Channel Profitability Analysis: Your Biggest Channel Is Rarely Your Best Investment
Every board deck tells the same story: the biggest channel gets the biggest investment. Revenue by channel, sorted descending, with the implied instruction to double down at the top. The logic is intuitive. It is also wrong, and a channel profitability analysis shows why.
Revenue measures what comes in. It does not measure what it costs to earn. A channel generating $3M with a 22% all-in trade rate and a $120K deduction tail keeps 74 cents of every gross dollar. A channel generating $1.8M at 9% trade with faster cash cycles keeps 91. The first channel produces more absolute dollars and looks better in every quarterly review. The second returns more per dollar put into it, and the per-dollar number, not the absolute one, is what should steer the next dollar of investment.
The reason most brands have never seen this is that the data to calculate it sits in four different systems (the ERP, the trade management tool, the deduction ledger, and the retailer portals) and nobody has ever joined them at the channel level.
The cost stack nobody builds
Trade spend is the second-largest line on the CPG P&L after cost of goods. It runs 15-30% of gross sales for brands in conventional grocery, and 73% of all CPG marketing spend flows through retailers in the form of trade promotion, shopper marketing, and retail media. But those percentages are not evenly distributed.
National retail accounts carry layered cost structures: contracted promotional allowances, scan-based payments, volume rebates, slotting fees, and compliance penalty programs. Walmart's OTIF program charges 3% of COGS per non-compliant shipment against a 98% threshold. Retailer deductions, valid and otherwise, run 5-15% of gross sales. The cash conversion cycle stretches 90-120 days from production to deposit.
Distribution channels carry their own trade costs, typically 6-12%, but the structure is simpler: fewer penalty layers, fewer promotional requirements, and cash cycles that run 30-45 days shorter than retail. The per-dollar trade cost of earning revenue through distribution is structurally lower than through retail. That alone does not make distribution the more profitable channel: distributors buy at lower wholesale prices, so COGS claims a larger share of every distribution dollar before contribution is counted. Cheaper to serve and less profitable per dollar can both be true of the same channel, which is why the ranking has to be computed from the full cost stack, not assumed from the trade rate.
Nobody sees this in a standard report because no standard report stacks trade, deductions, compliance, and float by channel. Brands track trade spend as a lump budget. They track deductions as an AR problem. They track compliance as a supply-chain problem. The channel-level P&L that would combine all three does not exist.
Why this misdirects your capital
When the only metric available is revenue by channel, every allocation decision flows from it. Where to place new SKUs. Where to increase trade investment. Where to approve slotting commitments. Which accounts to prioritize for fill rates and on-time delivery.
The problem is that revenue rank and return rank are different lists, and the gap does not run in one predictable direction. A smaller account can out-return a bigger one. A low-trade distribution channel can out-return retail on cost-to-serve, or retail can out-return distribution when its higher price realization more than offsets that cost. Which way it breaks is what the analysis has to measure rather than assume. If you are allocating capital by revenue, you are sending more money to the channel that costs more to serve, and underinvesting in the one that produces better returns.
This compounds. The retail account gets the new SKU placements, the incremental trade dollars, the broker attention. The distribution partner gets what is left. Over time, the capital structure drifts further from the return structure, and the gap between where the money goes and where the money works widens. Channel allocation is one of the Ten Decisions where the return on getting it right is measurable and direct.
The cash yield per invoiced dollar varies by channel for the same reason. A channel with heavy deductions and slow payment terms converts a smaller percentage of invoiced revenue to cash than one with clean remittances and shorter cycles. If your margin calculations use gross revenue as the denominator, they overstate the economics of your highest-cost channels and understate the economics of your simplest ones.
What the analysis looks like
The channel profitability analysis is straightforward in concept and tedious in execution, which is why most brands have never done it.
Pull the full cost stack per channel. Not just the trade spend commitment, that is the visible number. The analysis needs promotional scan allowances, slotting and shelving costs, compliance chargebacks, deductions by type, and the timing cost of working capital float. Each comes from a different system.
Normalize to capital deployed. Revenue cannot compare channels. What matters is contribution per dollar deployed, with deployed capital including inventory, trade investment, and operational cost allocated to that channel. A channel producing $500K in contribution on $2M deployed is outperforming one producing $700K on $4M, even though the second number is bigger.
Separate structural trade from operational waste. Structural trade (contracted rates, scan allowances, promotional commitments) is the cost of shelf access. You negotiated it. Operational waste (chargebacks driven by data-quality defects, deductions that went unchallenged past the dispute window, duplicate billings nobody caught) is the cost of bad process. The first is a business decision. The second is recoverable, what the deduction recovery process targets.
The output is a capital allocation map: contribution margin per dollar deployed, by channel, with the cost stack broken into structural and preventable layers. It tells you which channels earn their investment, which channels cost more than they return, and where reallocation would improve net contribution without changing a single sales number.
See it worked through
I built a synthetic $25M specialty food brand (Cinderhaven Provisions, 50 SKUs across ten sell-through channels) to model this analysis end to end. Real cost structures, real trade rates, no client data exposed. The finding: retail channels returned about $54,000 more in contribution per $1M of revenue routed than distribution, because retail's higher price realization more than offset its heavier trade and compliance load. The brand's highest-revenue account had one of its lowest contribution margins per dollar in.