贝塔斯曼的销售分析报表（精品）.ppt

BBC Member Segmentation Summary * Member Segmentation Basic Rule - I Criteria: Purchase Frequency (activity rate of recent 4 cycle) Rev. Contribution (PKU of recent active cycle) (C135: data up to 03 Jun; C24: up to 03 May) Member stayed =3 commitment cycles = N=884K (62%) Best / Normal / Poor Member stayed = 2 commitment cycles = N=165K (12%) Best / Normal / Poor Member stayed = 1 commitment cycle = N=148K (10%) Best / Poor Member stayed = 0 cycle (join-in cycle) = N=222K (16%) Pending to track performance (ZP/IP) Total N=1.42Mio Effective MB * Member Segmentation Basic Rule - II Split member into groups by PKU / active cycle Activity rate% in recent 4 cycles Define Best / Normal / Poor group according to the avg. profitability of each sub-group Best: profit = RMB10; Normal: profit = RMB[0,10); Poor: profit RMB0 * Member Stayed More Than 3 Cycles TTL: 884K Effective MB Best: 300K (34%) Normal: 232K (26%) Poor: 351K (40%) Profit / MB # MB MB share% * Member Stayed 2 Cycles TTL: 165K Best: 79K (48%) Normal: 49K (30%) Poor: 37K (22%) Profit / MB # MB MB share% * Profit / MB # MB MB share% Member Stayed 1 Cycle TTL: 148K Best: 43K (29%) Normal: 31K (21%) Poor: 74K (50%) * Main Findings From Segmentation and Profile Analysis - I Member’s purchase behavior varies a lot across different age group Young members aged below 20 performed much better than the mature group in the first 2 cycles after recruitment. Meanwhile after stayed more than 3 cycles, these young members activity rate and PKU are worse. But mature group are more stable although the activity rate are not so high as young teens Member aged 21~25 is the most unstable group mainly due their address are not easy to deliver, so address maintenance is very important for all those school or college address City difference: COD cities performed quite poor, nearly 50% of member belong to poor group, especially NJ, TJ, Suzhou etc. Though some new COD