Fashion & Apparel · High-SKU · Tight Margin

Some of your
best-performing channels
are probably losing you money.

Every platform says it's your top channel. Platform ROAS says everything is working. But when you account for margin, returns, and fulfilment, the picture looks very different. This playbook shows you how to find out.

−54%CAC, Girav, Germany
−40%Ad spend to achieve it
+4%New customers, same period
Based on real results from Girav × Kickbite
Part One — The Real Problem

Every platform says it drove the sale.
They can't all be right.

Your Meta dashboard shows a winning campaign. Google shows the same sale. Pinterest claims it too. Every platform is telling the truth, about its own numbers. None of them are telling the truth about your business.

This is not a tracking problem. It is a structural conflict of interest. Every platform is incentivised to report as many conversions as possible. The result is that the same sale gets counted three times, your reported ROAS looks strong, and your P&L looks flat.

"A campaign can generate a 10× ROAS and still create zero incremental revenue. The question is never what the platform reports, it's what the business actually kept."

Without independent attribution
Meta
I drove the sale. ROAS: 4.2×
Google
I drove the sale. ROAS: 3.8×
Pinterest
I influenced the sale. ROAS: 2.0×
Reality
One customer. One purchase. Three platforms taking credit.
Founder
Why is profit flat if everything is performing?
With Kickbite profit attribution
One conversion
One attribution model
One profit number
POAS: 1.4× after margin & returns
One budget decision
Based on what the business actually kept
Outcome
−54% CAC. +4% new customers. −40% spend.

The return tax your
ROAS is hiding from you

Marketing is the biggest block in your P&L. But most brands don't include returns in that block. The moment you do, your true acquisition cost changes dramatically.

In fashion and apparel, return rates of 20–40% are standard. Every campaign you run on a returnable product is being evaluated on revenue that will partially disappear.

What your dashboard shows
4.0×
Reported ROAS
CAC: €50
Return rate: not included
→
What your business actually has
2.6×
Return-adjusted ROAS
Return-adjusted CAC: €78
35% return rate applied
ScenarioReported ROASReturn rateActual POASReality
PMax scaling on sale itemsDiscounted stock · 15% margin4.1×12%0.6×Scaled, losing money per order
Meta prospecting, outerwear32% margin · 35% return rate2.8×35%0.9×Strong ROAS, negative profit
Pinterest driving accessories78% margin · 3% return rate2.1×3%1.4×Closest to real profitability
Brand search cannibalisationWould have converted organically6.0×—−∞Looked great, pure cost

See it in your account

What does your return-adjusted POAS actually look like?

Kickbite shows you margin and profit at ad level across Meta, Google, and Pinterest, refreshed daily.

Book a demo
Part Two — Proof

How Girav cut CAC by 54%
without spending more

Girav is a D2C men's clothing brand running 850 SKUs and over 2 million product variations. They were scaling confidently, until they looked at where conversions were actually coming from.

Case Study — Girav × Kickbite
Growing profitable without spending more
−54%
Customer Acquisition Cost
−40%
Ad Spend
+4%
New Customers

The problem: After iOS14, Girav's Google Analytics was severely under-reporting upper-funnel channels, particularly social paid. Meta's Ad Manager was over-attributing its own performance. In Q4 2022, CAC in Germany rose 143% while new customer volume missed target.

What they found: After implementing Kickbite, the team recovered 37% of iOS data loss and discovered that direct traffic was overcounted by 47%, while social paid conversions were undercounted by 2,356%. The entire channel mix assumption was wrong.

What they did: Social channels were repositioned as primary retargeting layers. Low-CAC, high new-customer campaigns were scaled. Budget moved toward what was actually working.

ROAS almost doubled. CAC dropped below target. Girav generated more revenue while spending less, not by finding new budget, but by reallocating existing budget based on what the data actually showed.

Part Three — The Framework

What is POAS, and why does
it fix budget allocation?

Profit on Ad Spend measures what the business keeps from every euro of ad spend, after margin, returns, and fulfilment. It is the only metric that connects marketing decisions directly to P&L outcomes.

The CFO Test

If your CFO asked: "Where should we put the next €100,000?" could your team answer using profit data rather than platform reports?

If the honest answer is no, your budget allocation is based on incomplete information. POAS gives every person in the business the same answer to that question.

The Formula
POAS = (Revenue × Margin%) − Returns − Fulfilment ÷ Ad Spend
Calculated at ad level by Kickbite, refreshed daily, using your actual SKU-level margin data synced from your product feed.

How the signal flows

Kickbite tracks profit at ad level across all three channels from a single, independent source of truth, independent of what any platform reports about itself.

Input
SKU-level margins
Feed synced from your catalogue. Every product has a true margin attached.
Layer 2
Return adjustment
Historical return rates per category reduce phantom revenue before attribution.
Layer 3
Kickbite attribution
Cross-channel profit signal at ad level, independent of platform self-reporting.
Output
POAS per ad
One profit number per ad, per channel, per day.
Decision
Scale / Hold / Cut
Rules-based budget allocation. Defensible to finance.

The budget allocation table

ChannelPlatform ROASKickbite POASBudget decision
Meta, Prospecting4.2×1.1×Hold, margin destroyed by returns
Google, PMax3.8×0.9×Cut, brand cannibalisation suspected
Pinterest, Accessories2.0×1.7×Scale, highest margin, lowest returns

The channel with the lowest platform ROAS is the one that deserves the most budget. That is the insight ROAS alone will never give you.

Part Three, continued

The decision framework:
Scale, Hold, Cut

Every ad, every day, falls into one of three zones. Calibrate to your margin structure in the first 30 days with Kickbite.

Scale
POAS > 1.5×

Generating more gross profit than it costs. Increase budget or expand reach.

  • Increase daily budget 20–30%
  • Expand audience or lookalike radius
  • Duplicate to new placements
  • Raise Google asset group bids
  • Brief similar creative variants
Hold
0.8 – 1.5×

Marginal. Maintain spend but don't increase. Investigate before any changes.

  • Freeze budget for 7 days
  • Audit audience overlap
  • Check SKU mix, is margin dragging?
  • Test creative refresh before scaling
  • Flag for weekly review
Cut
POAS < 0.8×

Destroying profit. Pause immediately unless a strategic reason is documented.

  • Pause ad or campaign
  • Reallocate budget to Scale zone
  • Document reason if keeping live
  • Review SKU promotion strategy
  • Escalate if systemic across channel
Part Three, continued

Channel-specific
operating rules

Each channel has a distinct role in your profit architecture. The POAS threshold is the same, the operating logic is not.

G
Google, PMax & Shopping
Conversion capture

Optimisation rules

  • Use Kickbite POAS to set effective tROAS targets.
  • Segment asset groups by margin tier. Never let PMax blend them.
  • Exclude brand terms from PMax. Measure brand search separately.
  • If POAS < 0.8 on a product group, suppress those SKUs from the feed first.

Common failure modes

  • PMax scaling on sale items because ROAS looks strong, actual margin is 15%.
  • Budget on brand cannibalisation, high reported ROAS, zero incremental value.
  • Single asset group blending accessories (78% margin) with outerwear (32%).
  • Modelled conversions inflating ROAS during iOS tracking gaps.
f
Meta, Prospecting & Retargeting
Demand generation

Optimisation rules

  • Separate prospecting and retargeting. POAS diverges significantly between the two.
  • Judge prospecting on a 14-day POAS window, not 7-day.
  • Use creative-level POAS to kill underperformers before Meta phases them out.
  • Watch for audience overlap with Google, double-counting is highest here.

Common failure modes

  • Scaling Advantage+ Shopping while PMax runs simultaneously, both claim the same conversion.
  • Optimising on CTR instead of POAS, high-click creatives driving low-margin purchases.
  • 7-day attribution killing upper-funnel prospecting prematurely.
  • Retargeting pool inflated by Google-driven visitors.
P
Pinterest, Discovery & Upper Funnel
Margin-efficient discovery

Optimisation rules

  • Pinterest ROAS will always look weak. Kickbite captures the assisted conversions platforms miss.
  • Prioritise high-margin categories. Intent-to-browse match is strongest here.
  • Use a 30-day POAS window. Consideration cycles are longer than Meta or Google.
  • First to cut in a squeeze if in Hold zone, last to cut if in Scale zone.

Common failure modes

  • Killing Pinterest because last-click ROAS looks low, ignoring its discovery role.
  • Running Meta creative on Pinterest, requires editorial, lifestyle-led visuals.
  • Not suppressing sale SKUs from Pinterest Shopping.
  • Evaluating on a 7-day window, guaranteed to look like a failing channel.
Part Three, continued

The operating cadence:
daily, weekly, monthly

FrequencyWhoWhat to reviewDecisions made
DailyPerformance Manager
  • 15 min max
  • Kickbite dashboard
  • Ads crossing into Cut zone
  • Ads moving into Scale zone
  • Spend pacing vs budget
  • Pause Cut-zone ads
  • Increase Scale-zone budgets
  • Flag anomalies for weekly
WeeklyHead of Performance
  • 45–60 min
  • Cross-channel view
  • 7-day POAS by channel
  • Budget allocation review
  • Creative ranked by POAS
  • Reallocate budget
  • Kill underperforming creatives
  • Adjust tROAS targets
MonthlyHead of Performance + Founder / CFO
  • 60–90 min
  • P&L reconciliation
  • 30-day POAS vs P&L
  • Channel mix efficiency
  • Threshold calibration
  • Adjust thresholds
  • Strategic channel decisions
  • Set next month targets
Part Four — Building Alignment

The questions every leadership
team asks — answered

Moving to profit-based attribution is a strategic decision. Whether you're the founder setting direction or the performance lead implementing it, these are the questions that need clear answers before the business commits.

Our ROAS looks strong. What's the actual problem?

ROAS measures revenue attributed to ads. It doesn't account for margin per SKU, return rates, or fulfilment costs. A ROAS of 4× on discounted stock with 15% margin is not a win, it's a loss that looks like a win. POAS closes that gap by measuring what the business actually keeps.

How do we know the attribution data is reliable?

No attribution model is perfect. The question is whether it's more reliable than platform self-reporting, where every channel takes credit for the same sale. In the first 30 days, Kickbite runs alongside existing reporting. Every month, POAS decisions are reconciled against actual P&L.

Will this change how the marketing team operates?

Yes, it will make decisions faster and more defensible. With POAS thresholds, daily decisions are rules-based and take 15 minutes. Weekly reviews have a clear agenda. Monthly reviews connect directly to P&L. The framework reduces decision friction across the entire team.

We've invested in tools before. What makes this different?

Most tools add a reporting layer. This changes the decision rule. Every budget call is made against one consistent profit metric rather than each platform's self-reported ROAS. Within 90 days, the business has a measurable before-and-after: what was spent, cut, and what the profit impact was.

"The shift to POAS is not about trusting a new tool. It's about building a shared language between marketing and finance, where every spending decision is connected to a number the whole business can stand behind."

Ready to allocate budget based on profit?

See how Kickbite tracks margin and profit at ad level across Meta, Google, and Pinterest, the same way Girav cut CAC by 54% without spending more.

Book a demo with Kickbite