企业 AI 转型有三重障碍：人才缺口、信任黑箱、非标准需求。数治网围绕“做正确的事 → 正确地做事 → 把事做得正确”，提供按职能场景、按行业定制、按成熟度阶段的三套入口，并附 7 大核心业务场景的课程映射速查。每一个方案都回答三个问题：技术上能不能做到？组织上愿不愿意接受？商业上值不值得做？Enterprise AI transformation faces three barriers: talent gap, trust black box, and non-standardized needs. Around the narrative "Do the right thing → Do things right → Do things correctly," shuzhi.me offers three entries — by function, by industry, by maturity stage — plus a 7-scenario course-mapping quick reference. Every solution answers: Can it technically be done? Will the organization accept it? Is it commercially worth it?
不是缺工具，是缺“既懂业务又懂 AI”的复合型人才。传统培训学完了用不上，用上了没法衡量效果。六大职能 × 三赛道 = 18 个认证序列、162 个技能节点，让人才培养直接对齐业务需求。It is not a tool shortage — it is a hybrid talent shortage. Training without application, application without measurable results. Six functions × three tracks = 18 sequences, 162 skill nodes, aligning talent development with business needs.
从经验驱动到数据驱动、从事后救火到事前预警。认证序列：战略型（运营数字化总监）/ 应用型（智能运营专家）/ 创新型（智慧运营场景设计师）。典型技能：流程自动化、智能排产、经营预警、RPA、BI、IoT。From experience-driven to data-driven, from firefighting to early warning. Tracks: Ops Digital Director / Intelligent Ops Expert / Smart Ops Scenario Designer. Skills: process automation, smart scheduling, operational early warning, RPA, BI, IoT.
从创意经验到数据驱动的精准营销、从单点创作到 AIGC 规模化内容生产。认证序列：战略型（营销数字化总监）/ 应用型（AI 增长营销分析师）/ 创新型（智能营销 / AIGC 内容设计师）。典型技能：CDP、MA、归因建模、客户旅程地图。From creative experience to data-driven precision marketing, from single-point creation to AIGC-scaled production. Tracks: Marketing Digital Director / AI Growth Marketing Analyst / Intelligent Marketing - AIGC Content Designer. Skills: CDP, MA, attribution modeling, journey mapping.
从成本中心到价值中心，线索跟进、客服体验、质检全覆盖。认证序列：战略型（销售数字化总监）/ 应用型（智能客服训练师）/ 创新型（对话式销售 / 客户体验设计师）。典型技能：CRM、线索评分、对话式 AI、NLP 质检、RAG 知识库。From cost center to value center. Tracks: Sales Digital Director / Intelligent CS Trainer / Conversational Sales - CX Designer. Skills: CRM, lead scoring, conversational AI, NLP QC, RAG knowledge base.
从功能交付到价值验证，缩短交付周期、降低缺陷率。认证序列：战略型（研发数字化总监）/ 应用型（AI 研发效能专家）/ 创新型（AI 原生产品经理）。典型技能：AI 辅助编程、智能测试、A/B 测试、CI/CD。From feature delivery to value validation. Tracks: R&D Digital Director / AI R&D Efficiency Expert / AI-Native Product Manager. Skills: AI-assisted programming, smart testing, A/B testing, CI/CD.
从事后核算到事前预测、从手工报表到智能分析、从合规被动到风险主动预警。认证序列：战略型（财务数字化总监）/ 应用型（智能财务专家）/ 创新型（业财一体 / 数据资产模式设计师）。典型技能：RPA、财务预测建模、数据资产估值、智能风控。From post-event accounting to pre-event prediction, from manual reports to intelligent analysis. Tracks: Finance Digital Director / Intelligent Finance Expert / Business-Finance Integration Designer. Skills: RPA, forecast modeling, data asset valuation, smart risk control.
从事务 HR 到数据 HR、从管人到管人才资本。认证序列：战略型（HR 数字化总监）/ 应用型（人才数据分析师）/ 创新型（组织体验与内部知识生态设计师）。典型技能：AI 辅助招聘、人才流失预测、技能市场运营、组织能力诊断。From transactional HR to data HR, from managing people to managing talent capital. Tracks: HR Digital Director / Talent Data Analyst / Org Experience - Knowledge Eco Designer. Skills: AI-assisted recruiting, churn prediction, skill market ops, org diagnosis.
不是通用模板，是按行业痛点定制的专属动态技能树。Not generic templates — industry-pain-point-customized Dynamic Skill Trees.
痛点：设备故障停机损失大、质检依赖人工经验（漏检率 3%+）、OT/IT 数据孤岛。价值：预测性维护降低停机 40%+、AI 质检提升良品率 2–5 个百分点、数字孪生优化排产 +25%。技能树：工业数据意识 → 预测性维护 → AI 质检 → 数字孪生 → 工业安全。Pain: downtime = millions per incident, human-reliant QC (3%+ miss rate), OT/IT silos. Value: predictive maintenance cuts downtime 40%+, AI QC lifts yield 2-5 pts, digital twins optimize scheduling +25%. Skill tree: industrial data awareness → predictive maintenance → AI QC → digital twins → industrial safety.
痛点：风控模型更新慢、合规报告人工编制耗时、流失预警滞后。价值：实时风控评分（毫秒级）、自动化合规报告（效率 ×10）、智能投顾提升客户 AUM 15%+。技能树：金融数据分类 → 风控模型 → 智能投顾 → 金融合规风险 → 模型风险。Pain: monthly risk-model updates, 2-day manual compliance reports, lagging churn alerts. Value: millisecond real-time risk scoring, 10x automated compliance, intelligent advisory lifts AUM 15%+. Skill tree: financial data classification → risk models → intelligent advisory → compliance risk → model risk.
痛点：库存周转慢、会员数据沉睡（80% 会员 90 天未互动）、线上线下割裂。价值：需求预测优化库存周转 +30%、RFM 分层激活复购 +20%、全渠道打通客单价 +15%。技能树：租户 RFM 分析 → 需求预测 → 全渠道运营 → 供应链优化 → 智能定价。Pain: 60+ day inventory turnover, 80% dormant members, disconnected channels. Value: demand forecasting lifts turnover +30%, RFM segmentation lifts repurchase +20%, omnichannel data lifts basket +15%. Skill tree: tenant RFM analysis → demand forecasting → omnichannel ops → supply chain optimization → smart pricing.
商业地产行业服务频道（含六职能场景导航与案例） →CRE Industry Service Channel (scenario navigation and cases) →
从“做正确的事”到“把事做得正确”——每一步都有数治网陪你。不是让企业一步到位，而是帮企业找到当前阶段最该做的事。L1→L5，每一步都有对应的测评、培养、认证、验证方案。From "Do the right thing" to "Do things correctly" — shuzhi.me at every step. Not leaping forward overnight, but finding the most important thing to do at your current stage.
| 级别Level | 阶段名称Stage | 核心问题Core Question | 数治网对应方案shuzhi.me Solution | 主线一映射Narrative 1 | 主线二映射Narrative 2 |
|---|---|---|---|---|---|
| L1 | 探索者Explorer | “AI 能不能做？”——个别部门试点，缺方法论"Can AI do it?" — pilots without methodology | 能力测评诊断包 + 2 职能试点Diagnostic Package + 2-function pilot | 做正确的事（认知升级）Do the right thing | 从听说到明白From hearing to understanding |
| L2 | 应用者Applicator | “AI 怎么做？”——有成功单点项目，缺复制能力"How to do AI?" — successful pilots, lacking replication | 定制培养包 + P2 项目交付Custom Training + P2 delivery | 正确地做事（行为升级）Do things right | 从知道到做到From knowing to doing |
| L3 | 适配者Adapter | “AI 怎么复制？”——成功经验可跨团队复制"How to replicate AI?" — success replicable across teams | 认证运营包 + 技能市场上架Certification Ops + Skill Market listing | 正确地做事（行为升级）Do things right | 从知道到做到From knowing to doing |
| L4 | 构建者Builder | “AI 怎么持续优化？”——常态化运营，缺价值度量"How to optimize AI?" — normalized, lacking value metrics | 价值验证包 + 业务价值仪表盘Value Validation + value dashboard | 把事做得正确（能力升级）Do things correctly | 从纸面到地面From paper to ground |
| L5 | 架构师Architect | “AI 怎么驱动增长？”——组织级智能驱动，缺生态"How to drive growth with AI?" — org-wide, lacking ecosystem | 生态共建包 + 组织变革咨询Ecosystem Partnership + org change consulting | 把事做得正确（能力升级）Do things correctly | 从纸面到地面From paper to ground |
7 大核心业务场景——每个痛点都有对应的课程、认证出口与预期成效。Seven core business scenarios — every pain point has a matching course, certification exit, and expected outcome.
| 业务场景Scenario | 核心痛点Pain Point | 关联课程Courses | 认证出口Exit | 预期成效Expected Outcome |
|---|---|---|---|---|
| 数据质量差Poor data quality | 数据孤岛、口径不统一Silos, inconsistent definitions | ADG-001, ADA-004, CVI-101 | DLP | 数据可用率 +40%Data availability +40% |
| 合规压力大Compliance pressure | GDPR/个保法落地难GDPR/PIPL implementation challenges | BCF-201, BGDPR-202 | DLP→DLE | 合规审计通过率 100%Audit pass rate 100% |
| 客户流失高High churn | 缺乏客户洞察Lacking customer insight | ADA-004, CVI-101, CJM-102 | ALP→ALE | 流失率 -25%Churn rate -25% |
| 安全事件频发Security incidents | 安全体系不健全Immature security | BSO-203, BBCM-204 | DLE | 安全事件响应时间 -60%Response time -60% |
| AI 产品落地难AI landing difficulty | AI 应用停留在 POCStuck at POC | AFC-004, AGOV-006, CAI-104 | ALE | POC→生产周期 -50%POC-to-production -50% |
| 数字化转型推不动Transformation stall | 缺乏顶层设计Missing top-level design | ADG-001, BRM-205, CDEO-102 | DLE→DLM | 转型项目成功率 +35%Project success +35% |
| 数据资产无法量化Unquantified data assets | 数据入表无标准No assetization standard | ADM-006, ADG-002 | DLE | 数据资产估值可审计Auditable asset valuation |