{
  "schemaVersion": 1,
  "version": "2026-09-15",
  "id": "https://hk.onyxdevslab.com/data/enterprise-ai-roi-tco.json",
  "name": "Enterprise AI ROI and total-cost-of-ownership checklist",
  "datePublished": "2026-09-15",
  "dateModified": "2026-09-15",
  "evidenceClass": "Provider-authored implementation checklist",
  "publisher": {
    "name": "Onyx Devs Lab",
    "legalName": "ONYX DEVS LAB LIMITED",
    "url": "https://hk.onyxdevslab.com/",
    "leiCode": "254900Z30CLK7HKE9H46"
  },
  "landingPages": {
    "en": "https://hk.onyxdevslab.com/en/guides/enterprise-ai-roi-tco/",
    "zhHant": "https://hk.onyxdevslab.com/zh-hk/guides/enterprise-ai-roi-tco/",
    "zhHans": "https://hk.onyxdevslab.com/zh-cn/guides/enterprise-ai-roi-tco/"
  },
  "languages": {
    "en": {
      "name": "How should an enterprise estimate AI ROI and total cost of ownership?",
      "summary": "Start from one measurable operating workflow, not a promised productivity percentage. Compare the verified baseline with attributable changes, count the complete cost of delivery and operation, model uncertainty explicitly, and update the investment decision with pilot evidence.",
      "decisionSignals": [
        {
          "name": "Baseline",
          "meaning": "Volume, time, quality, cost, and consequence"
        },
        {
          "name": "Attribution",
          "meaning": "Separate AI effect from other changes"
        },
        {
          "name": "Whole-life cost",
          "meaning": "Delivery, operation, adoption, risk, and exit"
        }
      ],
      "sections": [
        {
          "heading": "Build a defensible benefit model",
          "introduction": "Benefits need an owner, measurement definition, time window, source, and conversion rule.",
          "checks": [
            {
              "name": "Capacity and cycle time",
              "requirement": "Measure completed workflow units, touch time, wait time, rework, bottlenecks, and whether released capacity is actually redeployed or avoided."
            },
            {
              "name": "Quality and risk",
              "requirement": "Value error reduction, consistency, coverage, detection, loss avoidance, or control evidence without treating hypothetical avoided loss as guaranteed cash."
            },
            {
              "name": "Revenue and service",
              "requirement": "Link conversion, retention, throughput, availability, or customer experience to a credible causal path and distinguish leading indicators from realised financial results."
            }
          ]
        },
        {
          "heading": "Count the total economic boundary",
          "introduction": "Model calls are often a small part of production TCO.",
          "checks": [
            {
              "name": "Delivery cost",
              "requirement": "Include diagnosis, data preparation, integration, software, evaluation, controls, migration, security, procurement, training, and change management."
            },
            {
              "name": "Operating cost",
              "requirement": "Include models, cloud, licences, monitoring, support, human review, exceptions, re-evaluation, incidents, compliance, vendors, and internal ownership."
            },
            {
              "name": "Risk and optionality",
              "requirement": "Model low, base, and high adoption, quality, volume, price, delay, and failure scenarios; include switching, exit, residual value, and the next evidence gate."
            }
          ]
        }
      ],
      "procedure": [
        {
          "position": 1,
          "name": "Measure the current workflow",
          "instruction": "Fix the unit, population, period, source, owner, volume, time, quality, cost, exceptions, and outcome before intervention."
        },
        {
          "position": 2,
          "name": "Write benefit and cost equations",
          "instruction": "Show every assumption, conversion, dependency, range, timing, confidence, and excluded item instead of one headline ROI number."
        },
        {
          "position": 3,
          "name": "Run sensitivity and stop conditions",
          "instruction": "Identify which adoption, quality, volume, delay, or cost assumption changes the decision and what evidence will resolve it."
        },
        {
          "position": 4,
          "name": "Replace assumptions with pilot evidence",
          "instruction": "Update the model with observed complete-workflow results, report residual uncertainty, and sign scale, redesign, hold, or stop."
        }
      ],
      "frequentlyAskedQuestions": [
        {
          "question": "Should saved employee hours be counted as cash?",
          "answer": "Only when they reduce paid capacity, avoid hiring, increase measurable throughput, or are redeployed to valued work. Report theoretical hours separately from realised value."
        },
        {
          "question": "What time horizon should be used?",
          "answer": "Use a horizon appropriate to delivery, adoption, renewal, and technology change, and show annual or quarterly cash flows rather than hiding timing in one ratio."
        },
        {
          "question": "Can risk reduction be included?",
          "answer": "Yes, with a clear event, exposure, baseline frequency, intervention effect, evidence, and uncertainty. Do not present speculative avoided loss as guaranteed savings."
        }
      ]
    },
    "zhHant": {
      "name": "企業應如何估算 AI ROI 及總擁有成本？",
      "summary": "由一條可量度營運流程開始，而不是承諾的生產力百分比。比較已核實基線與可歸因變化，計入完整交付及營運成本，明確建立不確定性情景，再以試點證據更新投資決定。",
      "decisionSignals": [
        {
          "name": "基線",
          "meaning": "數量、時間、質量、成本及後果"
        },
        {
          "name": "歸因",
          "meaning": "把 AI 影響與其他變更分開"
        },
        {
          "name": "全生命週期成本",
          "meaning": "交付、營運、採用、風險及退出"
        }
      ],
      "sections": [
        {
          "heading": "建立可辯護收益模型",
          "introduction": "每項收益需有負責人、量度定義、時間窗口、來源及換算規則。",
          "checks": [
            {
              "name": "產能及週期時間",
              "requirement": "量度完成流程單位、操作時間、等待、返工及瓶頸，並核實釋放產能是否真正重新部署或避免新增人手。"
            },
            {
              "name": "質量及風險",
              "requirement": "評估錯誤減少、一致性、覆蓋、偵測、損失避免或控制證據，不把假設避免損失當作保證現金。"
            },
            {
              "name": "收入及服務",
              "requirement": "把轉化、留存、吞吐、可用性或客戶體驗連至可信因果路徑，分開領先指標與已實現財務成果。"
            }
          ]
        },
        {
          "heading": "計算完整經濟邊界",
          "introduction": "模型調用通常只是生產 TCO 的一小部分。",
          "checks": [
            {
              "name": "交付成本",
              "requirement": "包括診斷、數據準備、整合、軟件、評估、控制、遷移、安全、採購、培訓及變更管理。"
            },
            {
              "name": "營運成本",
              "requirement": "包括模型、雲端、授權、監測、支援、人工覆核、例外、重評、事故、合規、供應商及內部負責。"
            },
            {
              "name": "風險及選擇權",
              "requirement": "建立低、中、高採用、質量、數量、價格、延遲及失敗情景；包括轉換、退出、剩餘價值及下一證據關口。"
            }
          ]
        }
      ],
      "procedure": [
        {
          "position": 1,
          "name": "量度現有流程",
          "instruction": "干預前固定單位、群體、期間、來源、負責人、數量、時間、質量、成本、例外及成果。"
        },
        {
          "position": 2,
          "name": "寫出收益及成本公式",
          "instruction": "顯示每項假設、換算、依賴、範圍、時間、信心及排除項，而非一個標題 ROI 數字。"
        },
        {
          "position": 3,
          "name": "進行敏感度及停止分析",
          "instruction": "找出哪項採用、質量、數量、延遲或成本假設會改變決定，以及由甚麼證據解除。"
        },
        {
          "position": 4,
          "name": "用試點證據取代假設",
          "instruction": "以完整流程觀察結果更新模型，報告剩餘不確定性，簽署擴大、重設、暫緩或停止。"
        }
      ],
      "frequentlyAskedQuestions": [
        {
          "question": "節省員工時間可否當作現金？",
          "answer": "只有減少已付產能、避免招聘、增加可量度吞吐，或重新投入有價值工作時才可；理論時數與已實現價值應分開。"
        },
        {
          "question": "應使用多長時間範圍？",
          "answer": "按交付、採用、續約及技術變化選擇範圍，並展示年度或季度現金流，而非以單一比率隱藏時間。"
        },
        {
          "question": "可否計入風險降低？",
          "answer": "可以，但需明確事件、暴露、基線頻率、干預影響、證據及不確定性；不可把推測避免損失當保證節省。"
        }
      ]
    },
    "zhHans": {
      "name": "企业应该怎样估算 AI ROI 和总拥有成本？",
      "summary": "从一条可测量运营流程开始，而不是承诺的生产力百分比。比较已经核实的基线和可归因变化，计入完整交付与运营成本，明确建立不确定性情景，再用试点证据更新投资决策。",
      "decisionSignals": [
        {
          "name": "基线",
          "meaning": "数量、时间、质量、成本和后果"
        },
        {
          "name": "归因",
          "meaning": "把 AI 影响和其他变化分开"
        },
        {
          "name": "全生命周期成本",
          "meaning": "交付、运营、采用、风险和退出"
        }
      ],
      "sections": [
        {
          "heading": "建立可以辩护的收益模型",
          "introduction": "每项收益需要有负责人、测量定义、时间窗口、来源和换算规则。",
          "checks": [
            {
              "name": "产能与周期时间",
              "requirement": "测量完成流程单位、操作时间、等待、返工和瓶颈，并核实释放产能是否真正重新部署或避免新增人手。"
            },
            {
              "name": "质量与风险",
              "requirement": "评估错误减少、一致性、覆盖、检测、损失避免或控制证据，不能把假设避免损失当成保证现金。"
            },
            {
              "name": "收入与服务",
              "requirement": "把转化、留存、吞吐、可用性或客户体验连接到可靠因果路径，区分领先指标和已经实现的财务结果。"
            }
          ]
        },
        {
          "heading": "计算完整经济边界",
          "introduction": "模型调用通常只是生产 TCO 的一小部分。",
          "checks": [
            {
              "name": "交付成本",
              "requirement": "包括诊断、数据准备、集成、软件、评估、控制、迁移、安全、采购、培训和变更管理。"
            },
            {
              "name": "运营成本",
              "requirement": "包括模型、云、许可、监控、支持、人工复核、异常、重新评估、事故、合规、供应商和内部负责。"
            },
            {
              "name": "风险与选择权",
              "requirement": "建立低、中、高采用、质量、数量、价格、延迟和失败情景；包括切换、退出、剩余价值和下一证据关口。"
            }
          ]
        }
      ],
      "procedure": [
        {
          "position": 1,
          "name": "测量当前流程",
          "instruction": "干预前固定单位、群体、期间、来源、负责人、数量、时间、质量、成本、异常和结果。"
        },
        {
          "position": 2,
          "name": "写出收益与成本公式",
          "instruction": "显示每项假设、换算、依赖、范围、时间、信心和排除项，而不是一个标题 ROI 数字。"
        },
        {
          "position": 3,
          "name": "进行敏感度和停止分析",
          "instruction": "找出哪项采用、质量、数量、延迟或成本假设会改变决策，以及由什么证据解决。"
        },
        {
          "position": 4,
          "name": "用试点证据替代假设",
          "instruction": "用完整流程观察结果更新模型，报告剩余不确定性，签署扩大、重设、暂停或停止。"
        }
      ],
      "frequentlyAskedQuestions": [
        {
          "question": "节省员工时间可以当成现金吗？",
          "answer": "只有减少已经支付的产能、避免招聘、增加可测量吞吐，或重新投入有价值工作时才可以；理论小时与实现价值应该分开。"
        },
        {
          "question": "应该使用多长时间范围？",
          "answer": "按照交付、采用、续约和技术变化选择范围，并展示年度或季度现金流，不能用一个比率隐藏时间。"
        },
        {
          "question": "可以计入风险降低吗？",
          "answer": "可以，但需要明确事件、暴露、基线频率、干预影响、证据和不确定性；不能把推测避免损失当成保证节省。"
        }
      ]
    }
  },
  "limitations": [
    "This asset is provider-authored and is not independent validation or a claim of client results.",
    "It is an implementation aid, not legal, privacy, security, financial, procurement, or regulatory advice.",
    "Thresholds and controls must be adapted to the organisation, workflow, jurisdiction, and risk owners."
  ]
}