Campaign Manager - Segments
Resolution LevelsThe resolution table is a key concept for the Alterian Marketer and Analyst. Quite simply, any count of records in any Alterian application is a count of a number of rows on a specific table and each table is a logical groups of records. The Customer table is 1 row per Customer, an Order or Basket table is one row per Order etc. So if you set a table of resolution of Customer on a Segment document, the final result of that segment will be a count of Customers. A database will have many tables and these are linked to allow queries across these tables. A "higher" level database level has a one-to-many link with a lower database level, and a lower database level a many-to-one link with a higher database level. Therefore, when you change your resolution level your count may increase or decrease. In a typical database, the database levels could be structured with Household at the top of the hierarchy and Order Details at the bottom. A Household can contain more than one customer, each customer can make more than one order, and each Order may contain more than one item (Order Detail). Therefore, when changing the resolution level from Household to Order you would expect your count to increase, and when changing the resolution level from order to customer you would expect your count to decrease. In the example here there are 4 tables, Household, Customer, Order and Order Detail. Very logically a Household can have more than one Customer, so that is a one-to-many link. A Customer can have placed more than one Order, another one-to-many link and an Order can contain more than one item (Order Detail) another one-to-many link. It is important to be sure of this question at any stage of analytics, What am I counting?
The following shows a query with the top level table of resolution, seen after the "Applies to" table, set to Customer. There are 12,876 Customers between 25 and 35.
In the above screenshot, the count on the right appears twice, both showing 12,876 Customers. The top count is for the entire document, the lower count is against that specific row in the document. Say I now want Customers between 25-35 who have an Income Band of 30-40K. I add a second rule and see clear definition of the counts for each row and their table of resolution.
The count of Customers who meet both criteria is 391, but each Customer table count for each row is also visible and this is the isolated count for that single row and not a running total. So there are 26, 654 Customers with an Income between 30-40K regardless of tiier age or any other considerations. This is very powerful as you can see the units that go to make up your total. The power of the Table of resolution comes when looking at different tables in conjunction and when using Groups. Say the required result is now Customers aged 25-35 with an Income Band of 30-40K who placed an order in 2017 with a value greater than 50.00 There are two Order criteria here, firstly that the Order must have been made in 2017 and secondly that the Order was of a value > 50.00. These are both Order table columns so must be AND'ed or combined at the Order level first to give a count of Orders hat meet both criteria. At this point those Order can be counted at the Customer level to see the Customers who places those Order. The screenshots below shows this clearly but some further grousing has been introduced as there are two tables in consideration:
The Segment Document This must always bet set first and is simply what you want to count in this segment, i.e. 168 Customers
Customer Group As there are two rules both at Customer, i.e. Age Band 25-35 and Income Band 30-040K, we must ensure these are combined at the Customer table so create a Group with a Table of Resolution of Customer.
Order Group As there are two rules both at Order, i.e. Order Year YYYY = 2017 and Order Value > 50.00, we must ensure these are combined at the Order table so create a Group with a Table of Resolution of Order. The counts are clearly labeled.
So the final segment is a count of Customer who are between 25 & 35, have an Income Band of 30-40K and placed and Order in 2017 of a value greater than 50.00, bu the counts are each stage are clear and visible.
Document Breakdown 168 - Count of Customers who meet the Segment document rule definition. 391 - Count of Customer who meet the rule definition of the Customer Group i.e. Customers Age 25-35 and Income Band 30-40K 12,876 - Count of Customers Age 25-35 26,654 - Count of Customers with Income Band of 30-40K 63, 872 - Count of Orders that meet the rule definition of the Order Groups i.e. Order placed in 2017 with Order Value > 50.00 108,799 - Count of Orders in 2017 149,017 - Count of Orders with a value > 50.00
|
|
© Alterian. All Rights Reserved. | Privacy Policy | Legal Notice |   
|