E-commerce website user Analysis

When the user has the purchase behavior on the electronic Commerce website, then becomes the value customer from the potential customer to the website. E-commerce Web site generally will users of the transaction information, including the purchase time, purchase goods, purchase quantity, payment amount and other information stored in their own database, so for these users, we can based on the operation of the site data on their trading behavior analysis to estimate the value of each user, and the possibility of expanding marketing (lead Generation) for each user.

　　Indicators to evaluate user value

For the selection of evaluation indicators This follows 3 principles:

1, the index can be quantified: no way, to do quantitative analysis, this is the most basic premise;

2. As comprehensive as possible: according to the underlying data to choose as many indicators as possible, so that from a multi-angle analysis and evaluation;

3, Linear Independence: that is, the indicators as far as possible to maintain irrelevant. For example, if you choose the number of users to buy and total consumption, then must be the more purchase number of users the total consumption is higher, that is, resulting in the evaluation dimension overlap, and the choice of purchase times and the average value of each transaction can avoid the drawbacks of this correlation.

Based on the above principles, select the following indicators (also based on the characteristics of the site to select the appropriate statistical time period):

1, the most recent purchase time: The user's most recent purchase from the current number of days;

2, Purchase Frequency: Users in this period of time to buy the number of times;

3, the average amount of each transaction: the user in this period of total consumption/purchase times;

4, a single maximum turnover: the user in this period of time to buy the highest amount of words paid;

5, buy the type of goods: the user in this period of time to buy the type of goods or large categories of goods.

　　Demonstration of user evaluation model

The same, can also be demonstrated by radar charts, but also using the deviation of the standardized method for each indicator to eliminate the unit of measurement of the 10 score. The following is an example of a radar map:

Through this radar map, we can read more information than user loyalty. The above 3 indicators-the latest time to buy, the frequency of purchases, and the type of merchandise you buy-can be used to evaluate user loyalty, and the following 2 indicators-the average value of each transaction and word maximum can be used to measure the user's spending power. As above, the user 1 although the purchase frequency and the breadth of the purchase is not high, but its ability to consume a strong, and 2 users are frequent buyers, the site has a certain degree of loyalty, but its consumption capacity in general. Therefore, the upper half of the graphic user has a higher degree of loyalty, while the lower part of the larger area of users with higher consumption capacity. These two types of users are valuable customers of the site, but due to their different types of marketing strategies can be treated separately.

　　The significance of user transaction behavior analysis

1, found the site's High-value customers (VIP), Customer Relationship Management (CRM) and maintain a valuable customer support;

2, explore the Web site can develop users, for some new customers or potential customers for targeted marketing;

3, timely detection of potential loss of customers, timely adoption of effective measures;

4, according to user transactions to subdivide customer base, the implementation of targeted marketing strategy.

Source: Website Data analysis Contributor: Joegh