Cajees Lighting & Electrical

CEO Dashboard · Ads, store traffic and sales in one place

3 ad campaigns live

Data through 8 Aug 2026
Currency: ZAR
Meta 1746036432952526 · Google Ads 740-061-9686
GA4 property 505967287

Viewing month

Compare with

Combined performance

Revenue the platforms claim

—

Total ad spend

—

Blended ROAS

—

Conversions claimed

—

Platform split —

Meta Ads

Facebook & Instagram

Spend

—

Revenue

—

ROAS

—

Purchases

—

Clicks

—

CTR

—

CPC

—

Add to cart

—

Checkouts

—

Google Ads

Shopping & Performance Max

Spend

—

Revenue

—

ROAS

—

Conversions

—

Clicks

—

CTR

—

CPC

—

Impressions

—

Cost / conv.

—

Six-month trend

Spend vs revenue by month

Both platforms combined. Selected month highlighted.

ROAS by month, per platform

Return on every R1 spent.

Meta vs Google

Where the money went

Where the revenue came from

Month against month

	Selected	Comparison	Difference

Google Ads campaigns —

Campaign	Type	Spend	Clicks	CTR	CPC	Conv.	Revenue	ROAS

Meta ad scoreboard — July 2026 (last complete month)

Ad-level detail for July, the last complete month. August has only R246 of spend so far, too little to read. All nine ads sit in one campaign, Cajees - Sales - 7 Jan.

Ad	Status	Spend	CTR	CPC	Purchases	Revenue	ROAS
Ad 8	Carrying the account	R1,674	6.50%	R0.78	9	R15,395	9.19×
Ad 9	Converting	R1,451	5.87%	R0.84	2	R2,188	1.51×
Ad 7	No sales	R725	5.66%	R0.82	0	R0	—
Ad 12	No sales	R236	5.80%	R0.78	0	R0	—
Ad 10	No sales	R158	6.08%	R0.71	0	R0	—
Ad 2	No sales	R49	8.68%	R0.86	0	R0	—
Ad 5	No sales	R21	8.36%	R0.74	0	R0	—
Ad 6	No sales	R22	5.11%	R1.55	0	R0	—
Ad 4	No sales	R9	7.69%	R0.86	0	R0	—

CEO action list — ads

1. Google is the bigger engine

Since February Google claims R188,644 on R27,724 spend against Meta's R73,974 on R23,147. Similar budgets, roughly 2.1× the claimed return.

2. Fix the March tracking gap

March recorded 15 Google conversions but only R5 in conversion value. The sales happened; the value did not report. Verify conversion value tracking in Google Ads and Shopify.

3. Feed Ad 8

Across all of July Ad 8 returned 9.19× on R1,674 with 9 of the account's 11 purchases. Ad 9 spent R1,451 for 1.51×. Move budget across and build the next creatives off Ad 8's angle.

4. Seven Meta ads make no sales

Seven of nine live ads spent R1,220 across July with zero tracked purchases. Pause them and concentrate budget on the two that convert.

5. The gap is checkout, not clicks

Meta CTR climbed from 2.55% in February to 6.10% in July and CPC fell from R1.56 to R0.80. The ads are working. The leak is at checkout. See the Site tab.

6. Performance Max does the selling

Performance Max produced the bulk of Google's July revenue while Shopping delivered far more clicks for a fraction of it. Review the Shopping feed and bidding before adding budget there.

Store performance

Sessions

—

Visitors

—

Store revenue (actual)

—

Conversion rate

—

The buying funnel —

Counted in people, not events. Each bar shows how many of the previous step carried on.

Traffic and conversion over six months

Sessions vs conversion rate

Bars are visits. The line is the share of visits that bought.

Store revenue by month

What the store actually recorded, all channels included.

Who is arriving —

Traffic by channel

Phone, computer or tablet

Channel scoreboard —

Channel	Type	Sessions	Share	Purchases	Revenue	Conv. rate

What the ad platforms claim vs what the store recorded

	Meta & Google claim	Store actually recorded	Gap

Why the numbers differ. Meta and Google each count a sale they believe they influenced, so one order can be claimed twice. Google Analytics counts the order once, on the store. Treat the store figure as the truth for revenue, and the platform figures as a guide to which ads are working relative to each other.

CEO action list — store

1. Conversion tracking cleaned up — done 9 Aug

A Google Analytics rule was inventing a purchase on every checkout page view, inflating Google Ads purchase reporting about 13×. The rule is deleted. Separately, Smart Bidding was optimising for a blend of sales, phone calls, forms and newsletter signups — only 36% of the signal was actual sales. Purchases are now the sole bidding goal. Expect reported conversions to drop sharply and correctly from here.

2. The leak is the delivery step

In July 146 people started checkout, only 46 reached the payment screen, and 16 bought. 68% are lost before payment is even shown. That gap is email, delivery address and shipping cost. Payment itself converts at a normal rate, and a live test on a real phone completed without a problem.

3. Test a difficult delivery address

Checkout works fine to a served address. The question is what a customer far from Trichardt sees: a large delivery quote, or no shipping option at all. Put a rural postal code through checkout and watch what comes back. That is the cheapest way to confirm the 68%.

4. Half your visitors are inside Facebook's app

51% of sessions run in the Facebook in-app browser, where 23.5% throw a JavaScript error against 0–5% in normal browsers. A live checkout test in that browser still completed, so this is not blocking sales, but the product-page scripts are worth cleaning up.

5. Paid Shopping brings crowds, not buyers

Paid Shopping delivered 4,190 July sessions and 1 order. Cross-network delivered 1,240 sessions and 8 orders worth R42,303. Now that purchases are the only bidding goal, give Google 2–3 weeks to re-learn before judging Shopping again.

6. Turn on abandoned checkout emails

Around 130 people a month reach checkout and leave. Shopify can email them automatically: Settings → Checkout → Abandoned checkouts. Offer free delivery rather than a discount, since delivery is the suspected cause. Even 5% recovery is meaningful against 21 orders a month.