MarkSnodgrass Coach

You should be able to do a recursive dataflow on this pretty easily and store the datetime for the task at different flags. When you join the new data with the recursive data, join it on task ID AND flag. This will allow you store the task datetime when the flag is 0 and then, in a new row, the task datetime when the flag…

@grandele glad it worked for you. If you can mark my answer as the accepted solution, it will help others in the community. Have a great day!

You will want to use a case statement and the datediff function. It would look like this: CASE WHEN DATEDIFF(EmailDate,FirstDate) <= 8 THEN 'True' ELSE 'False' END This should do it for you.

In the General Chart Properties, for the Divide Value by property, set it to None.

One way to accomplish this would be to use the rank and window tile in Magic ETL and get your totals and rank them in order. You can then use the rank to show your top 10 and then use a beast mode (or formula tile in Magic ETL) to assign anyone above 10 to the other group. Use this new field as your series. You can also…

ok. you have things opposite of what I do. Below in the code block is my dataflow. Take what is in this code block and copy and paste it into a new empty dataflow. It might not like the input output datasets in your instance, but it should allow you to at least see the tiles that are important and click around. (Magic ETLs…

What criteria are you using for your filter tile? Based on what you are doing it should be where the date is null from your right table. Here's a screenshot of a basic recursive dataflow that I did, and yours appears to be set up similarly.

If your new dataset doesn't have any beast modes built off it, then there won't be any duplication because nothing is there to duplicate. If you do have a beast mode with the same name on both datasets, it will rename one by adding something like "from dataset...." to the end of the beast mode name. It will also alert you…

Unfortunately, I am not seeing an option to modify this. You could submit this as an enhancement request in the ideas exchange section of the Dojo and it might get included in a future release.

You are going to want to watch @jaeW_at_Onyx video on rolling averages https://www.youtube.com/watch?v=cnc6gMKZ9R8 You will need to ask your CSM to enable Windows Functions in Beast Modes if it is not already enabled in order to what he explains. Your tiers would just be a case statement built on top of the rolling average…

@ljb18 what field is being used to count the number of tracks? That field likely has blanks in it and the summary number will not count blanks as part of its total. Find a column that does not have any blanks or nulls and use that as your count for the summary number and you should see the same number as your dataflow row…

One issue you are running into is using the week function and your dates cross years. Since the week numbers start over at the beginning of each year, you are going to have issues with the math. You would need to incorporate if the year of both dates are the same or not and make the necessary adjustments. I had to do a…

I would recommend using the Domo CLI Tool, which you can download from the Admin - Tool Downloads section and perform these bulk actions.

@GrantSmith will know the answer to this one.

Domo has an API that you can work with to automate the process. You could be a custom api that extracts data from one of your systems and updates Domo.

That's hard to say without seeing some sample data, but rather than doing a full join, you should consider doing an append and include all columns from both datasets. It is a commonly recommended approach instead of joining the data.

A full join is going to give you all the records from both tables and join matching records where it can. An inner join would only give you the matching records and exclude all other rows. This KB article has some good visuals that explain the various types along with a video.

You would want to follow the steps outlined here:

Yes, you can split your data and rejoin the data in Magic ETL. You might also want to look make use of the select columns tile while you are in the ETL and remove columns that you don't think will be used, or branch columns off into separate datasets. Keep in mind, if you want to use a table card, that card type is limited…

Is that error referencing a column name, or the data in a particular column?

Domo's MySQL is version 5.6, which doesn't support window functions. I would suggest using Magic ETL to do this with the rank and window tile. If you have Magic ETL 2.0, it will run incredibly fast. Rank and Window KB Article: Magic ETL 2.0 KB Article:…

Here is an idea for you to try using Magic ETL, and if you are using 2.0, it will run very fast. Here's a sample visual of what I am proposing for you and will explain it further below. Look at your data and determine what the maximum number of columns are. In my example, I am assuming the largest row has 10 columns, so I…

The Domo Webform is a very quick way to enter in some data and test with it. You can find it when you go to the data center and click on cloud app and then enter in webform

If your beast mode is calculating correctly in the pivot table, you can use that same beast mode as the y axis in the bar graph and the sales date in the x-axis and then use the date range filter in the top right to tell it to graph by month.

@MattGo good point. I pulled that example from a flex table card, which is what I often to use to compare prior years and I don't have to select the compare to previous year date option. I agree that it would be nice if it was just a baked in option on the summary numbers, especially for period over period cards.…

You can create a custom summary number to get this information. Here is what I wrote for one of my cards to show the trend. ROUND(((SUM(CASE WHEN YEAR(`datereported`) = YEAR(CURRENT_DATE()) THEN 1 ELSE 0 END ) - SUM(CASE WHEN YEAR(`datereported`) = YEAR(CURRENT_DATE())-1 THEN 1 ELSE 0 END ))/ SUM(CASE WHEN…

Have you tried writing it like this: SELECT p.`id`, p.`name`, p.parent_id, c.`name` as childname, FROM `products` p INNER JOIN `products` c on p.id = c.parentid You could change the inner join to a left join in case there are products with no children and you want to still include those.

I'm assuming you are doing this at the card level. You can do this a couple different ways. If you don't need the blank records, drag the ID field into the filter section and tell it to exclude blank records. Then, your count distinct should work as is. If you need the blanks but don't want to count them you can write a…

If you know that your text is always going to start with "Email somedate" then you can use the substring function to extract the date. In this example, you would do SUBSTRING(textfield,7,10) to get the date. You could also wrap the STR_TO_DATE function around the substring function to format it in a particular way if you…

You need to join these two datasets together to make one dataset. The easiest way to do this is with Magic ETL. The below example shows you more than just joining, but the idea is that you can use the join tile to join your two datasets together. You would join on week. You might want to review this KB article to better…