Retail & Consumer Goods

Hi @Sushi almeda​ . When dealing with ticket data, it is best to re-shape your data into one row per event, with an event being an open or close action. You can use Custom SQL (UNION ALL) or Tableau's union feature to self-union two copies of your ticket data together.

I exported your Jira data into a CSV and joined it with itself:

Then I aliased the auto-generated Table Name field to distinguish the two identical copies:

Then I created a combined Date field that is either the Created date or the Resolution date.

// can't use aliases except in GUI
IF RIGHT([Table Name],1) != "1" THEN [Created]
ELSE [Resolutiondate]
END

Filter on Date to exclude Null and then it's easy to graph what you want.

You can also graph the number of open tickets at any point in the past using this technique. Create a new field called Delta that is either +1 or -1.

IF RIGHT([Table Name],1) != "1" THEN 1 ELSE -1 END

Now you can plot the Running Total of Delta right up to the current week.

I used to analyze a lot of Jira ticket data.

I have a similar problem, and it has to do with using the week function on the week begin date of last week of the year. For example in 2019/2020 transition week, the Gregorian calendar treats the week beginning 12/29/2019 as week 53, but the following week (beginning 1/5/2020) as week 2 because technically week 1 is hidden as it begins in the middle of the week (1/1/2020 - 1/4/2020). The ISO-WEEK calendar is intended to correct for it by only allowing 7 day weeks, but ISO weeks start on Mondays and most companies weeks begin on Sundays. I don’t have a solution but zero null is probably not the only reason that can happen.

Hi Morgan,

Use the tool below to export a Tableau friendly dataset to map your dates to the NRF Calendar. Join on the "date" value and you should be good to go.

https://public.tableau.com/app/profile/fearghal.gunning/viz/NRFCalendarExportTool/454Visualization

Grant

I have created parameter to select year, currently selected 2020, displayed week number and sales. Using LOD found 2019 last week sales. finally difference between them.

Added line chart based on weeknumber of current year.

Please find attached packaged workbook.

Hope it will help !

Hi @Sanober Khan​ 

You must have defined your filter on a particular field right, add that field in your statement.

I believe your need this statement for the header of the report, If yes then you should add the field to the detail shelf for it to be available. Also your filter should be single value list type to pass one value at a time.

Let me know if you still have any doubts.

Thanks and Regards,

Pallavi Naik

For performance, you have to think through the entire analytics stack:

-Storage

Network (Network transport can have issues within each layer)

-Database

Network

-Tableau Server

Network

-Dashboard

Network

Starting from a Tableau Dashboard perspective, @Bruce Segal​ comment above is the place to start, Performance Recording is your friend.

The links below will help from a Desktop/Dashboard performance perspective and what to think about.

Desktop Performance Recorder: https://help.tableau.com/current/pro/desktop/en-us/perf_record_create_desktop.htm

Tableau Workbook Performance Checklist: https://help.tableau.com/current/pro/desktop/en-us/perf_checklist.htm

If your dashboard is well designed, and it is slow, it can be the database. I would say 90% of the performance issues that I see are at the database level. Most are on-premise databases or companies that have moved from on-premise to online databases, but they didn't reshape the data to take advantage of newer database technology.

Not sure what your database is but I included a few whitepapers below for Snowflake and Google Big Query.

Database Performance:

BigQuery Best Practices: https://www.tableau.com/learn/whitepapers/google-bigquery-tableau-best-practices

Rapid Fire Analytics w/ Snowflake: https://www.tableau.com/learn/whitepapers/delivering-rapid-fire-analytics-snowflake-and-tableau

Tableau+Snowflake Blog: https://bigdatadave.com/blog-feed/

Hi @Sanober Khan​ 

I had already posted my solution to your earlier question about joining 2 tables (one

with Raw SQL timestamp and the other with Julian date).

As far as this question goes, you can convert the raw sql date to Julian date using this

formula that I came up with using Interpolation of my earlier results (Astronomy) over

a decade and half ago.

2451545 + DATEDIFF ('day', DATE("2000-1-1"), [RawSQL Date] )

Basically, in the above formula 2451545 is the Julian date corresponding to 1st Jan 2000

(which is the date value you see in the above equation). I also tested this with the value in

your example with a Julian date value of 2458878.

The logic is very simple as 2458878 - 2451545 = 7333 days which would be equivalent to

20 years and 28 days.

We had 5 leap years between 1st Jan 2000 and 1st Jan 2020 which adds 5 days

to 20 x 365 thus giving a total of 7305 days and subtracting this from the earlier value of

7333 gives 28 days. This was the difference so the actual date will be 1st Jan 2020 + 28 days

= 29th Jan 2020 (In case this is getting too confusing 😕 , forget this explanation and just use

the earlier formula which does exactly the same thing). Take care and Best wishes !

Sincerely,

Soumitra

Hi Janine,

Have you seen the following which may help you?

Custom Shapes | Tableau Software

How to use custom shapes as filters on your dashboard | Tableau Software 

• Ryan Sleeper's Odds of Going Pro in Sports
• Clara Siegel's Where Have Dinosaurs Been Found?
• Angel Pilares's Copa America
• Dash Davidson Best Fantasy Factories

https://tableaumagic.com/tableau-qt-custom-shapes/

Hope it all helps!  If it does, please mark as correct so that others may find it useful in the future.  Best, Don

Here is the link to my viz. I didn't have too much time to play with it, but wanted to submit it before I forgot!

Tableau Public

My insights:

• Los Pollos Hermanos is the supplier with the highest % shrink, followed by Wayne Enterprises, then Barry Allen Shipping.
• Eggs is the product with the highest % shrink, followed by flour, milk, then sugar
• Barry Allen Shipping is the supplier with the lowest average shipping length (makes sense with the Flash), followed by Los Pollos Hermanos, then Barry Allen Shipping
• Sugar is the product with the longest shipping length followed by eggs, milk, and flour
• Barry Allen Shipping has the largest average ship length for eggs, but the lowest % shrink
• Wayne Enterprises has the longest average ship length for sugar
• Los Pollos Hermanos has the largest % shrink for eggs
  • There is some correlation between ship time and % shrink with eggs. The quicker ship lengths tend to have the lowest percent shrinks
  • There are a couple outliers for shipments in the month of September. The average ship length is 9 days but the % shrink is extremely high for those shipping lengths. Something out of the ordinary happened with these (Did the chickens revolt? Were the "eggs" really in the shipment or was it made to look like they were? Did someone interfere and take from the shipment?)
  • Average shrink for eggs is 21% the shipments from Los Pollos during September have up to 3 times that shrink