Your only real ROAS is total revenue over total spend
Meta reports 4.2x. Snapchat reports 3.1x. TikTok says 2.8x. Google says 6x. Add them up and the platforms are claiming more revenue than actually landed in your bank account. The reason is simple: one customer sees your Snap ad on Tuesday, your Instagram Reel on Thursday, searches your brand name on Google on Friday, and buys a 28 KD order. Four platforms, one order, four claims of credit. Seven-day click and one-day view windows make the overlap worse. The fix is not a smarter attribution model. It is arithmetic — total revenue for the week divided by total spend for the week, agency fee and influencer payments included.
Your blended number will always be lower than the platform numbers, and that is correct, not a problem. Most Kuwait stores spending 1,000 to 4,000 KD a month land between 2.5x and 4x blended while their dashboards add up to 6x or 8x. The gap is double counting, not fraud. Then subtract the sales you would have made anyway: repeat buyers, the WhatsApp regulars who order every month, the people who already know you from your Salmiya branch. If you do not separate new customers from returning ones, you will congratulate yourself for paying to acquire people who were coming back regardless.
None of this works unless orders live in one system you own. If a third of your orders come through Instagram DMs, a third through a WhatsApp catalogue and the rest through a delivery app, there is no honest numerator and no honest denominator. Running the store on Shopify with a KNET-enabled checkout gives you one order table, one revenue figure and a report you can pull in a minute every Saturday. Then log each platform's spend in a plain spreadsheet: four rows, one total. That spreadsheet is more honest than any dashboard anyone has shown you.
Ready to start your Shopify store?
Start a free trial and try the platform for yourself.
A reported 4x can still be a losing month
Run the real numbers on a 25 KD average order. Product cost at 45 percent takes 11.250 KD. Delivery costs 1.500 to 2 KD depending on whether you use your own driver or a courier. Packaging is around 0.350. KNET and card fees take roughly 0.250. That leaves about 11.400 KD of contribution before advertising. Now allow for cash on delivery: if 60 percent of orders are COD and 15 percent of those are refused at the door, you are paying the courier both ways on roughly one order in eleven. Returns at 8 percent cost you again. Real contribution lands near 11 KD.
Now bring in the ad cost. At the 4x Meta reports, each 25 KD order looks like it cost 6.250 KD to acquire, leaving about 4.750 KD of profit per order — a healthy store. At a blended 2.4x, the same order actually cost 10.400 KD, leaving 0.600 KD. Sell 400 orders in a month and that is 240 KD of contribution to cover an agency retainer, a warehouse, staff and subscriptions. You are working for the platforms. Nothing changed in your marketing between those two lines. Only the arithmetic changed, and only one version of it is true.
The fix is usually margin, not media. Push prepaid KNET over cash on delivery — a 1 KD discount for paying online is cheaper than a failed delivery, and it typically moves 15 to 25 percent of COD orders across. Raise the average order with a two-item bundle or a free-delivery threshold set about 40 percent above your current average. Cut returns with honest sizing photos and a WhatsApp confirmation before dispatch. Every dinar you add to contribution per order is worth more than a point of reported ROAS, and unlike the reported number, that dinar is real.
Track one number every Saturday morning
Every Saturday morning, open one sheet and fill three cells. Last seven days of revenue. That revenue multiplied by your contribution margin percentage, which you calculate once and revisit each quarter. Total ad spend across Meta, Snapchat, TikTok, Google, influencer payments and your agency fee. Subtract the third number from the second. What is left is what marketing actually earned you last week, in dinars, and it is the only number that belongs in the meeting. If it is positive and growing, spend more. If it has been negative three weeks running, cut spend before you touch the creative.
Track one supporting number beside it: the share of orders from first-time customers. Healthy Kuwait stores sit between 45 and 65 percent new. Below 40 percent you are paying to reach people who would have bought anyway, and your blended figure is flattering you. Above 75 percent you are not retaining anyone, and acquisition cost will eat you within two quarters. Repeat rate matters more than any ad metric, because the second order carries no acquisition cost at all. That is why the honest way to raise blended return is usually email, WhatsApp and a reason to come back.
Compare against the same period last year, never against last month. Kuwait's calendar swings hard: Ramadan and the week before Eid can double order volume, Hala February lifts anything gift-shaped, and July and August empty out as families travel. A store that judges its return in late July will cut budget exactly when costs are seasonally high and revenue is seasonally low. Give any change 60 to 90 days and at least 200 to 300 orders before you rule on it. The first four weeks of a new campaign are the platform learning, and you are paying the tuition.