WizardOz Contributor

Hello, Valiant, I used this to get to a new page, it works perfect. However, in my new page, I have some cards built from another dataset. The second dataset has the same column name as the filter, but the cards did not pick up the fitler. Do you know how to let the fitler also works for the second dataset? Thank you.…

Hi, Superman, Wonder if you can give me some ideas: I am using this same approach to get to a new page, however, in the new page, I have a few cards built from another data set. The filter did not pass to these cards. In the second dataset, I used a beast mode to set up a column matching with the filter column name. Also,…

Hi, Superman, GuitarHero, Thank you for your ideas. Like SuperMan said, they both have limitations: - I do need to allow users to select date range, country, so using the flow to get the count won't work. - Sorting the count descendly, the problem is I do not know the top #, it changes all the time. I put an exchange ideas…

I know 50 M is not "Big Data" at all, it does not make sense to me too. We've been talking with Domo for a while, but they said not much can be done. I use Redshift, I looked at histry of running: The excution time (meaning left join, some calculations) is about 4 hrs, but it takes 5 hours to load data to AWS and 7 hours…

Hi, Superman, Thanks for suggesting to get the daily cost, but a monthly is good enough for us. As for how I want to use the data: 1. provide the spend for last month, 2. Compare Mom andYoY for spend. 3. Provide spend /visit for last month 4. Compare MoM and YoY for spend/visit. Currently, I used EndDate as the date to sum…

Hi, Superman, Since this is a drill down card to show who subscribed to our mailing list, I realy do not want to show anything else. If I add another aggregated fiedd in a table chart, can I somehow hide that column from viewing? Thank you. Olivia

I added an indicator in the dataflow. Every code when it is the first time presneted in the data, the indicator is 1, then I sumed up based on this indicator. sum( case when `code_indicator` = 1 then `mc_spend` else 0 end )

sum (distinct 'cost' ) could not give me the correct sum either. My dataset has 70m rows and 200 columns, and it is updated daily, I have many other calcualtions (combinations) I would like to do. I hope I can do it in beast mode instead of in dataflow.

As you can see sum in the table has code 1 sumed up as 20, so I will get total of 43. But the actual sum is 33, in other words, I only want to sum for each code once. in adition, after I get the sum, I will do some other calculations such as cost per visit, cost per download etc.

This is what I found out -- the beta of page embed will not start beta until late this year.

Thank you. Valiant. I will try Date Filter Card. Olivia

Hi, Superman, For this particular case, I only have 2 datasets, and one has 3 fields and the other has 4 fields for filters. Unfortunately, not all fields are same, therefore, the option to "apply it to all" does not work. But continue on this "Apply to all" approach: for the 2 datasets, I can change one of the field to…

Hi, Aaron, Thank you for your great suggestion!. I am going to see how to optimize my data flow like that Olivia

Thanks for let me know about FUSION ETL. I am tryng to get that one. Olivia

Hi, Aron, Thank you for your comments. The data with 123 columns and 54 million rows is the original source data. The example table A I gave above is only a subset of this original data - I have another query to select those columns. I have a data flow based on the originally dataset and ran 10 hours in redshift. I already…

Hi, dthierjung, your suggestion works - it takes about 3.5 hours to get it finish. But I think this is still too long. Wondering how do you guys run ad hoc queries? Thnak you

thank you - dthierjung, I will try what you suggested. May I ask what is your average run time for what size of data? But here is an example of a data flow I did yesterday and it is still running after 17 hours - I missed my deadline for an analysis :(. this is what in this data flow: 3 input tables: table A: 54million…

Hi, DDalt, Thank you for your help. I did as what's in your data flow. However, there is another issue now: In the data flow, the total is for all 'Sent' for all data I have, however, in my card, I only wanted to show for a certain time period, even better, I would like to give user the ability to select the date range,…

Hello, DDalt, thank you for your help. My data is stored in option 1 in your pdf. Unfortunately, what I am showing is what my stakeholders would like to have. I hope I can do it in the way they want. Any other ideas? Thank you. Olivia

Hi, Valiant, thank you very much for the advide. I have not used Designed studio yet, this will be a good opportunity to learn it. Also, good to know the new Storytelling/Layout functionality, hope it will be released soon. Best, Olivia

Hi, Superman, do you mean to create a dashboard app? thank you for the idea. Best, Olivia

Hi, Valiant, Thank you for sharing the ideas, I heard about the embed, but never tried before. Will get more info about it. Thanks, Olivia

Hello, Valiant, I did the % calculation as you suggested and got the %. However, for my table card, if I only want to see top 10 rows and % calcualtion is based on the total of the top10. Is there a way to do this? Thank you. Olivia

Thanks, Matt. But nothing come out.

Hi, Matt, I rewrote the statment like this, but still nothingcme out :(. Can you help? Olivia (case when `date`> (CURDATE()-7) then SUM(case when `Category` IN ('Views','Likes', 'Dislikes', 'Comments', 'Shares', 'Subscribers') then `Total` END) when `date`> (CURDATE()-7) then AVG(case when `Category` IN ('Avg Watch Time…

Hi, Matt, This is what I wrote , but I did not get any output. Can you please help me -- What is missing here? Thank you very much. Olivia CASE WHEN (SUM(CASE WHEN `date`> (CURDATE()-7) and `Category`<>'Avg Watch Time (Sec)' and `Category`<>'Avg Watch Percentage (%)' THEN `Total` END)) = 0 THEN 0 ELSE(AVG(CASE WHEN `date`>…

Hello, Smurray, thank you for your help. I've looked at this report, but it only gives me the title for each individual videos, what I need is the playlist titles so that I can add a filter for the playlist. For example, our youtube channels has playlists eventA, eventB, eventC etc, under each playlist, there are several…

NO, I am still hoping someone can help me with it. Thanks.