First of all I'd like to thank you for sharing this, it's absolutely amazing!
I'm just having a little problem that I can't figure out how to solve on my own.
For some reason, sometimes my lines cross each other for no apparent reason, did this ever happen to you, and would you know how to solve it?
Here is a printscreen to illustrate better the situation:
Rodrigo Kusminsky Herscu.
2 of 2 people found this helpful
The issue you are describing is usually due to table calc misconfiguration. « Ln Position » and « Ln Index » should all be computed at « level n » whereas « Ln Position Max » should be computed at the deepest. This applies for all starting and ending points of each curve.
I hope this will help you. Let me know !
Bien cordialement, Best Regards, Mit freundlichen Grüssen,
Olivier CATHERIN | CEO & Co-founder
Thank you for sharing, it's so cool!
Now I am trying to learn how to make.
But I can not link with the order and Polygonic model,
But also T can not be displayed.
Is I lost what steps it?
1 of 1 people found this helpful
Hi, just took me a bit to figure it out. We need to fix the Excel file in order to be able to look correctly link the twbx to the Excel:
- The 'Link' column has a capital in 'orders' whereas it does not in the poly spreadsheet, change all to 'Link' in order to fix it.
- Also, the 'T' column is spelled without a capital in the spreadsheet whereas Tableau is looking for a capital 'T';
@Oliver, this is amazing. I'm fiddling around with it, got tons of use-cases!
One question, would it be possible to order the gantt charts (and, thus, the curves) by something other than the total percentage?
In other words, would I be able to move the 'starting point' (on the right) all the way to the bottom in order to make the graph more intuitive?
1 of 1 people found this helpful
You can certainly sort the dimension the way you want it. Just be careful to apply your sorting options to both curves and gantt chart. Just right click on the dimension pill and click "sort". If there is a specific order to apply, use "sort manually".
This should do the work ;-)
hi oliver, im very new to tableau, can i do sankey diagram with blended data sources?
I want to manipulate the underlying data for some more calculations. I am stuck on getting the actual counts since the data is duplicated. If I wanted to write a Fixed LOD to get the actual customer counts without duplication, how would I go about it?
Wow you are Great!.
Thank for the sharing this nice information.
Very awesome VIZ! I was able to recreate your workbook using your data, but when I tried to make my own with my data the curves failed to draw. The Sigmoid values are also not as varied as they are in your example. Much confusion. Do you happen to have any insight on what I might be doing wrong? I'd be most grateful!
Thanks for this, and especially for the Sankey version.
I've managed to recreate the polygonic decision tree and it's really useful. However, I'm having a bit of difficulty replicating the Sankey version (even with your data set). Is it possible to have some pointers (or even a simiar guide to the above) on how to create it?
I can provide workbooks and a detailed explanation if required, but a guide would probably benefit the community more!
This is fantastic! I've been trying to figure out this exact viz type for a while now and something would always end up slightly off every time. Your walk through is much appreciated!
This is simply fantabulous! Am in no mood of waiting for parts of this to be implemented at my work place for an Investment banking client's IAM application!
Many thanks, and look forward for more such learnings!
This is the best resource I've found on these charts in Tableau so far! I'm trying to recreate something similar, but I don't need the curve sizes to be based on any measures, I just need them to act as connecting lines, as my main focus is to outline hierarchical relationships.
So the members inside my "Source" dimension would be the starting point (could be an event or a person) and then the next level would be the members of my "Source Description" dimension (name of event or person) and that would branch out to the names of the people these sources brought in to the business.
We'd like to design the dataset in such a way that new 'levels' are automatically generated inside the database if those new customers happen to refer / bring in more customers, and so on. So ideally, Tableau would need to be able to accommodate those extra levels if / when they come into play.
I've tried creating a sample dataset template to use in conjunction with the article resources here but I cannot seem to get it to work; as it seems an integral part of the article's chart involves linking curve size and position to each other; and mine will not need to be that dynamic. And as silly as it sounds, I can't figure out how to eliminate that factor in this example and still get it to work.
Is there a template somewhere I could use to achieve what I'm looking for?
Thank you! Olivier! This chart is amazingly inspiring!
Just learning this technique, trying to replicate the steps and everything looks accurate however I have an issue (image below) . I have narrowed it down to the right side points all meetings close together. In the tooltip the measures values for curve a, b and c are all really close can anyone point me in the direction to troubleshoot this issue?