起步顺序Order of work 先做库存感知的推荐,再做批发订货助手,最后做门店级预测。Stock-aware recommendation first, then wholesale ordering help, then store-level forecasting.
公开事实Public facts 已核对Checked
- C$7.37B2025 财年营收Revenue, fiscal 20252025 财年年报Annual report FY2025
- 690全省门店Stores across Ontario2025 财年FY2025
- 1M官网每周访问量(约)Weekly website visits, about其数字官在 Coveo 访谈中所述Per its digital chief, Coveo interview
- 531提供当日自提的门店Stores with Same-Day Pickup2024 财年FY2024
需要完成的三个转变Three shifts to make 判断Judgment
- 统计搜索点击Counting search clicks → 统计完成的订单Counting orders completed
- 推荐任何商品Recommending any product → 推荐我的门店有货的商品Recommending what is in stock at my store
- 供应商常见问题机器人A supplier F A Q chatbot → 面向批发客户的订货帮手Ordering help for wholesale buyers
专家组裁决Expert panel 依据公开观点的推演Inference from public views
Erik Brynjolfsson
统计完成的订单而非点击;设基线与对照组。Count orders completed, not clicks; set baselines and a control group.
Ethan Mollick
用真实门店库存测出推荐的失败点;员工反馈边界。Test where recommendations fail on real store stock; staff report the edge.
Ziad Obermeyer
写明真实目标,不只是销量;负责任销售规则请与法务确认。Write the real target, not only sales; confirm responsible-sales rules with counsel.
Andrew Ng
先选一个:一个区域的库存感知推荐;三到六个月出数字。Pick one: stock-aware recommendations in one region; numbers in three to six months.
Thomas Davenport
网站、门店与批发要连起来;点名跨渠道业务负责人。This joins the website, stores and wholesale. Name a business owner across them.
Foteini Agrafioti
省属采购流程较长;动手前先把采购规则写成一页。Crown procurement runs early; put purchasing rules on one page before any build.
Ethan Mollick:现在就在一个区域用真实库存测试。Put it on real stock in one region now.
Ziad Obermeyer:酒类个性化之前,先写清目标。Name the target before personalising alcohol.
→ 一个区域测试;放大前须过法务与负责任销售评审。Test in one region; counsel and responsible-sales review gate roll-out.
六位专家是谁、怎样思考、有什么铁律 →Who the six experts are, how they think, their rules →
三个目标Three priorities 建议Recommendation
让推荐买得到Make recommendations buyable
在个性化推荐中加入门店库存,并统计订单而非点击。Add store stock to personalised recommendations, and track orders, not clicks.
帮批发客户订货Help wholesale buyers order
面向餐厅、酒吧与杂货店的订货助手,复用供应商机器人的经验。An ordering assistant for restaurants, bars and grocers, reusing supplier-chatbot lessons.
按门店做预测Forecast by store
按门店、季节与活动做预测,先确认采购规则。Store, season and event forecasts, with procurement rules checked first.
场景组合Scenarios 建议Recommendation
| 场景Scenario | 做法Approach | 状态Status |
|---|---|---|
| 库存感知推荐Stock-aware recommendations 增收Revenue体验Experience | 只推荐顾客附近门店或自提可得的商品。Show only what is available for pickup or at the shopper's nearby store. | 建议 · 扩展已上线搜索Proposed · extends live search |
| 批发订货助手Wholesale ordering assistant 增收Revenue提效Efficiency | 帮助餐厅、酒吧与杂货店复购与选品。Help restaurants, bars and grocers reorder and choose products. | 建议Proposed |
| 带订单状态的供应商助手Supplier assistant with order status 提效Efficiency降风险Risk | 让供应商机器人在权限控制下查看真实订单状态。Let the supplier chatbot see real order status, with access controls. | 建议 · 目前无实时数据Proposed · today it has no live data |
| 门店与季节预测Store and season forecasts 提效Efficiency降风险Risk | 按门店、季节与活动,覆盖门店与区域仓库做销量预测。Forecast sales by store, season and event across stores and regional warehouses. | 建议 · 先核对采购规则Proposed · check procurement rules |
12 个月路线12-month roadmap 建议Recommendation
奠基Foundation
- 「订单完成」指标Order-complete metric
- 门店库存数据对接Store stock feed mapped
- 采购规则一页Procurement rules on one page
区域测试Region test
- 一个区域库存感知推荐Stock-aware picks in one region
- 对照组测量Control group measured
- 员工用真实库存测试Staff test on real stock
上线Go live
- 全省推广Roll out province-wide
- 批发助手试点Wholesale assistant pilot
- 负责任销售评审Responsible-sales review
扩展Extend
- 门店级预测Store-level forecasts
- 供应商订单状态Supplier order status
- 年度重排优先级Annual re-ranking
治理与合规Governance 建议Recommendation
- 负责任销售与隐私规则请与法务确认Confirm responsible-sales and privacy rules with counsel
- 动手前核对采购与省属机构规则Procurement and Crown rules checked before any build
- 每个面向客户的助手都有专人负责A person owns each customer-facing assistant
三层指标Three-layer metrics 建议Recommendation
| 采用Adoption | 使用个性化推荐的顾客Shoppers using personalised picks · 使用助手的批发客户Wholesale buyers using the assistant |
|---|---|
| 过程Process | 推荐商品有货的比例Share of picks in stock · 复购耗时Time to reorder · 助手转人工比例Assistant escalation rate |
| 结果Outcome | 对比基线的完成订单Orders completed against baseline · 对比基线的缺货Stock-outs against baseline |
目标值由企业自定。Targets are the company's to set.
风险Risks 判断Judgment
推荐没货的商品Recommending what is not there
推荐无货商品,比不推荐更伤信任。Out-of-stock picks cost trust faster than no picks.
把点击当成成果Clicks counted as wins
搜索点击上升,但订单才是关键数字;先设基线。Search clicks rose, but orders are the number that matters; set baselines.
个性化优化了错误目标Personalising the wrong target
只优化销量可能与负责任销售冲突;请与法务确认。Optimising sales alone may clash with responsible-sales goals; confirm with counsel.
前 90 天First 90 days 建议Recommendation
| 第 1–2 周Weeks 1–2 | 点名网站、门店、批发的业务负责人Name a business owner across web, stores, wholesale |
|---|---|
| 第 3–4 周Weeks 3–4 | 定义「订单完成」指标与基线Define the order-complete metric and baseline |
| 第 5–10 周Weeks 5–10 | 对接门店库存;选一个区域Map store stock feed; pick one region |
| 第 13 周Week 13 | 首次复盘First review |
我们可能错在:We could be wrong if: 门店库存数据太滞后,撑不起推荐;或顾客上网站主要是浏览而非购买。store stock data is too stale to power recommendations, or shoppers use the site to browse, not buy.
出处Sources
本页是基于公开资料的分析,与所述企业没有合作关系,不代表其官方立场或内部计划;企业数字多来自其自述或媒体转述;「专家组」是依据六位专家公开观点的推演,不是他们的原话,也不代表他们或其雇主的立场。资料整理日期:2026-11-14。This page is an analysis based on public information. It has no affiliation with the company and does not represent its official position or internal plans. Company figures come mostly from its own statements or media reports. The expert panel is an inference from six experts' public views, not their words, and does not represent them or their employers. Compiled 2026-11-14.
想给你的公司也做一份?Want one for your company?
同样的方法:公开事实、专属判断、12 个月路线和前 90 天。公司大小都可以聊,中文沟通。Same method: public facts, a company-specific judgment, a 12-month roadmap and the first 90 days. Any size of business; we can talk in Chinese or English.