Spent
₺42,276
Came back
₺152,372
Counted from the shop's own orders
Return
3.60×
Every lira in, against every lira the order book found
Orders
275
What happened
Computed- The account spent ₺42,276 and the order book found ₺152,372 behind it, a return of 3.60×. Every revenue figure on this sheet is ours; what Meta and Google claim for themselves is larger and lives on the attribution page.
- Brand search took 9 per cent of the spend and produced 27 per cent of the revenue. That is the best line in the account and it is the one worth arguing about, because people typing the shop's name were mostly going to arrive anyway. The honest question is not whether it returns eleven times — it is how much of that would have happened for nothing.
- ₺2,275 of the spend is Brewing guide · traffic, which runs on a traffic objective. It is inside the total above and outside the return, and it is named here rather than quietly left out so this sheet and the desk add up to the same thing.
Where it went, day by day
Ad platformsHow it split, and what came back
Ad platforms · Order storeBy platform
Meta Ads
₺22,15852%
₺39,674 back · 1.96× — Across the judged part. ₺2,275 of it runs on a traffic objective and is outside this rate.
Google Ads
₺20,11748%
₺112,697 back · 5.60×
Marketplace advertising is not in this split and not in any figure on this sheet. Neither marketplace exposes it in the seller API, so the money the shop spends there is spent outside this module.
By campaign
Autumn Blend · launch
₺11,98728%
Meta Ads · ₺27,020 back · 2.25×
Performance Max · catalogue
₺9,86623%
Google Ads · ₺55,374 back · 5.61×
Subscription · always on
₺7,09317%
Meta Ads · ₺10,790 back · 1.52×
Search · non-brand
₺6,30015%
Google Ads · ₺16,659 back · 2.64×
Search · brand
₺3,9519%
Google Ads · ₺40,665 back · 10.29×
Brewing guide · traffic
₺2,2755%
Meta Ads · ₺715 back · not counted — Runs on a traffic objective. What came back is counted; dividing it by what went out would mean nothing.
Back to work · coupon
₺8032%
Meta Ads · ₺1,148 back · 1.43×
What it cost
Computed- Return across the campaigns that are judged on it
- 3.79×
- Spend on a campaign not judged on return
- ₺2,275
- Cost per order
- ₺154
- Cost per click
- ₺7.24
- Share of spend on brand search
- 9.3%
- Share of revenue from brand search
- 26.7%
The headline figure above includes the traffic campaign's spend. This one does not, and the two are not close.
Brewing guide · traffic. It runs on a traffic objective at the Radar's request, and dividing what came back by what went out would produce a number that means nothing.
Advertising cost only. What the coffee cost is in the Sales Cockpit.
People typing the shop's name. Most of them were going to find it anyway.
Which is why the account's return looks the way it does: 9 per cent of the money produced 27 per cent of the revenue.
What was done about it
Entered by handThree changes were approved in this period, putting ₺2,600 a day more behind the account and ₺900 a day taken out of it. Nothing here was sent to Meta or Google until a person approved it in writing, and the sentence they wrote is kept with it.
12 Sep
the owner
BudgetNon-brand search · daily budget
₺700 → ₺900
+₺200 a day
“Two hundred a day to stop going dark at four. Look at it again on Friday and tell me whether the extra two hundred bought anything.”
Since: It still caps, now at about half past five. Cost per order held. Worth another look.
7 Sep
the owner
StartAutumn Blend · launch
₺2,400 a day to 30 Sep
“Start it on the 8th with the two frames that passed. The label is still held, so nothing on the ad says what is in the bag.”
Since: Broad is buying learning rather than orders, as expected. Retargeting is carrying the campaign.
7 Sep
marketing
PauseBack to work · coupon
₺900 a day → Ended
“Off with the coupon, same hour. Do not leave it running over the weekend the way we did in March.”
Since: Stopped cleanly. The March mistake cost about eleven thousand lira, which is the reason this one has a rule behind it now.
What this sheet is counting
Order storeRevenue is counted from the shop's own orders, everywhere in this module, including on the screen you keep open during a campaign. The platform's figure is shown beside it and never instead of it, and the two are never averaged. There is no comparison on this sheet: the account does not go back far enough for a period this long, and an infinite rise is worse than no figure at all.