed. The 4.28x and 93 purchases are the current Meta account alone (1 Jan to 13 Nov 2026), which does not count showroom sales closed after WhatsApp. The 17.2x and 22.3x are single ads, not the account average.

The early store: categories, offers and the product gridOffer first: 25% off
The early store: categories, offers and the product grid
Day zero: the Homzo sign over an empty space, before the showroom openedDay zero: the sign goes up
Day zero: the Homzo sign over an empty space, before the showroom opened
The early Instagram: the opening announcement and countdownCountdown to opening
The early Instagram: the opening announcement and countdown
Meta Ads Reporting: campaigns by spend, purchases and conversion value56,660 → 584,618 EGPTotal: 232,248 → 993,694 EGP
Meta Ads Reporting: campaigns by spend, purchases and conversion value
Google Ads: campaigns, budgets, status and bid strategyPMax Store: 3,000 EGP a day
Google Ads: campaigns, budgets, status and bid strategy
@homzo.egypt profile and highlights155 posts · 1,232 followers
@homzo.egypt profile and highlights
The grid, 13 November 2026Black Friday, in brand colors
The grid, 13 November 2026
Campaigns ranked by ROAS (Ads Manager)ROAS, best first
Campaigns ranked by ROAS (Ads Manager)
Campaigns ranked by spend (Ads Manager)55,410 EGP · 47 buys · 10.55x
Campaigns ranked by spend (Ads Manager)
Top campaigns by ROAS10.48x on 55,810 EGP spend
Top campaigns by ROAS

Sales on five Fridays in October

77K2 Oct
193K9 Oct
118K16 Oct
212K23 Oct
345K30 Oct

EGP, showroom plus online invoices, from the store’s own system.

From product view to purchase

Product views30K
Add to cart2,830
Checkout started1,817
Payment info780
Purchases93

Meta account events, whole period.

The story in 5 steps

  1. Where it started

    Before the first ad, a brand was being built.

    More

    The Homzo showroom in Zahraa El Maadi was still being fitted out and people had not met the brand yet. I was involved from that stage, setting up the sales and marketing system around it.

  2. What was in the way

    A shop and a stock of furniture do not make, on their own, a brand people recognize or a way of buying that can be measured and grown.

    More

    Those had to exist before the spend did.

  3. 3.4Kproduct catalog connected

    The move

    First the presence people would see, the store they would buy from and a 3.4K-product catalog connected to Meta through Pixel and Conversions API.

    More

    Then the offers: the monthly installment as the way in ("368 EGP / month, zero down payment"), separate paths for buying online and for visiting the showroom, and Google beside Meta.

  4. 14 timesthe spend

    What happened

    Across the whole engagement, about 28M EGP of sales on about 2M EGP of ad spend: sales at 14 times the spend, across Meta, WhatsApp campaigns and Google.

    More

    The current Meta account alone, 1 January to 13 November 2026: 232,248 EGP of spend and 993,694 EGP of purchases it tracked (4.28x).

  5. What stays with me

    Wider is not always better. When the Friday audience was widened from Maadi to all of Cairo and Giza, conversations went up but showroom invoices fell from 12 to 4, so the local radius came back.

01 / 10

Step 1 of 10

Day zero

Before the first campaign, the showroom was still being fitted out: the Homzo sign over an empty space. I was involved from this stage. Every number further down this page grew from here.

Illustration: the showroom before opening, one hero piece ready
Illustration: the showroom before opening, one hero piece ready
Before opening
Day zero: the Homzo sign over an empty space, before the showroom openedDay zero: the sign goes up
Look at Day zero: the sign over an empty space, before the showroom opened and before the first campaign.

Before the first campaign there was work to do: the showroom was still being fitted out and people had not met Homzo yet. The work started with the social presence and the opening posts, the store with its categories and offers, and an e-commerce catalog (WooCommerce connected to Meta, 3.4K products) ready for the campaigns.

People had not met Homzo yet. So the brand had to show up first.Next ↓
02 / 10

Step 2 of 10

A brand people could meet

The opening posts and countdown on Instagram, in one orange, and a store with categories, offers and a product grid, ready before traffic was bought.

Instagram
The early Instagram: the opening announcement and countdownCountdown to opening
Look at The early Instagram: the opening announcement and the countdown.
Website
The early store: categories, offers and the product gridOffer first: 25% off
Look at The early store: offer banners first, then categories and products.
Before spend went up, the store had to be able to teach the algorithm.Next ↓
03 / 10

Step 3 of 10

Ready to measure

3.4Kproduct catalog

Pixel and Conversions API connected to a 3.4K-product catalog, browser and server events side by side. Only then did the first campaigns run, so they produced signals the platform could learn from.

Illustration: from the ad signal to the invoice
Illustration: from the ad signal to the invoice
Google Tag Manager
Google Tag Manager: the GA4 and Meta Pixel tags and their triggersEcommerce events to GA4
Look at Tag Manager: GA4 ecommerce events from the store data layer, beside the Meta Pixel.
Meta Events Manager
Meta Events Manager: the dataset, 28 days of events and their match qualityMatch 6.1/10
Look at Events Manager: the store events Meta receives every day, and their match quality.
Google Analytics 4
GA4: active users, key events and purchases14 purchases tracked
Look at GA4: 14 purchases tracked in the last 7 days.
Microsoft Clarity
Microsoft Clarity: sessions, scroll depth, rage and dead clicks12.34% dead clicks: fix list
Look at Clarity: 1,459 sessions in 3 days, and 12.34% dead clicks to fix.

Everything below is the real setup, screenshotted on 13 November 2026. The dataset "Homzo Main" receives Meta Pixel and Conversions API events (158.6K in the last 28 days). Google Tag Manager runs GA4 event tags for contact, lead, search and ecommerce from the data layer, plus the Meta Pixel. GA4 reports the same store, and Microsoft Clarity records sessions, scroll depth and rage and dead clicks.

Then came the offer that changed how people saw the price.Next ↓
04 / 10

Step 4 of 10

The installment as the way in

The ads that did best show a clear product and what it costs a month, not the full price. In Ads Manager, single ads built this way returned 17.2x and 22.3x. Today the same idea runs in reels filmed in the showroom: real people, the payment partner on screen, and one clear offer.

Instagram reel
Pay with valU: a reel filmed on a showroom sofa.
Look at Pay with valU: a reel filmed on a showroom sofa.
Instagram reel
Installments with Souhoola, said by a person, not a banner.
Look at Installments with Souhoola, said by a person, not a banner.
Instagram reel
Black Friday all November, in the brand orange.
Look at Black Friday all November, in the brand orange.
Instagram reel
Come to the showroom: the call to visit, with the card in hand.
Look at Come to the showroom: the call to visit, with the card in hand.
Every step of the buyer’s path is measured, up to the store’s own invoice.Next ↓
05 / 10

Step 5 of 10 · Measurement

I track the customer from the first ad to the cash register

Every step of the buyer’s path is measured by a tool built for that step, and the last one is the store’s own invoice.

  1. Meta + Google Ads

    The click

    Every campaign carries its goal and audience in its name, so every click arrives already labelled.

    43Meta campaigns in 2026
    6,225messaging conversations
    Google Ads: Performance Max and Search, side by side
    Google Ads: Performance Max and Search, side by side
  2. Meta Pixel + Conversions API

    The signal

    The browser and the server both send each event, so ad blockers and lost cookies do not blind the algorithm. The launch campaigns alone tracked 13,127 landing-page views down to 580 add-to-carts and 54 checkouts.

    158.6Kevents in 28 days
    580add-to-carts tracked
    54checkouts started
    Meta Events Manager: the Homzo Main dataset
    Meta Events Manager: the Homzo Main dataset
  3. Google Tag Manager

    The wiring

    One container reads the store’s data layer and fires every tag: contact, lead, search and ecommerce for GA4, plus the Meta Pixel.

    4GA4 events from the data layer
    1container for every tag
    Tag Manager: the tags and their triggers
    Tag Manager: the tags and their triggers
  4. Google Analytics 4

    The path

    Who arrived, from which channel, and who bought, measured against the same week before.

    3.7Kactive users in 7 days
    15Kpage views in 7 days
    GA4: users, key events and purchases
    GA4: users, key events and purchases
  5. Microsoft Clarity

    The behavior

    I watch real sessions: how far people scroll, where they click and where they get stuck. Dead clicks become the developer’s fix list.

    1,459sessions in 3 days
    63.85%average scroll depth
    12.34%dead clicks, the fix list
    Clarity: sessions, scroll depth, rage and dead clicks
    Clarity: sessions, scroll depth, rage and dead clicks
  6. The store’s sales system

    The sale

    The loop closes at the counter. Decisions are made on invoices, not on what a platform claims.

    420showroom invoices
    11.29M EGPshowroom sales, Jan to Oct 2026
That is the measurement. Now open the biggest campaign it feeds.Next ↓
06 / 10

Step 6 of 10

Inside the account

6,753conversations

Once the full tracking, the store and platform integrations and HYPD were in place, the next 11 weeks matched the whole year before them. From 1 September to 16 November 2026 the account ran 45 campaigns, 84 ad sets and 194 ads: 250,610 EGP of spend, 99 website purchases worth 1,105,450 EGP and 6,753 conversations, more than the 93 purchases and 993,694 EGP the earlier reporting showed for January to November. Every campaign is named by goal, audience, offer and date. The biggest one, 3 ad sets and 13 ads, returned 10.62x; the catalog campaign runs 5 ad sets and 35 ads. The second half is a live screen recording: Ads Manager, the Clarity funnel and click heatmap, GA4 traffic by channel and the Google Ads campaigns.

A 40-second cut: the account map, then Ads Manager, Clarity, GA4 and Google Ads, recorded live
That is the account from the inside. Here are the year’s results.Next ↓
07 / 10

Step 7 of 10

The results, dated

6,225messaging conversations

Meta account only, from 1 January to 13 November 2026: 232,248 EGP of spend, 93 purchases worth 993,694 EGP, and 6,225 messaging conversations. The purchase campaign alone returned 584,618 EGP.

Meta Ads Reporting
Meta Ads Reporting: campaigns by spend, purchases and conversion value56,660 → 584,618 EGPTotal: 232,248 → 993,694 EGP
Look at First row: 56,660 EGP of spend, 50 purchases, 584,618 EGP of value. Bottom row: the account total.

Meta Ads Reporting for the Homzo account from 1 January to 13 November 2026: 232,248 EGP of spend, 93 purchases worth 993,694 EGP, and 6,225 messaging conversations across 43 campaigns. The purchase-focused campaign alone returned 584,618 EGP on 56,660 EGP of spend, and the new-chair campaign 118,340 EGP on 11,961 EGP.

Behind that total, these campaigns did the work.Next ↓
08 / 10

Step 8 of 10

What the numbers said

Ranked by ROAS, retargeting and the Cairo–Giza purchase campaign won. Ranked by spend, you can see where the budget actually sat. Both views together decide where the next pound goes.

Meta Ads Manager
Campaigns ranked by ROAS (Ads Manager)ROAS, best first
Look at The top of the list is where the return is.
Meta Ads Manager
Campaigns ranked by spend (Ads Manager)55,410 EGP · 47 buys · 10.55x
Look at The same campaigns ranked by spend.
Meta Ads Manager
Top campaigns by ROAS10.48x on 55,810 EGP spend
Look at The strongest campaigns by ROAS.
Meta Ads Manager
Results summary993,694 EGP in purchases4.37x purchase ROAS
Look at The summary view of the account.
  • 12.17xRetargeting (cart + checkout + viewers)12.17x.
  • 10.62xBiggest campaign, autumn 202610.62x across 3 ad sets and 13 ads, 1 September to 16 November 2026.
  • 10.3xAdvantage+ purchase campaign for Cairo and Giza10.3x on 56,660 EGP, 584,618 EGP in purchase value.
  • 10.10xDiscount-versus-installment campaign10.10x. New-chair sales: 9.89x. Bed sales: 8.57x.
  • 22.3xBest single ads22.3x for a bedroom set from 1,967 EGP/month, and 17.2x for bed HZOR9 from 368 EGP/month on 15,244 EGP spend (each ad’s own result in Ads Manager, not the account average).
  • 368 EGPWhy they wonboth lead with the monthly installment ("368 EGP / month, 24 months with valU, zero down payment") next to a clear product shot, instead of the full price.
  • A separate WhatsApp funnel for showroom traffic around Maadi, with cash-discount and installment offers.
  • Structured testsautomatic vs manual audiences, budget set per campaign vs per ad set, one catalog campaign per category, Black Friday prep.
Meta is only one of the channels. Google runs beside it.Next ↓
09 / 10

Step 9 of 10

Google beside Meta

Performance Max for the showroom and for the product feed, and Search campaigns split by category, brand and showroom.

Google Ads
Google Ads: campaigns, budgets, status and bid strategyPMax Store: 3,000 EGP a day
Look at Campaign type and bid strategy per campaign, with the daily budget.

The campaign list of the Homzo Google Ads account on 13 November 2026: Performance Max for the store and for the product feed, and Search campaigns split by category, brand and showroom.

All of it feeds one brand, and this is how it looks today.Next ↓
10 / 10

Step 10 of 10

The brand today

A grid that mixes reels, installment posts and Black Friday. The last two months are reels shot in the showroom with real people: the app, ordering from home, and the rooms themselves.

Instagram reel
The Homzo app: the catalog on the phone.
Look at The Homzo app: the catalog on the phone.
Instagram reel
Order online from where you are.
Look at Order online from where you are.
Instagram reel
The living room, played for fun.
Look at The living room, played for fun.
Instagram reel
A walk through the showroom.
Look at A walk through the showroom.
Instagram
@homzo.egypt profile and highlights155 posts · 1,232 followers
Look at Highlights for the soft opening, deliveries and the showroom.
Instagram
The grid, 13 November 2026Black Friday, in brand colors
Look at Reels and offers share one visual language.

The brand’s Instagram today: reels, Black Friday and installment posts in one grid.

More from this account 6 sections · 11 to 16

11The sales system proves the showroom sales

The numbers that matter come from the store’s own sales system, not from an ad platform. Its export holds 420 showroom invoices worth 11,288,949 EGP between 3 January and 23 October 2026, plus 78 online orders worth 1,038,141 EGP. Of the showroom total, 8,557,777 EGP across 346 invoices is already closed.

  • Every figure is an invoice total, so it counts what was actually sold at the counter, which Meta and Google cannot see.
  • Customer names, phone numbers and invoice rows are never shown here.

12Creative production

Offers only win if the creative matches them, so I ran creative production alongside the media. That meant 176 campaign images in eight batches. The batches covered Friday cash and installment offers, the showroom, the November campaign, a 12-design Friday pack, 69 cash-offer winners and the Black Friday launch and final sets. On top of that came a Black Friday pack of 22 ad files with copy, and 16 videos. A written analysis maps every post to its offer and to a separate campaign goal: website purchase, showroom WhatsApp or app install.

13A control room built from invoices, not platform numbers

Platform dashboards disagree with each other and with the cash register, so I built an hourly control room for the account. Small Python collectors pull Meta, Google Ads, GA4, Clarity, competitor ads and the offline sales from the store's own sales system. One dashboard then shows the top insights, the proposed decisions, spend pacing every 2 hours, tracking health, daily spend against sales, orders and the visitor funnel.

The Black Friday target was set in the same system: 20M EGP in sales on 1M EGP of spend (5%), 1–30 November.

14The Friday showroom playbook

Showroom WhatsApp campaigns aimed at a 10-mile radius around Maadi, with cash-discount and installment offers, turned Fridays into the biggest sales day of the week.

  • 944,130 EGPFive Fridays in October944,130 EGP across five Fridays, showroom plus online, from the sales system: 2 Oct 76,711 · 9 Oct 192,835 · 16 Oct 118,070 · 23 Oct 212,007 · 30 Oct 344,507 EGP.
  • 1830 Oct18 invoices (12 showroom, 6 online) for 344,507 EGP on 15,427 EGP of spend: about 22x, spend at 4.5% of sales. These are same-day sales against same-day spend, not all from that day’s ads. Sales were up 62% on the previous Friday with spend up 60%.
  • 10The next Friday the audience was widened from 10 miles around Maadi to all of Cairo and Giza. Conversations rose 21%, but real conversations (3+ customer messages) fell about 30% and showroom invoices dropped from 12 to 4. The 10-mile targeting was restored.
  • 166WhatsApp quality checkof 166 conversations at about 14 EGP each, only 19% reached a third message. A new qualifying ad set fixed it.

15Diagnosing a revenue drop

When sales fell, I compared 24–25 Oct against 17–18 Oct using the sales system, not the ad platforms.

  • 6 to 2Online invoices fell from 6 to 2 (−66.7%) and their value from 93,068 to 24,025 EGP (−74.2%).
  • 12,907 vs 12,717 EGPMeta spend was flat (12,907 vs 12,717 EGP) but the share going to website-purchase campaigns fell from 68.8% to 47.7%.
  • 7.42% to 4.93%Product view to checkout start fell from 7.42% to 4.93%.
  • 3,990 EGPAn offer mismatchthe ad said 3,990 EGP while the site showed 3,995 EGP, and the discount code stopped working when installments were chosen. A prioritized fix list went to the store's developer.
  • 4GA4 reported more purchases than the sales system had invoices, because GA4 fires at order creation. Decisions were made on invoices.

16Across channels (GA4 + Google Ads)

On the site, Paid Social is the largest revenue channel in GA4: 507,020 EGP from 36 orders, about 46% of all purchase revenue. Google Ads runs beside it, including a Performance Max store campaign for Maadi, Mokattam and New Cairo and a "Friday Showroom 20% Cash" search campaign. GA4 and Meta count orders differently, so the figures are shown separately.

NextTravelTravitell EgyptThe client kept raising spend from $1K a month to $1K–$1.5K a day, over about 3 years, because the sales kept coming.Continue Travitell Egypt← All work