Operating Intelligence for Retail: Omnichannel Profitability

Retail profitability is an omnichannel problem. Customers browse on mobile, convert in store, return online, and repurchase through whichever channel is most convenient. Each touchpoint carries a different cost structure. Each fulfillment method produces a different margin. And in most retail organizations, the data for all of it lives in separate systems — POS, e-commerce platform, WMS, ERP, marketing stack — that no one has connected into a coherent operating picture.

The result is a common and costly pattern: retailers know their blended gross margin. They do not know whether their digital channel is profitable. They do not know which stores are margin-positive and which are dilutive. They do not know which SKUs are turning at healthy rates and which are quietly consuming working capital and storage cost. They find out when they close the books — or when they need a markdown to clear excess inventory.

Operating intelligence for retail is the discipline of fixing that visibility problem before it compounds into a profitability problem. This article covers the five operating areas where the gap between data availability and decision-making is most costly: omnichannel performance visibility, inventory profitability, channel margin comparison, store-level performance, and supply chain cost intelligence.

Omnichannel Performance Visibility

The defining challenge of omnichannel retail is not connecting channels — most retailers have done that at the transaction level. The challenge is measuring profitability across channels in a way that accounts for how each channel actually operates. A digital order fulfilled from a distribution center has a completely different cost structure than a store pickup order fulfilled from on-hand inventory. Treating them as equivalent in a blended margin calculation produces a number that is accurate in aggregate and misleading in every specific decision it informs.

Retailers offering curbside pickup reported online conversion rates of 3.9% in 2024 — nearly a full percentage point above the 3.1% average for retailers with no omnichannel fulfillment options, according to Digital Commerce 360's analysis of the Top 1,000 retailers. That conversion advantage is real and material. But the full profitability picture requires knowing whether that incremental conversion comes with incremental cost that offsets the revenue gain.

Unified omnichannel intelligence tracks three layers simultaneously:

• Channel attribution: Which channel originated the customer, which channel closed the order, and which channel processed the return. Omnichannel customers shop 70% more frequently and spend roughly 16% more per order than single-channel shoppers — but that value is attributed to a channel mix, not a single source.
• Fulfillment cost allocation: Per-order fulfillment cost by method (DC-fulfilled, store-fulfilled, BOPIS, curbside). Retailers who have achieved high maturity in unified commerce report 27% lower fulfillment costs — but that improvement requires knowing your current cost per order type first.
• Return rate by channel: Online return rates run 20–30% in most apparel categories versus 8–10% in-store. A channel showing strong gross sales with a high return rate has a materially different net revenue profile than the top-line suggests.

BOPIS specifically has shifted from a convenience feature to a strategic margin lever. BOPIS sales hit $154.3 billion in 2025, up 16.2% year over year. When fulfilled correctly — using on-hand store inventory that would otherwise require markdown — BOPIS can improve both inventory turnover and digital margin simultaneously. When fulfilled poorly — requiring split shipments or distribution center involvement — it erodes the cost advantage that justified the model.

Inventory Profitability

Inventory is where retail profitability is most frequently destroyed — not through a single bad decision, but through the accumulation of small decisions made without real-time visibility into turn rates, carrying cost, and markdown risk by SKU.

Industry benchmarks for inventory turnover vary substantially by category:

• Grocery: 15–20 turns per year. The high turn rate is a structural requirement given perishability, but it also means margin is razor-thin — grocery gross margins run 20–25%, with net margins of 1–3%. Any deterioration in turn rate has an outsized impact on profitability.
• Fashion and apparel: 6–12 turns per year. Fast-fashion operators target the high end of that range; specialty retailers run closer to 6–8. Apparel gross margins of 45–55% provide more buffer — but slow-moving fashion inventory carries accelerating markdown risk as seasonality creates a hard clearance deadline.
• Electronics: 4–6 turns per year. Gross margins of 15–25% leave little room for carrying cost. Technology obsolescence adds a dimension beyond simple slow-turn risk — a product that turns 4 times per year may be worth significantly less at turn 4 than at turn 1.
• Furniture and home goods: 3–5 turns per year. Lower turn rates are expected given price point and purchase cycle, but they require precise forecasting to avoid tying up working capital in slow categories.

The decision that operating intelligence enables is not just "what is our current turn rate" but "which specific SKUs are turning below threshold, where are they concentrated, and what is the carrying cost of inaction." A SKU turning at 2x in a category that benchmarks at 8x has a calculable cost — and that calculation needs to be surfaced before the markdown decision becomes forced rather than strategic.

Inventory Metrics That Belong in the Operating View

The inventory metrics that matter for operating decisions are not the same as the inventory metrics that appear in most retail dashboards. Stock availability and weeks of supply are reported widely. The metrics that drive margin decisions — gross margin return on inventory investment (GMROI), sell-through rate by category and location, and SKU-level contribution after carrying cost — are frequently absent from the operating review because they require data from multiple systems to calculate.

GMROI = Gross Margin / Average Inventory Cost. A healthy GMROI varies by category — apparel retailers typically target 3.0 or higher; grocery runs lower given margin compression. Tracking GMROI at a SKU and location level, rather than as a blended average, is what separates inventory optimization from inventory monitoring.

Channel Margin Comparison

Most retailers have a mental model that digital is lower-margin than in-store because of fulfillment cost — and they are right directionally. But the actual margin gap varies significantly based on order economics, and understanding that gap at a product and segment level is what determines whether digital growth is accretive or dilutive to overall profitability.

A useful framework for channel margin comparison accounts for four layers of cost:

• Gross margin by channel: Does the same product sell at the same price across channels, or are digital prices lower due to competitive pressure? Are digital promotions eroding margin on units that would have converted in store at full price?
• Fulfillment cost by order type: DC-fulfilled ground shipping typically runs $8–12 per order for mid-size retailers. Store-to-door and expedited options add $4–8. BOPIS, when using on-hand inventory, can reduce fulfillment cost to near zero. The spread between order types is material at volume.
• Customer acquisition cost by channel: Paid digital acquisition — search, social, affiliate — has a direct, measurable cost per order. In-store foot traffic does not have a zero CAC; it is just allocated differently through rent, staff, and marketing. Channel margin comparison requires consistency in how CAC is attributed.
• Return cost by channel: Processing a return costs $5–15 per unit in labor and logistics before accounting for any inventory value impairment. At a 25% return rate in apparel digital, that cost is embedded in every order's margin calculation whether it appears in the channel P&L or not.

Retail operating expense ratios — SG&A as a percentage of revenue — run 20–30% across most retail formats. The channel-level version of that metric is what determines whether adding volume in a given channel is worth pursuing. Retailers using Fairview to connect their POS, e-commerce, and fulfillment cost data into a single channel margin view frequently surface a gap of 8–15 percentage points between their assumed digital margin and the actual contribution after full cost allocation.

Store-Level Performance Intelligence

Same-store sales growth is the headline metric for physical retail performance. It answers the right strategic question — are established locations growing or declining — but it answers it too slowly and too broadly to be actionable at the operating level.

By the time same-store sales trends negative for a location, several more specific things have already gone wrong: traffic may be declining, conversion may be dropping, average transaction value may be compressing, or a specific category may have lost share to a nearby competitor. The intervention required for each root cause is different. Responding to a blended same-store sales number without knowing which driver is responsible produces generic responses to specific problems.

Store-level operating intelligence disaggregates same-store performance into its components:

• Traffic and conversion rate: In-store conversion rates average 25–45% depending on retail format, with luxury formats at the high end and discount formats at the low end. A location tracking below its historical conversion rate — with stable traffic — signals a floor, merchandising, or staff execution problem. Declining traffic with stable conversion signals a location or marketing problem.
• Average transaction value by category: Movement in ATV often reflects assortment shifts, promotional depth, or mix changes before it shows up in same-store sales. A location where ATV is falling in apparel but stable in accessories may be losing a specific brand or size range to a competitor.
• Labor productivity: Revenue per labor hour, by location and day-part, determines whether staffing levels are matched to demand patterns. Overstaffed off-peak periods and understaffed peak periods are both visible in the data — but only if labor cost is connected to transaction data at the hourly level.
• Shrink and loss rate: Shrink typically runs 1.5–2% of revenue for general merchandise retailers and higher for grocery. Location-level variation in shrink rate is a significant operating cost that rarely appears in the weekly operating review.

Moving from Location Reporting to Location Decision-Making

The difference between a store-level report and a store-level decision system is specificity. A report shows that Location 14 had a 4.2% decline in same-store sales this month. A decision system surfaces that Location 14's conversion rate dropped from 32% to 27% in the last three weeks, concentrated in weekend afternoons, coinciding with a reduction in floor staff coverage. The first observation requires investigation. The second is a decision ready to be made.

Supply Chain Cost Intelligence

Supply chain cost is the third major operating lever — after gross margin and labor — in retail profitability. For most retailers, it is also the one with the least visibility. Landed cost per unit, inbound freight variance, and DC throughput cost are calculated periodically, not continuously. Carrier performance data lives in a TMS that does not connect to the P&L view. Vendor compliance issues that create receiving delays and rework costs are tracked in operations and never surface in the financial operating review.

The supply chain metrics that most directly affect profitability are:

• Landed cost per unit vs. planned: Variance between purchase order cost and actual landed cost — including freight, duties, and handling — determines the true gross margin of every unit in inventory. A 3% inbound freight increase on a category with 20% gross margins has a disproportionate margin impact.
• On-time and in-full (OTIF) rate by vendor: Late or incomplete shipments create stockouts, emergency replenishment cost, and downstream markdowns on adjacent inventory that overstocks because of the shortage in a related category. Retailer OTIF penalties — common with major mass-market retailers — are a direct cost of vendor performance failures.
• DC throughput and order cycle time: Labor productivity in fulfillment operations directly affects the cost per unit shipped. Retailers who have reduced unified commerce fulfillment costs by 27% have done so primarily through DC optimization and store-fulfillment routing improvements — not carrier contract renegotiation.
• Returns processing cost: Return logistics cost is frequently underestimated because it is allocated to a central operations budget rather than attributed back to the channel or product category that generated the return. This misallocation masks the true profitability of high-return categories and channels.

Connecting supply chain cost data to the operating P&L — in real time, not through month-end allocation — is what transforms supply chain management from a cost-control function into an active profitability lever. Fairview's approach to this problem is to ingest vendor performance, freight cost, and DC throughput data alongside channel revenue and store-level metrics, so that operators see the full margin picture rather than isolated slices of it.

Building the Operating Intelligence System

The five areas covered in this article — omnichannel visibility, inventory profitability, channel margin, store performance, and supply chain cost — are each individually addressable with category-specific tools. POS analytics platforms handle store performance. Inventory management systems track turn rates. TMS platforms monitor carrier performance. Each does its job in its lane.

The operating intelligence problem in retail is not a shortage of tools. It is the absence of a layer that connects all of them into a single operating view where decisions about margin, inventory, channel mix, and store performance are informed by the same data at the same time. The COO reviewing weekly performance should not be reconciling five dashboards from five systems to get to a coherent picture of where the business is making money, where it is leaking margin, and what to address first.

That connected layer is what operating intelligence is — and in retail, where margin is thin, channel economics are fragmented, and inventory risk compounds quickly, it is what separates operators who stay ahead of problems from those who discover them after the fact.