大数据分析的速度.pptx

大数据分析的速度以手机网络数据获取匿名预付费用户群的基本信息及使用偏好以进行绅分营销日程分析无处不在一切有他自己的时间分析速度的价值分析和灵活案例在中国市场热数据变现实践2标注所有您的多渠道客户旅程，带有10条线路的条码, 并理解在哪和为什么您正在丢失销售一个单独的视觉化向您显示百万交易隐 藏的错误？在2天理解 10个影响您客户的事宜，他们对这些 感觉如何，您需要什么来解决他们？仅仅6周在您的分析数据投资上产生500% 回报?分析无处不在典型案例 – 基于实时数据Will Cage, TeradataHas used 90% of BundleOld model HandsetMedium Value CustomerJust Dropped a callPassing Orange ShopCell 0014Call into the Orange shop to see new phones and earn 10 points per 10Euro Top Up !Cell0004Cell 0005Cell 0013Cell 0003Cell 0002Cell 0012Sorry for the problems – tickets for you in the next hour !Cell 0001Cell 0011Has movie channel at homePassing movie theatreReal-time offerHigh Value CustomerJust Dropped a callCell 00108典型的时间和价值:客户服务, 只是及时道歉Resolution and notification before customer knowsBusiness Event – e.g., a Customer Service Incident Apology within 15 minutesApology within 3 hoursApology within 24 hoursTimeApology within 3 daysNo ApologyPerception of Resolution by CustomerGains ValueLoses Value典型的时间和价值: 客户呼吁或检查竞争，然后开始制造过程Customer is given a special offer in the retention platform or called right after churnEffectiveness of retention Campaign3x retention and early winback campaign effectivenessBeforeAfter3X62Time (hours)24 hoursEARLY WINBACK of the customers who called the competition is only effective within 24 hours during which the call must be discovered and acted uponCustomer calls or checks competitor and churns典型的时间和价值: 客户步入专项区域，此处有他感兴趣的零售提供Effectiveness of retention CampaignCustomer is within 20m of the cinemaT1: Time to discover the relevant eventT2: Time to relate it to locationT1T2Time (minutes)Customer walks inside the ring fenceRELEVANT LOCATION BASED OFFER requires spee in reaction to predefined rules but also fast discovery of relevant eventsCustomer is checking info about the Spiderman Premier分析提升的类型Enhance existing classification or predictive power with new methods that require new speedFast discovery of a new piece of data that explains and allows id