PKR 42.6M in DTC sales at -45% CAC
Ali Rashid · DTC E-commerce
The challenge
A Shopify DTC operator scaling Meta spend on a half-broken pixel — iOS, ad-blockers and COD orders had stripped 30–40% of conversion signal, so Advantage+ was optimizing blind. Reported ROAS looked healthy but contribution margin kept shrinking, and nobody could tell which SKUs actually paid for themselves once the 18% COD return rate cleared.
The mechanism
Meta CAPI rebuild + POAS bidding + Klaviyo retention. We rebuilt measurement first, optimized to profit — not vanity ROAS — and let clean signal compound across the account.
The result
PKR 42.6M in DTC sales at -45% CAC — measured, attributed and sustained in the Pakistan / GCC market.
Ali Rashid's Shopify DTC brand cleared PKR 42.6M in sales at a 45% lower CAC after a Meta CAPI rebuild, POAS bidding, and a Klaviyo retention layer turned a half-broken pixel into clean profit signal. Event Match Quality jumped from 4.2 to 8.1 inside a week.
The Brutal Problem
Ali had built the brand to PKR 28M annualised on instinct, persistence and a tolerance for chaos that had finally run out. By month 14 of scale, Meta spend was at PKR 3.2M/month and the bank account was contracting, not growing. Reported ROAS sat at 2.4x; Ali's reconciliation against bank deposits and the 18% COD return rate said true contribution margin was negative 8%. He had stopped drawing a salary in March. His older brother had loaned him PKR 1.8M in May to clear supplier dues — a conversation he described to us as "the worst hour of my life." His wife had stopped going to family gatherings because the same uncle kept asking when Ali would "finish playing with the website." His warehouse manager and one of his two CSRs were quietly looking elsewhere. He was running the 2am dashboard refresh ritual — staring at Meta's ROAS column as if it would change. The product was real: 4.8 average rating, 22% organic UGC, two GCC distributors had reached out unsolicited. But the unit economics were lying to him, and every additional rupee of spend was making the lie louder. He was three months from shutting down.
What Made It Worse
Five compounding failures. (1) The browser-only pixel was missing 30-40% of purchases — iOS, ad-blockers and the WhatsApp/COD flow all dark to Meta. (2) Advantage+ was learning against the wrong signal because PMax shared learning pools with retargeting. (3) The 18% COD return rate was never netted out — "ROAS" included gross orders, not delivered. (4) Klaviyo had 31K contacts and one flow. (5) Creative was being rotated quarterly; fatigue scores were red on every top-spending asset.
The Diagnosis
Quantified leaks. EMQ 4.2 (Meta optimizing against ~half-signal). True purchase capture 62% of actual orders. Reported ROAS 2.4x → POAS-adjusted 0.92x → contribution margin -8%. PMax top three SKUs by spend were the bottom three by margin. Two highest-margin SKUs received 6% of spend. Klaviyo revenue share 4.1% (benchmark 25-32%). CAC PKR 1,840; LTV PKR 2,210 (1.2:1 — brands need 3:1 to compound). Creative fatigue: top-3 ads at frequency 4.6, CTR collapsed 41% vs 30-day prior. Diagnosis: the brand was acquiring the wrong customers profitably-on-paper and losing money on every one of them in cash. Not a scaling problem. A signal-and-retention problem.
The Solution Stack
12 weeks, 11 steps:
- Week 1 — Stape CAPI gateway + dedicated sGTM container. Hashed user data, deduplicated events, server-side Purchase / AddToCart / Lead / COD-Confirm firing.
- Week 1 — EMQ verification — climbed 4.2 → 8.1 inside 7 days.
- Week 2 — Margin + return-rate feed piped into Meta and Google via offline conversion uploads.
- Week 3 — POAS bidding activated on top campaigns; revenue-based ROAS deprecated as a KPI internally.
- Week 4 — PMax margin fencing — three tiers, low-POAS SKUs blocked from scaling.
- Week 5 — Creative Velocity — 14 new variants/month, 72-hour kill window.
- Week 6 — Klaviyo lifecycle — welcome (7), AC (3+SMS), post-purchase (14d arc), win-back (45/90/120), VIP.
- Week 7 — WhatsApp COD recovery flow.
- Week 8 — Audience hygiene — separated retargeting from prospecting learning pools.
- Week 10 — GCC market test — replicated stack into UAE and Saudi audiences with Arabic creative.
- Week 12 — Daily profit dashboard reconciled to bank deposits.
The Inflection Point
Week 5. POAS bidding had been live for two weeks. On a Tuesday, the algorithm shifted overnight: three campaigns Ali had considered hero campaigns got starved (they had been driving low-margin SKUs); two campaigns he'd been considering pausing surged (high-margin, lower revenue, higher net). Daily delivered orders climbed from 41 to 78 inside a week, with spend flat. The GCC test in week 10 produced a single AED order from a Dubai buyer — the first international sale. Within two months that segment was 19% of revenue. Ali messaged the team a screenshot of his bank balance going positive for the first time in five months. No comment, just the screenshot.
Final Numbers
| Metric | Before | After | Change |
|---|---|---|---|
| Sales (12 mo) | ~PKR 18M | PKR 42.6M | +137% |
| CAC | PKR 1,840 | PKR 1,012 | -45% |
| EMQ | 4.2 | 8.1 | +93% |
| POAS | 0.92x | 2.7x | +193% |
| Klaviyo revenue share | 4.1% | 32% | +7.8x |
| Contribution margin | -8% | +17% | +25 pts |
| LTV:CAC | 1.2:1 | 3.6:1 | 3x |
What We Learned / Replicable Playbook
Pakistan and GCC DTC brands pay the highest signal-tax in the world: iOS share, ad-blocker share and COD-share all stack against browser pixels. The replicable sequence: (1) server-side CAPI before anything else, (2) verify EMQ above 7, (3) feed margin and COD return-rate into bidding, (4) tier PMax by margin, (5) build the lifecycle layer before scaling further, (6) only then test new markets. Skipping (1) and (2) and jumping to "more creative" stays stuck at break-even. Sequence is the moat. Brands that do it in order go profitable inside 60 days.
Daily-Life Outcome
Ali repaid his brother in month six. He drew his first salary in month five. He hired back the warehouse manager and added a second CSR in month seven. His wife went back to family gatherings — and the uncle stopped asking about the website. He sleeps. He's now running a 2027 plan that includes a GCC distributor agreement and a second product line.
FAQ
How fast does EMQ move after CAPI deployment?
Inside 7-10 days for well-instrumented stores. We verify weekly until it sits at 8+.
Does this work without Shopify Plus?
Yes. Standard Shopify + Stape + sGTM + Klaviyo is the canonical stack.
What about TikTok?
Same stack. TikTok Events API replaces Meta CAPI, signal flows are identical.
Driving services: DTC Ecommerce Growth · Server-Side Tracking · SEM & Paid Advertising. Further reading: Meta CAPI Setup Guide · POAS vs ROAS.
Get a free 48-hour audit → We'll show you the exact conversions Meta is missing on your account — and the CAC drop you'd see if it weren't.
The service behind this result.
This outcome was driven by Meta CAPI rebuild + POAS bidding + Klaviyo retention. Explore the service and the thinking behind it.
Your account has a story like this in it. We'll find it.
A free 48-hour audit shows you exactly where the growth is hiding.