Data Analytics Journey Phase 3

After the modern data platform has been well-implemented, we can enhance the analytics with ML (Machine Learning) to get more insights.

Such as:

1. Customer segmentation by ML
2. Set up the scoring system
3. Recommendation Engine

Scoring to Customer and Segmentation

Customer scoring is based on the segmentation of your customer database

Customer scoring draws on customer data. The score assigned to each of your customers is calculated from the data you have on them. You can use all types of data to build a scoring system:

• Socio-demographic data: age, sex, marital status, profession
• Psychological data: interests, opinions
• Behavioral data: purchase history, data of last purchase, purchase frequency, Apps behavior, Campaign conversion rate, etc.

Example for Segmentation

Segmentation by Machine Learning

Found 4 types of customers

1. Low Annual income with Low spending score (Smart shopper are the one highly targeted by sales/coupons/promotion)
2. Low Annual income with High spending score (Satisfied Shopper are least interested in due to their spending habits)
3. Average income with Average spending score (Require more data to figure out their buying decision)
4. High income with Low spending score (Unsatisfied shopper with the mall’s service. These are our target study group as we need to attract theses shoppers to increase sales demand)

Recommendation Engine

Remarks: Location data and Push Notification action will be handled by a 3rd party vendor.

Case Reference: Delivery Optimization  (Chain store scheduler)

Example usages of Machine Learning:

• Customer Scoring
• Customer Segmentation
• Customer Behavior Analysis
• Data Profiling
• Abnormal Detection

#MachineLearning #ML #DataAnalyticsJourney #DataJourney #機器學習 #機器自學 #數據分析之旅 #數據之旅 #數據分析旅程 #數據分析旅程 #香港數據分析 #HongKongDataAnalytics #HongKongDataAnalysis