On this page本页目录
How do you identify an unprofitable product?如何识别亏损商品？
Calculate product contribution after sales reversals, sale-time cost of goods, forward and reverse fulfillment, payment and marketplace fees, handling, support, write-down, and evidenced recovery. Show both total contribution and contribution per fulfilled unit across a comparable mature period. Then separate avoidable from committed shared costs, test missing-data and allocation assumptions, quantify inventory and capacity consequences, and review bundle, acquisition, retention, brand, safety, and customer roles. A negative allocated margin is a screening signal—not an automatic discontinue decision.
计算销售冲回、销售时商品成本、正向与逆向履约、支付与平台费、处理、客服、减值及有证据回收后的商品贡献。在可比成熟期间同时展示总贡献与每履约件贡献。再区分可避免与已承诺共享成本，测试缺失数据和分配假设，量化库存与产能后果，并复核捆绑、获客、留存、品牌、安全与客户角色。负分配利润只是筛查信号，不是自动下架决定。
“Unprofitable” depends on scope and time horizon. Gross profit, contribution, cash contribution, fully allocated profit, incremental decision profit, and lifetime/customer economics answer different questions. This guide is management decision support, not audited accounting, tax advice, valuation, or a promise that discontinuing a product will improve company profit.
“亏损”取决于范围与时间跨度。毛利、贡献、现金贡献、完全分配利润、增量决策利润及生命周期/客户经济回答不同问题。本文用于管理决策支持，不是审计会计、税务建议、估值，也不保证下架商品会提高公司利润。
Evaluate seven decision layers评估七个决策层
A product can be negative under one allocation and positive under an avoidable-cost view. It can also support profitable baskets or consume scarce capacity. Keep the layers visible and do not collapse them into a hidden score.
一个商品在某种分配下可能为负，但在可避免成本视图下为正；也可能支持盈利购物篮或占用稀缺产能。应让各层可见，不要压缩为隐藏分数。
Net sales, sale-time COGS, direct forward/return costs, fees, write-down, and realized recovery.净销售额、销售时商品成本、直接正向/退货成本、费用、减值与已实现回收。
Fulfilled and retained units, total contribution, seasonality, lifecycle, cohort maturity, and trend.履约与保留件数、总贡献、季节性、生命周期、群组成熟度与趋势。
Direct variable, step-variable, committed, shared, sunk, exit, and transition costs by horizon.按时间跨度划分直接变动、阶梯变动、已承诺、共享、沉没、退出与过渡成本。
Working capital, aging, space, pick/pack touches, support workload, supplier commitments, and alternatives.营运资金、库龄、空间、拣包触点、客服负荷、供应商承诺与替代用途。
Acquisition, repeat behavior, attachment, substitution, retention, accessibility, and customer harm.获客、复购、连带、替代、留存、可访问性与客户伤害。
Fix, reprice, change promotion, bundle, route, source, policy, service level, limit, monitor, or retire.修复、改价、调整促销、捆绑、路由、采购、政策、服务等级、限制、监控或退出。
Coverage, estimates, allocation sensitivity, owner, approval, experiment, rollback, and post-decision review.覆盖、估计、分配敏感性、负责人、批准、实验、回滚与决策后复核。
Discontinuing a product does not automatically eliminate allocated overhead. Estimate which costs and constraints actually change, when they change, and where demand, inventory, customers, and workload go.
下架商品不会自动消除已分配间接费用。应估计哪些成本与约束真正变化、何时变化，以及需求、库存、客户与工作量转移到哪里。
Choose the profitability view that matches the decision选择与决策匹配的盈利视图
Publish multiple bridges rather than one ambiguous “profit.” Every view should name its included revenue, costs, recovery, allocation, time horizon, currency, cohort, and coverage.
发布多个核对桥，而不是一个含糊“利润”。每种视图都应声明包含的收入、成本、回收、分配、时间跨度、币种、群组与覆盖。
| Profitability view盈利视图 | Useful decision适用决策 | Main caution主要注意 |
|---|---|---|
| Gross profit毛利 | Pricing and merchandise baseline定价与商品基线 | Omits fulfillment, returns, support and other drivers省略履约、退货、客服等驱动 |
| After-returns contribution退货后贡献 | Compare operating economics by product比较商品运营经济 | Depends on complete direct cost and recovery coverage依赖完整直接成本与回收覆盖 |
| Incremental / avoidable contribution增量/可避免贡献 | Fix, limit or retire decision修复、限制或退出决策 | Time horizon and step-cost thresholds matter时间跨度与阶梯成本阈值重要 |
| Fully allocated profit完全分配利润 | Accountability and portfolio view责任与商品组合视图 | Negative value may not mean costs disappear on exit负值不表示退出时成本消失 |
| Customer/basket economics客户/购物篮经济 | Traffic, attachment and retention role流量、连带与留存角色 | Association is not incremental causation关联不等于增量因果 |
Build a decision-ready product portfolio table建立可用于决策的商品组合表
- Reconcile product identity核对商品身份
Use stable variant and parent keys, SKU aliases, effective dates, bundle components, substitutions, launch and lifecycle.使用稳定变体与父级键、SKU 别名、生效日期、捆绑组成、替代、上市与生命周期。 - Calculate complete contribution计算完整贡献
Join sale-time revenue/COGS, forward costs, return/refund legs, reverse costs, disposition, recovery, fees, and late adjustments.关联销售时收入/商品成本、正向成本、退货/退款环节、逆向成本、处置、回收、费用与迟到调整。 - Classify cost behavior分类成本行为
Mark direct, allocated, variable, step-variable, committed, sunk, exit and transition costs for each horizon.按时间跨度标记直接、分配、变动、阶梯变动、已承诺、沉没、退出与过渡成本。 - Add inventory and capacity加入库存与产能
Measure on-hand and inbound units, aging, purchase commitments, storage, touches, support load and alternative use.衡量现有与在途库存、库龄、采购承诺、存储、触点、客服负荷与替代用途。 - Measure customer and basket role衡量客户与购物篮角色
Analyze attachment and repeat behavior with matched or experimental evidence where possible; preserve substitution paths.可行时用匹配或实验性证据分析连带与复购，并保留替代路径。 - Run sensitivity and scenarios运行敏感性与情景
Vary missing costs, recovery, allocation drivers, demand transfer, price, rate, supplier, service and exit timing.变化缺失成本、回收、分配驱动、需求转移、价格、退货率、供应商、服务与退出时间。 - Govern a reversible decision治理可逆决策
Record owner, option, expected mechanism, guardrails, test cohort, approvals, rollback, review date and realized outcome.记录负责人、方案、预期机制、护栏、测试群组、批准、回滚、复核日期与已实现结果。
The CSV includes product identity, after-returns contribution, cost behavior, inventory and capacity, customer/basket role, data coverage, scenario assumptions, alternatives, decision owner, test, rollback, and realized outcome.
CSV 包含商品身份、退货后贡献、成本行为、库存与产能、客户/购物篮角色、数据覆盖、情景假设、替代方案、决策负责人、测试、回滚与已实现结果。
Download CSV template下载 CSV 模板 ↓Estimate the incremental effect of each option估计每个方案的增量影响
Show the base case and plausible ranges for volume, price, return rate, recovery, variable and step costs, substitution, inventory liquidation, timing, and customer effects. Do not add fully allocated cost savings unless the cost actually changes under the selected horizon.
展示销量、价格、退货率、回收、变动与阶梯成本、替代、库存清算、时间及客户效应的基准与合理范围。除非成本在所选时间跨度下确实变化，否则不要加入完全分配成本节省。
| Output输出 | Required context所需背景 | What it can support可支持内容 |
|---|---|---|
| After-returns contribution退货后贡献 | Reconciled direct ledger, mature period, coverage已核对直接台账、成熟期间、覆盖 | Screen economic performance筛查经济表现 |
| Avoidable contribution可避免贡献 | Cost behavior, time horizon, thresholds, commitments成本行为、时间跨度、阈值、承诺 | Estimate limit/retire effect估计限制/退出影响 |
| Inventory and transition value库存与过渡价值 | On-hand/inbound, aging, liquidation, cancellation, disposal, timing现有/在途、库龄、清算、取消、处置、时间 | Plan implementation cash and loss规划实施现金与损失 |
| Customer/basket effect客户/购物篮效应 | Attachment, substitution, matched cohorts or experiment, uncertainty连带、替代、匹配群组或实验、不确定性 | Avoid local optimization避免局部优化 |
| Option scenario方案情景 | Price, content, product, supplier, route, service, policy and capacity assumptions价格、内容、商品、供应商、路由、服务、政策与产能假设 | Compare reversible alternatives比较可逆替代 |
Worked example: a negative allocated margin is not the decision示例：负分配利润不等于决策
A synthetic product shows $80,000 after-returns contribution before shared overhead and −$20,000 after $100,000 of allocated warehouse and management cost. Only $15,000 of that allocation would be avoided within 12 months if the product were retired. Exit and inventory costs are estimated at $30,000, and some demand may transfer to a higher-contribution substitute.
一个模拟商品在共享间接费用前有 80,000 美元退货后贡献，分配 100,000 美元仓库与管理成本后为 −20,000 美元。如果 12 个月内退出，仅有 15,000 美元分配成本可避免，退出与库存成本估计 30,000 美元，部分需求可能转向贡献更高的替代商品。
| Synthetic decision line模拟决策项目 | Amount金额 | Behavior行为 | Implication含义 |
|---|---|---|---|
| After-returns direct contribution退货后直接贡献 | +$80,000 | Lost if all demand disappears需求全部消失则损失 | Retiring is not automatically beneficial退出不自动有利 |
| Allocated shared cost分配共享成本 | −$100,000 | Only $15,000 avoidable in horizon期间内仅 15,000 可避免 | Do not assume full saving不要假设全部节省 |
| Exit and inventory cost退出与库存成本 | −$30,000 | One-time and timing-sensitive一次性且对时间敏感 | Include transition case纳入过渡情景 |
| Substitution contribution替代贡献 | Scenario | Requires behavior evidence需要行为证据 | Test assortment change测试选品变更 |
The fully allocated view correctly signals a question but does not prove retirement improves total profit. The business should compare fixing, repricing, routing, limiting and retiring under avoidable-cost and substitution scenarios. A limited assortment or promotion test can reveal demand transfer and service effects before an irreversible exit.
完全分配视图正确提出问题，但不能证明退出会提高总利润。业务应在可避免成本与替代情景下比较修复、改价、路由、限制与退出。有限选品或促销测试可在不可逆退出前揭示需求转移与服务影响。
All store figures and records in this example are synthetic. They illustrate the method and do not represent InfiniSynapse customer results or industry benchmarks.本示例中的商店数字与记录均为模拟，仅用于说明方法，不代表 InfiniSynapse 客户结果或行业基准。
Classify the problem before choosing the option选择方案前先分类问题
A negative product can reflect price, discount, acquisition, forward fulfillment, returns, reverse cost, poor recovery, support load, inventory aging, allocation, or low volume. Each has a different remedy and side effects. Preserve the driver bridge so decision owners can see what must change and what remains fixed.
商品为负可能来自价格、折扣、获客、正向履约、退货、逆向成本、回收差、客服负荷、库存老化、分配或低销量。每种问题有不同补救与副作用。应保留驱动核对桥，让决策负责人看到必须改变与保持不变的内容。
A testable product, content, fulfillment, return, recovery or service driver creates avoidable loss.可测试商品、内容、履约、退货、回收或服务驱动造成可避免损失。
Contribution fails under specific price, discount, channel or acquisition terms.贡献在特定价格、折扣、渠道或获客条件下失效。
Low volume cannot cover step costs or the product displaces a better constrained use.低销量无法覆盖阶梯成本，或商品挤占更优的受限用途。
The product appears negative from shared cost that will remain under the contemplated decision.商品因拟议决策下仍会存在的共享成本而显得为负。
Retain a separate “insufficient evidence” state. A product with missing COGS, recovery, cost invoices, inventory or customer-role data is not proven profitable or unprofitable.
保留独立“证据不足”状态。缺少商品成本、回收、成本发票、库存或客户角色数据的商品，既未证明盈利，也未证明亏损。
Run ten controls before changing the assortment改变选品前完成十项控制
- Ledger reconciliation: sale, return, refund, cost, disposition and recovery records tie to source totals.台账核对：销售、退货、退款、成本、处置与回收记录与来源总数一致。
- Complete coverage: missing COGS, direct costs, late invoices, recovery and unmatched lines remain visible.完整覆盖：商品成本、直接成本、迟到发票、回收缺失与未匹配行保持可见。
- Mature and representative period: season, launch, promotion, policy and return window are controlled.成熟代表期间：控制季节、上市、促销、政策与退货窗口。
- Unit and total views: contribution per fulfilled/retained unit and total contribution are both shown.单位与总量视图：同时展示每履约/保留件贡献与总贡献。
- Cost behavior: direct, allocated, variable, step, committed, sunk and exit costs are labeled.成本行为：标记直接、分配、变动、阶梯、已承诺、沉没与退出成本。
- Time horizon: decision, contract, capacity and cost-change dates are explicit.时间跨度：明确决策、合同、产能与成本变化日期。
- Inventory transition: on-hand/inbound, aging, cancellation, liquidation, donation and disposal are modeled.库存过渡：模拟现有/在途、库龄、取消、清算、捐赠与处置。
- Customer interactions: attachment, substitution, acquisition, retention, accessibility and harm are assessed.客户交互：评估连带、替代、获客、留存、可访问性与伤害。
- Sensitivity: allocation, recovery, volume, price, rate, substitution and timing ranges are tested.敏感性：测试分配、回收、销量、价格、退货率、替代与时间范围。
- Reversible governance: owner, approval, test, guardrails, rollback and realized outcome are recorded.可逆治理：记录负责人、批准、测试、护栏、回滚与已实现结果。
Avoid nine unprofitable-product decision mistakes避免九个亏损商品决策错误
- Treating a customer-selected reason as a verified root cause.把客户选择的原因当作已核验根因。
- Combining customer reason, observed condition, disposition, and refund outcome in one field.把客户原因、观察状态、处置与退款结果混在一个字段。
- Changing code labels without versioning or remapping historical records.更改代码标签却不进行版本化或映射历史记录。
- Ranking percentages without counts, eligible denominators, value, or uncertainty.只按百分比排序，不展示数量、合格分母、金额或不确定性。
- Comparing products, channels, or periods with different question wording and missingness.比较问题措辞与缺失程度不同的商品、渠道或期间。
- Discarding “other,” free text, multi-reason, changed, or unknown responses.丢弃“其他”、自由文本、多原因、已更改或未知回答。
- Acting on correlation before reviewing cases and testing a mechanism.在检查案例并测试机制前就依据相关性行动。
- Assuming all allocated overhead disappears when one product is discontinued.假设下架一个商品后所有分配间接费用都会消失。
- Ignoring demand substitution, basket contribution, inventory liquidation, commitments and transition costs.忽略需求替代、购物篮贡献、库存清算、承诺与过渡成本。
Separate the diagnosis (“which driver makes this view negative?”) from the decision (“which feasible option improves total value?”). Record expected and realized effects so the portfolio process can learn instead of repeatedly trusting forecasts.
把诊断（哪个驱动使该视图为负）与决策（哪个可行方案改善总价值）分开。记录预期与已实现效果，让商品组合流程能够学习，而不是反复相信预测。
Test the least irreversible high-value option first优先测试最可逆且高价值的方案
Choose a product whose negative result persists across credible coverage and allocation scenarios. Identify the largest controllable driver and compare fix, price, promotion, source, route, bundle, policy, service, capacity, limit and retirement options. Start with a bounded reversible test when safety and legal obligations allow, then evaluate mature returns, contribution, inventory, capacity, customer and portfolio effects.
选择在可信覆盖与分配情景下负结果仍持续的商品。识别最大可控驱动，并比较修复、定价、促销、采购、路由、捆绑、政策、服务、产能、限制与退出方案。在安全与法律义务允许时，从有限可逆测试开始，再评估成熟退货、贡献、库存、产能、客户与商品组合影响。
| Signal信号 | Evidence to check待检查证据 | Safe next step安全下一步 |
|---|---|---|
| Negative after returns; one reason dominates退货后为负；一个原因占主导 | Reason/inspection coverage, mechanism, economics, intervention cost, guardrails原因/质检覆盖、机制、经济、干预成本、护栏 | Test the specific fix before retirement退出前测试具体修复 |
| Negative only after shared allocation仅在共享分配后为负 | Avoidable costs, step thresholds, capacity alternative, horizon, demand transfer可避免成本、阶梯阈值、产能替代、时间跨度、需求转移 | Use incremental scenario, not full allocation saving使用增量情景而非全部分配节省 |
| Negative direct contribution with severe harm直接贡献为负且伤害严重 | Safety/compliance, customer impact, inventory, obligations, alternatives安全/合规、客户影响、库存、义务、替代 | Escalate promptly; economics is secondary及时升级；经济性次要 |
Prepare a product decision evidence file准备商品决策证据文件
Export stable product/variant and bundle identity; sale and return cohorts; net sales, sale-time COGS, forward/return fulfillment, payment and marketplace fees, handling, support, write-down and realized recovery; contribution per unit and total; direct/allocated and variable/step/committed/sunk/exit/transition cost labels; currency and recognition status; inventory and purchase commitments; capacity use; attachment, substitution and customer role; coverage; scenario assumptions; options; decision owner, approval, test, guardrails, rollback and realized outcome. Return Compass can compare governed scenarios; it cannot approve accounting or assortment decisions.
导出稳定商品/变体与捆绑身份；销售与退货群组；净销售额、销售时商品成本、正向/退货履约、支付与平台费、处理、客服、减值与已实现回收；每件与总贡献；直接/分配及变动/阶梯/已承诺/沉没/退出/过渡成本标签；币种与确认状态；库存与采购承诺；产能使用；连带、替代与客户角色；覆盖；情景假设；方案；决策负责人、批准、测试、护栏、回滚与已实现结果。逆向罗盘可比较受治理情景，但不能批准会计或选品决策。
Open Return Compass打开逆向罗盘 →Identify Unprofitable Products FAQ识别亏损商品常见问题
Reconcile after-returns contribution by stable product, show total and per-unit results, classify avoidable and allocated costs, publish coverage, and compare realistic decision scenarios including inventory, capacity and customer effects.按稳定商品核对退货后贡献，展示总量与每件结果，分类可避免与分配成本，发布覆盖，并比较包含库存、产能与客户影响的现实决策情景。
Not automatically. Determine which allocated costs actually disappear, include exit and transition costs, assess substitution and portfolio effects, and compare reversible alternatives.不能自动下架。应判断哪些分配成本真正消失，纳入退出与过渡成本，评估替代及商品组合影响，并比较可逆方案。
Use the decision scope: sale-time COGS, discounts/reversals, payment and marketplace fees, forward and reverse fulfillment, handling, support, write-down, recovery, and clearly classified shared or transition costs.按决策范围使用销售时商品成本、折扣/冲回、支付与平台费、正向与逆向履约、处理、客服、减值、回收及清晰分类的共享或过渡成本。
Possibly through acquisition, attachment, substitution, retention, accessibility or strategic coverage, but association is not proof. Use matched or experimental evidence and include the effect in scenarios.可能通过获客、连带、替代、留存、可访问性或战略覆盖发挥作用，但关联不是证明。应使用匹配或实验性证据，并把影响纳入情景。
Test product/content fixes, repricing, promotion changes, bundling, sourcing, routing, return policy/service design, capacity changes, channel or customer limits, and monitored seasonal availability.可测试商品/内容修复、改价、调整促销、捆绑、采购、路由、退货政策/服务设计、产能变更、渠道或客户限制，以及受监控季节供应。
Sources, evidence labels, and limitations来源、证据标签与限制
- Shopify Help Center: Profit reports — Official documentation for product- and variant-level gross profit reporting, cost-per-item prerequisites, discounts, and refunds.Shopify 帮助中心：利润报告——关于商品及变体级毛利润报告、单件成本前提、折扣与退款的官方文档。
- Shopify Help Center: Analytics data points reference — Official definitions for gross profit, net sales, sales reversals, and cost-recording coverage in Shopify analytics.Shopify 帮助中心：分析数据字段参考——Shopify 分析中毛利润、净销售额、销售冲回与成本记录覆盖的官方定义。
- Microsoft Learn: Sales returns in Dynamics 365 Supply Chain Management — Official documentation for customer-selected reason codes, reason groups, return-line references, disposition actions, and inventory/credit implications.Microsoft Learn：Dynamics 365 供应链销售退货——客户选择原因码、原因组、退货行关联、处置动作以及库存/贷项影响的官方文档。
- Microsoft Learn: Return cost price and return lot ID — Official documentation showing why a return should reference the original sales line and cost at sale rather than silently using an unrelated current item cost.Microsoft Learn：退货成本价与退货批次 ID——官方文档说明退货应关联原销售行与销售时成本，而不是静默使用无关的当前商品成本。
- Shopify Help Center: Using SKUs to manage inventory — Official guidance that SKUs are internal inventory and reporting identifiers and should be unique for each product variant.Shopify 帮助中心：使用 SKU 管理库存——官方指南说明 SKU 是内部库存与报告标识，每个商品变体应使用唯一 SKU。
Evidence statement: Official commerce and product-data documentation supports stable SKU and variant identity, return-workflow stages, and evidence separation. Shopify and Microsoft sources support cost, profit-reporting, return-line and recovery concepts; no universal profitability threshold or guaranteed decision outcome is asserted. Examples are synthetic. No customer result, universal threshold, causal claim, or guaranteed improvement is asserted. Sources were reviewed September 15, 2026; named subject-matter review is required before publication.证据声明：官方商业与商品数据文档支持稳定的 SKU 和变体身份、退货工作流阶段与证据分离。Shopify 与 Microsoft 来源支持成本、利润报告、退货行与回收概念；本文不声称通用盈利阈值或保证决策结果。示例为模拟。本文不声称客户结果、通用阈值、因果结论或保证改善。来源核验于 2026 年 9 月 15 日完成；发布前需要具名领域审核。
Diagnose the loss, then compare feasible options先诊断亏损，再比较可行方案
Identify products from reconciled after-returns contribution, not revenue, refund amount, or allocated margin alone. Separate total and unit economics, avoidable and committed costs, missing data, inventory transition, capacity, and customer/portfolio effects. Treat a negative result as a decision question. Compare multiple scenarios, test the least irreversible valuable option, and measure realized outcomes before expanding or retiring the product.
应根据已核对退货后贡献识别商品，而不是只看收入、退款金额或分配利润。区分总量与单位经济、可避免与已承诺成本、缺失数据、库存过渡、产能及客户/商品组合影响。把负结果视为决策问题，比较多个情景，测试最可逆且有价值的方案，并在扩大或退出商品前衡量已实现结果。