Forum Discussion
Customer segmentation mix over time
I'm working with a sales transaction dataset with unique customer IDs for each customer, and want to graph customer segments overtime (defined various ways, but to start off I want to split into 2 segments - HighFrequencyCustomers, who have move than 6 transactions in the previous 12 months, and then all the rest).
TransactionsLast12Months = calculate(distinctcount(SalesRawData[TransactionCode]),filter(SalesRawData,SalesRawData[customerCode]<=max(SalesRawData[customerCode])),filter(all('Calendar'[Date]),'Calendar'[Date]>max('Calendar'[Date])-365 && 'Calendar'[Date]<=max('Calendar'[Date])))
IsHighFrequencyCustomer = [TransactionsLast12Months]>6
So far so good.
But how do I use the IsHighFrequencyCustomer as a filter or axis to be able to show the change over time in the mix between these two customer groups?
Would greatly appreciate guidance.
srl01 wrote:
I'm working with a sales transaction dataset with unique customer IDs for each customer, and want to graph customer segments overtime (defined various ways, but to start off I want to split into 2 segments - HighFrequencyCustomers, who have move than 6 transactions in the previous 12 months, and then all the rest).
TransactionsLast12Months = calculate(distinctcount(SalesRawData[TransactionCode]),filter(SalesRawData,SalesRawData[customerCode]<=max(SalesRawData[customerCode])),filter(all('Calendar'[Date]),'Calendar'[Date]>max('Calendar'[Date])-365 && 'Calendar'[Date]<=max('Calendar'[Date]))) IsHighFrequencyCustomer = [TransactionsLast12Months]>6
So far so good.
But how do I use the IsHighFrequencyCustomer as a filter or axis to be able to show the change over time in the mix between these two customer groups?
Would greatly appreciate guidance.
It is not possible to put a measure to axis. To show the change over time, you can put calendar dates as X-axis and create two measures in a linechart.
HighFrequencyCustomer = calculate( [some expression], FILTER('table',IsHighFrequencyCustomer)) NonHighFrequencyCustomer = calculate( [some expression], FILTER('table',NOT(IsHighFrequencyCustomer)))
1 Reply
srl01 wrote:
I'm working with a sales transaction dataset with unique customer IDs for each customer, and want to graph customer segments overtime (defined various ways, but to start off I want to split into 2 segments - HighFrequencyCustomers, who have move than 6 transactions in the previous 12 months, and then all the rest).
TransactionsLast12Months = calculate(distinctcount(SalesRawData[TransactionCode]),filter(SalesRawData,SalesRawData[customerCode]<=max(SalesRawData[customerCode])),filter(all('Calendar'[Date]),'Calendar'[Date]>max('Calendar'[Date])-365 && 'Calendar'[Date]<=max('Calendar'[Date]))) IsHighFrequencyCustomer = [TransactionsLast12Months]>6
So far so good.
But how do I use the IsHighFrequencyCustomer as a filter or axis to be able to show the change over time in the mix between these two customer groups?
Would greatly appreciate guidance.
It is not possible to put a measure to axis. To show the change over time, you can put calendar dates as X-axis and create two measures in a linechart.
HighFrequencyCustomer = calculate( [some expression], FILTER('table',IsHighFrequencyCustomer)) NonHighFrequencyCustomer = calculate( [some expression], FILTER('table',NOT(IsHighFrequencyCustomer)))