Visual Analytics- First.var Last.var

[gabras]

Hello,

I'm trying to simulate a Customer Journey in SAS VA.

I would like something like this:

ID	wayOfContact	Request	DOIT	reqBefStore	reqInStore	wocBefStore
1	Tel	A	0	-	-	-
1	Store	B	1	A	B	Tel
1	Tel	C	0	-	-	-
1	In	D	0	-	-	-
1	Store	B	1	D	B	In
2	Fax	D	0	-	-	-
2	Mail	C	0	-	-	-

The DOIT variable is calculated when the same ID contact the company by wayOfContact <> Store and then Store. DOIT is 1 if wayOfContact is Store, 0 otherwise. If an ID contact the company only going to Store wayOfContact must be 0.

And it is possible to hav also other column:
- object of the request before Store (reqBefStore)
- object of the request during Store (reqInStore)
- wayOfContact before Store (wocBefStore)

Please help me.

Thank you

[SASKiwi]

This is best done outside of VA using DATA step processing when preparing your VA load data. What have you tried so far?

[gabras]

Hi @SASKiwi,

Thanks for the reply but i need to do it in VA. I need somethin dinamic that change based on what the user choose as filter. Maybe the user wants to see that table only when wayOfContact in (Tel, Store) and so DOIT must change accordingly.
I hope i was able to explain what i meant for.

Thank you 😉

[SASKiwi]

Can you show using your sample data how your calculated columns would change if the filter was applied?

[gabras]

Hi,

if i wolud choose wayOfContact in (In , Store) the matrix would become:

ID	wayOfContact	Request	DOIT	reqBefStore	reqInStore	wocBefStore
1	Store	B	0	-	-	-
1	In	D	0	-	-	-
1	Store	B	1	D	B	In

Any suggestion?

[gabras]

I don't know if i'm wrong but i think that theese requirements could be reached using a Sankey Diagram.

But i have nevere used it.

[gabras]

@FalkoSchulz would you help me?

I would add that i can add a date variable to my table.

But i have questions:

- if my data has no time values, and since it could be duplicated data value, is it possible to create a DOIT variable according to the sort the table showed?

- is it possible, using sankey diagram options,filter event (request) values and show the diagram based on given filter? Look at the table below:

Table i have:

ID	wayOfContact	Request	DOIT	reqBefStore	reqInStore	wocBefStore
1	Tel	A	0	-	-	-
1	Store	B	1	A	B	Tel
1	Tel	C	0	-	-	-
1	In	D	0	-	-	-
1	Store	B	1	D	B	In
2	Fax	D	0	-	-	-
2	Mail	C	0	-	-	-
3	Fax	B	0	-	-	-
3	In	D	0	-	-	-
3	Store	B	1	D	B	In
4	In	E	0	-	-	-

Table after filtering (only "In" and "Store"):

ID	wayOfContact	Request	DOIT	reqBefStore	reqInStore	wocBefStore
1	In	D	0	-	-	-
1	Store	B	1	D	B	In
3	In	D	0	-	-	-
3	Store	B	1	D	B	In
4	In	E	0	-	-	-

Thank you

[gabras]

@FalkoSchulz i have this dataset:

ID	Date	wayOfContact	Request
1	01/01/2018	Tel	A
1	02/01/2018	Store	B
1	03/01/2018	Tel	C
1	04/01/2018	In	D
1	05/01/2018	Store	B
2	05/01/2018	Fax	D
2	10/01/2018	Mail	C
3	01/01/2018	Fax	B
3	02/01/2018	In	D
3	03/01/2018	Store	B
4	02/01/2018	In	E
5	04/01/2018	Store	A

and i would like something like this:

ID	Date	wayOfContact	Request	DOIT	reqBefStore	reqInStore	wocBefStore	nDaysPrevRow
1	01/01/2018	Tel	A	0	-	-	-
1	02/01/2018	Store	B	1	A	B	Tel	10
1	03/01/2018	Tel	C	0	-	-	-
1	04/01/2018	In	D	0	-	-	-
1	05/01/2018	Store	B	1	D	B	In	3
2	05/01/2018	Fax	D	0	-	-	-
2	10/01/2018	Mail	C	0	-	-	-
3	01/01/2018	Fax	B	0	-	-	-
3	02/01/2018	In	D	0	-	-	-
3	03/01/2018	Store	B	1	D	B	In	2
4	02/01/2018	In	E	0	-	-	-
5	04/01/2018	Store	A	0	-	-	-

If wayOfContact in ('Store','In') filter is applied could become:

ID	Date	wayOfContact	Request	DOIT	reqBefStore	reqInStore	wocBefStore	nDaysPrevRow
1	02/01/2018	Store	B	0	-	-	-
1	04/01/2018	In	D	0	-	-	-
1	05/01/2018	Store	B	1	D	B	In	3
3	02/01/2018	In	D	0	-	-	-
3	03/01/2018	Store	B	1	D	B	In	2
4	02/01/2018	In	E	0	-	-	-
5	04/01/2018	Store	A	0	-	-	-

Is it possible to have this in sas va?

is it possible to have it using sankey diagram ?

[SASKiwi]

If you prefer a Sankey diagram then how you are trying to restructure your data isn't required, I suspect. What version of VA are you using? Up to Version 7.4 Sankey diagrams are only available as visualisations which you can export to a VA report:

http://support.sas.com/documentation/cdl/en/vaug/69957/HTML/default/viewer.htm#n08zz749uilz3cn1u3hm8...

I'll leave it up to those who know more about these diagrams to comment further

[gabras]

Hi @SASKiwi
I think you are right.
But the re are some requirements that i don’t know i can do using sankey diagram. For example, is it possibile to add a label to the node? If i use the way of contact variabile for the node is it possibile to show also the request associated to it?

[FalkoSchulz]

Unfortunately there are some limitations in terms how row-based calculations can be done in VA. You have a related discussion here: Re: SAS Visual Analytics create a Sequence variable

You can certainly use your original table straight as input to path analysis (sankey diagram) - it would render similar to this:

Note, that a sankey diagram is really designed to be used on high volume data so you rarely look at an individual but rather at the overall customer journey here. For example, what percentage of my customers first contact by phone and then visit a store? How many customers do not come back (contact) after visiting a store? How many purchase a product? etc.

The sankey diagram shows you the total count (small numbers to the start and the end of a path) taking this particular step.

No, the labeling in a sankey diagram is done given the event identifier. in your example this is probably 'wayOfContact'. You can however use special path filters here to identify a particular path(s) of interest. For more detailed investigation - you can create other visualizations given a path or event selection (e.g. bar chart showing the customers behind this path). Such visualization can have additional labels, measures etc to describe this path subset.

Hope this helps. Falko