Jsun82 Member

Ok, that was weird. I didn't make any changes to the data, but just reran it and even though the error was showing before (I had A, B, and C and then changed C to D and it showed all 4. Now it only shows the corrected A, B, and D). I also can't find other examples where this issue existed so I'm going to mark your answer…

Hi, thanks for the reply. The datasets in question are using the replace method so I don't think the recursive ETL is necessary. I already use that for certain reports that I want to append. I'm the only one building anything in Domo at all so it's more of an annoying cleanup issue for myself more than anything.

If you use the formula, it should be populating the new beast mode field with a value. So, if the new column is called "Status", then for the cases when `option` 'A' it stores a value of 'A' 'B' it stores a value of 'B' 'C' it stores a value of 'C' Anything else it should store 'Not Updated' And then when you graph it, it…

I think I understand your question, but let me restate: You want a count of options A, B, C, and Not Updated to show on a chart. If Option has a value A, B, or C, then it is considered updated. If so: case when `option` = 'A' then 'A' when `option` = 'B' then 'B' when `option` = 'C' then 'C' else 'Not Updated' end Then you…

Yes, it is now working! I thought I was going crazy for a second, but it seems like it's back to normal. Thanks for responding.

Looking over it, the logic behind this makes sense. Is it possible that the [mode] part of the function is causing the mismatch? I'm not super familiar with the WEEK function, but I didn't see '22' as an option.

I would use a simple ETL transform "Collapse Columns". Here is a link to how to do it: https://knowledge.domo.com/Prepare/Magic_Transforms/ETL_DataFlows/02ETL_Actions:_Edit_Columns I've attached some files that hopefully will help.

Have you tried sorting with 2 parameters? Try it with date (to keep it as you have it) and then by the value (1-5). You may have to change the aggregation to get it to look right.

Ah, I see. Currently you have "Name" as your category, so that is why you are seeing the letters at the bottom. Also, you are using the "grouped bar" graph instead of the "single bar". Can you share what the base excel file looks like? I just need to know the format to try and get the information looking like you want it.

As stated by Valiant, if you go into analyzer, the fields you select to place into "x-axis" and "y-axis" make up the visualization. There is a default that Domo selects, but you can easily change all of the inputs. If you are still having issues can you send a screenshot of your analyzer?

Are you seeing the incorrect totals within ETL? I would make sure that you aren't experiencing a rollup later on. Also, another option would to multiply the values with a beast mode formula on your output. It could keep things cleaner.

Sorry, when I was saying your rounding, I was referring to your ETL transform version and not the beast mode calculation. If we go back to the beast mode formula, have you checked all of your inputs that they are strictly numbers? I did some testing and the SEC_TO_TIME worked fine for all postive whole numbers and decimal…

Have you tried converting the inputs to your calculation to decimals? This may prevent the rounding you are seeing if they are currently coming in as whole numbers.

Here is the rank and window link: https://knowledge.domo.com?cid=etlactionsrankandwindow And the video: https://university.domo.com/videos/watch/ozZWNnYzE6i9lZiWn5F2dpeCKiyFoQGI/cat/1d40efe9078f416a95ade66fab31ee53 It looks like you have set it up correctly (hard to tell without seeing the base data). It should be…

Are you saying that the data you have has two different categories, or are you saying that you are seeing 2 different names for the same set of data? In either case, you could fix this in the base data, or you could do this via beastmode. If you only have 2 distinct data sources and just want to rename it, this is what I…

This definitely works. Also, it may be easier to use the ETL transform "Rank and Window" with the subfunction average for the window of 7 days depending on how your data is laid out. If there is only one data point per day, or if you can group by day without affecting what you want to see in the output, it may be a…

This looks like there may be more than just the text field being sorted since I think the logical progression would start with 120. I would take a look at your sorting filters first. Then, in general, if you want to sort out the order in which the items are displayed, you could take the same case statement you have below…

I'm not sure if this can be done in beast mode, but I know it can be done in mysql and in ETL transform. If it were me, I would just "group by" as a transform for the relevant field and then run the same formula that I posted above. Is there any reason you would still need to see the individual entries? That would be the…

You could fix this in an ETL transform by using the rank function to create a new column. Then you could use that column as a filter to get the top 10, top 20, or whatever version of this you would like. Hope this helps!

If there is only one instance for each value (i.e. each customer name appears only once) then you could use: Title the beast mode formula something like "Adjusted Group" case when `Value` < 200 then 'Other' else `Group` end `Value` = your values you are checking to see if they are smaller than 200 `Group` = the grouping…

I don't think there is a way to do this currently, and I agree this would be helpful. However, there is a work around that I've used that may help. If you save the current version and then notate the version then delete the branch you want to pause, you can do whatever testing is required and then revert to the earlier…

I'm seeing the same issue. The data is fine when not compared, but when compared month over month, the dates do not match (i.e. instead of Feb being compared to Feb, Feb is being compared to Jan)