Alteryx Designer Desktop Ideas

I will sometimes open "New Window(s)" from a single or multiple Browse(s), and arrange them on a second monitor to dig through a module I am unfamiliar with, or when building out complicated models.  It would be helpful if I could label these tempary new windows so that I don't spend to much time trying to remember which one is which.

User Story:  As a data analyst, I want to be able to easily edit my data's meta-data, so that down-stream actions can easily consume and export my data. 

Currently, you can do a bulk edit of your data's metadata, however, this has to be done using the Multi-Field formula. However, this solution is non-intuitive, and seems to run counter to the solution design of Alteryx, where each and every step is it's own module. 

A single module handling bulk and single column/attribute metadata transformations and editing would be (a) much more intuitive and (b) save analysts time of hand editing meta-data. 

cheers! 

Currently the only encoding method supported by Alteryx Avro format is deflate.  Please consider improving this to include google snappy approach.

This codec method is referenced in the Avro documentation for supported codex.

https://avro.apache.org/docs/1.7.6/spec.html#Required+Codecs

it would  particularly interesting to develop a WMS support in Alteryx.
To include other Maps like bing, google, HERE instead of CloudMade to display geo informations.

Mathias

I would love the R tool editor to work like a standard text field....it might be better explained in this scenario.  Pretend the character text is a script youve written with the function being at the top.  Let's say you'd like to move the function closer to the script, look at the weird output.  This editor pastes text like we are pasting images.

The use case is that I like to break my code into mini functions that I work on in the r console with sample data.  Once it works, I post it into alteryx and experiment with it on a small sample, then a larger sample.  If I have to have a document for my overall r cost in notepad ++, my function, and the console, it’s a little nusance, especially since I usually have to go back and forth with multiple functions.  I am not askin for a full blown editor, I like my notepad ++ for that, just a text input that works conventionally. 

SampleFunction <- function(x)
{
print(x)
}

abcdefghijklmnopqrstuvwxyz abcdefghijklmnopqrstuvwxyz
abcdefghijklmnopqrstuvwxyz abcdefghijklmnopqrstuvwxyz
abcdefghijklmnopqrstuvwxyz abcdefghijklmnopqrstuvwxyz
abcdefghijklmnopqrstuvwxyz abcdefghijklmnopqrstuvwxyz
abcdefghijklmnopqrstuvwxyz abcdefghijklmnopqrstuvwxyz
abcdefghijklmnopqrstuvwxyz abcdefghijklmnopqrstuvwxyz
abcdefghijklmnopqrstuvwxyz abcdefghijklmnopqrstuvwxyz
abcdefghijklmnopqrstuvwxyz abcdefghijklmnopqrstuvwxyz
abcdefghijklmnopqrstuvwxyz abcdefghijklmnopqrstuvwxyz
abcdefghijklmnopqrstuvwxyz abcdefghijklmnopqrstuvwxyz
abcdefghijklmnopqrstuvwxyz abcdefghijklmnopqrstuvwxyz
abcdefghijklmnopqrstuvwxyz abcdefghijklmnopqrstuvwxyz

when you paste the function (or any other text in the middle, look at this funky output)

abcdefghijklmnopqrstuvwxyz abcdefghijklmnopqrstuvwxyz
abcdefghijklmnopqrstuvwxyz abcdefghijklmnopqrstuvwxyz
SampleFunction <- function(x)abcdefghijklmnopqrstuvwxyz abcdefghijklmnopqrstuvwxyz
{abcdefghijklmnopqrstuvwxyz abcdefghijklmnopqrstuvwxyz
print(x)abcdefghijklmnopqrstuvwxyz abcdefghijklmnopqrstuvwxyz
}abcdefghijklmnopqrstuvwxyz abcdefghijklmnopqrstuvwxyz
abcdefghijklmnopqrstuvwxyz abcdefghijklmnopqrstuvwxyz
abcdefghijklmnopqrstuvwxyz abcdefghijklmnopqrstuvwxyz
abcdefghijklmnopqrstuvwxyz abcdefghijklmnopqrstuvwxyz
abcdefghijklmnopqrstuvwxyz abcdefghijklmnopqrstuvwxyz
abcdefghijklmnopqrstuvwxyz abcdefghijklmnopqrstuvwxyz
abcdefghijklmnopqrstuvwxyz abcdefghijklmnopqrstuvwxyz
abcdefghijklmnopqrstuvwxyz abcdefghijklmnopqrstuvwxyz

Although we can write a temporary file with all the parameters used it would be very helpful to pass simple parameters (or go all the way) between the Apps. I imagine any parameter created using Interface Tools should be available throughout the whole Chained App. 

The field summary tool is an excellent resource to get an overview of the data and spot targets for analysis or data cleansing. 

Unfortunately it has limitations either in the number of fields included or some combination of the number of fields and one or more of its attributes.   There is nothing in the documentation I found to make a user aware of this. When you exceed N  fields selected the system just hangs, indicating it is running but there is no connection progress shown and nothing seems to happen, even if you limit input to 1 record.

Through trial and error I found an approximate limit in number of fields I can include and still have it work.

I request that Alteryx update the tool help info and devise enhancements to dynamically load balance the tool so it can scale to the number of fields requested or at least warn when the limit is reached or approached.  The latter warning could be similar to the red font warnings in the formula tool when you have a malformed expression.  However a load balancing version is most desired.

The issue as it stands results in users wasting a lot of time trying to make the tool work as expected, then report it to support as an apparent bug in the tool which can be argued both ways.

I realize in the real world there are limits but in this real world we are seeing the number of fields to analyze increase significantly, especially when you have a data license and integrate 3rd party data to you own native data adding a hundred or hundreds more fields.

How about adding the ability to split intersecting trade areas at the points of intersection to create two new spatial objects.  The two objects could then be used to process customer records and divide them into "territories" based on  the line of intersection.

I would like to suggest to add a widget which encapsulate an R script able to perform outlier detection, something similar like netflix did:

• RAD - Outlier Detection on Big Data
• Tracking down the Villains: Outlier Detection at Netflix

Thank you.

Regards,

Cristian

It would be nice to tie the labels to the spatial objects being labeled.

The community could benefit from easier integration of splitting and applying functions to grouped data. The summarize tool is great for splitting your data and applying summary statistical functions. It would be super useful to take that block just one step further, and allow users to apply any other (aggregate) function to their grouped data instead of just the built-in functions in the summarize tool. I would envision that aggregate function either being a custom function that is a combination of existing user-specified functions within Alteryx (e.g. in the formula tool) and/or even an interface that allows you to use other Alteryx macros on the grouped data.

Apply user-defined functions, or other powerful Alteryx macros to grouped and data is a very common operation in the data analyst's daily workflows and being able to apply them without reverting to batch/iterative macros in a seamless manner would be naturally helpful.

https://ramnathv.github.io/pycon2014-r/explore/sac.html

Currently using these macros from Adam Riley: http://www.chaosreignswithin.com/

The stuff he has over there is great and makes alteryx much more powerful.  Two tools in particular that I am using from there are Conditional Runners and the Parallel Block Until Done.  These are tools that for us are vital and it would be nice if Alteryx would pick them up and maintain them for stability.  

It would be really useful to have a Join function that updated an existing file (not a database, but a flat or yxdb file).

The rough SAS equivalent are the UPDATE and MODIFY Statements

http://support.sas.com/documentation/cdl/en/basess/58133/HTML/default/viewer.htm#a001329151.htm

The goal would be to have a join function that would allow you to update a master dataset's missing variables from a transaction database and, optionally, to overwrite values on the master data set with current ones, without duplicating records, based on a common key.

The use case is you have an original file, new information comes in and you want to fill in the data that was originally missing without overwriting the original data (if there is data on the transaction file for that variable). In this case only missing data is changed.

Or as a separate use case, you had original data which has now been updated and you do want to overwrite the original data. In this case any variable with new values is updated, and variables without new values is left unchanged.

Why this is needed: if you don't have a Oracle type database, it is difficult to do this task inside of Alteryx and information changes over time (customers buy new products, customers update profiles, you have a file that is missing some data, and want to merge with a file that has better data for missing values, but worse data for exisitng values (it is from a different time period (e.g. older)).  In theory you could do this with "IF isnull() Then replace" statements, but you'd have to build them for each variable and have a long data flow to capture the correct updates. Now is is much faster to do it in SAS and import the updated file back into Alteryx.

I would love to be able to double-click on an input or output file and have the file open. Second to that would be a clickable hyperlink to the filepath that could be used to open the file, or a "go to" button or something. Anything would be better than my current process of copying and pasting the filepath into an explorer window.

In the summary tool, we often use the summary tool to concatenate strings - we love this functionality.

However, we would also like to be able to concatenate just the unique values of strings. This could be done if we ran the preceding text field through the unique tool first, and then concatenate. But when we are doing this for multiple text variables and when we need to summarize other types of data at the same time, this becomes a very un-natural combination of joins & macros. 

Thanks for considering,

Jeremy

XGboost regression is now the benchmark for every Kaggle competition and seems to consistently outperform random forest, spline regression, and all of the more basic models. For those of us using predictive modeling on a regular basis in our actual work, this tool would allow for a quick improvement in our model accuracy. And I think, from a marketing standpoint, having a core group of users competing in Kaggle using Alteryx would be a great way to show off Alteryx's power.

It is readily available as an R package: https://cran.r-project.org/web/packages/xgboost/index.html

I think we would all benefit from having IntelliSense within Alteryx. Just think about how much quicker writing formulas wold be or any time you need to reference your data within the various tools.

When editing larger workflows, the "Find Tool" window is often a great help in terms of locating tools. However, having to close this window in order to make edits to the workflow is a bit inconvenient, especially when changing the sort method or using the search function. It would be great if users were allowed to edit the workflows and tools while still keeping this window open and maintaining the current settings.

It would be extremely useful if most Alteryx tools had the option to output error records seaparetly. This functionality is already present in most other ETL packages, even freeware ones like Pentaho - Kettle. From its wiki: http://wiki.pentaho.com/display/EAI/.09+Transformation+Steps

http://wiki.pentaho.com/display/EAI/.09+Transformation+Steps

Step error handling allows you to configure a step so that instead of halting a transformation when an error occurs, the rows that caused an error are passed to a different step.