jaeW_at_Onyx Coach

you CAN do a lag function in Analyzer! Youtube video explanation: https://youtu.be/cnc6gMKZ9R8 lag(sum(`Measure1`)) over( order by year(`Date`) * 100 + WEEKOFYEAR(`Date`) )

I just posted not to use a Recursive Dataflow. But if you're hell-bent. Don't use a JOIN to combine historical and new. Use an APPEND (b/c JOINS in MagicETL are slow especially with composite keys). Add a Row_Number() function via the RANK / WINDOW'ed tile where you number the rows based on how you define a duplicate row…

Sorry late to the game! But my ten cents. Don't build a recursive dataflow. Yes, a recursive dataflow will work for your use case, but recursive dataflows don't scale with data volume. If we're talking 100M+ row datasets, you'll start having to wait in excess of 20 minutes to i've seen 5, 10 and 23.999 hours. (23.999 b/c…

You can alter dataflows if you know how to monitor network traffic and send curl requests. OR you can update the JSON definition of a dataflow using the JavaCLI (my recommendation over CURL or Postman) and the list-dataflow command. This is definitely one of those ... use at your own risk kind of things. Creating a…

Clarifying question. Do you actually have future dated data? You could do something like date_format( `target_date`, '%Y-%m') to group your dates by yearMonth number and then take the count of each row, just to verify that you have future dated data. Assuming you do, you can follow @GrantSmith 's tip with a small twist If…

in the enumerator you only keep columns specific to engagement (in my case just libraryID b/c that's the PK for the table, userID, contentID and count_numerator.

Oh no!! Dang it. I'll have to record it again tonight. Sorry.

Gotcha covered @WizardOz , EDIT: updated with sound: https://youtu.be/Xb4QgKYgaqg the short answer, you want to create a dataset with the granularity one row per user per content. then JOIN it to the rows of actual engagement. that way you can calculate did a user engage or not and take the ratio. this model does not use…

can you make a new dojo topic for this? I LOVE THE QUESTION and am happy to dig in... but just to keep things separate. 

Building on Superman's response, From your screenshot it looks like you have a date that contains 'Offsite'. which suggests you have text in what you want to be a date column. when domo ingests data from excel it should autoconvert excel's integer representation of a date to ... date. If the connector encounters text, the…

Neeti, a blank row could just be a row with a space in it. Add it to the Measures table I described.

I'm not sure if i follow your requirement exactly but i think there's a couple things you might consider. I don't think Domo really shines in use cases where you try to push individual rows frequently, insofar as you're not sending data directly to a database... the data gets collected in a data lake-esque environment…

instead of relying on interactions, could you build a URL into card? you could use a CASE statement to control where the URL goes. And maybe use pfilters to apply filters to the card. It's no longer a drill down, but maybe it gests the job done.

EDIT:: https://youtu.be/YgevJkjeFqw video showing how to do it. To get it as a pivot table, I imagine you'd have to treat your 'headers' like you do hierearchies in Excel Pivot Tables. For example if you wanted to see a grouping for Customers by State, you have Customer and State value on each row, and then can use a Group…

TLDR version: the first option operates AFTER aggregation. the second option operates AT THE ROW LEVEL before aggregation. SUPER TLDR the first option is the one you probably want. Consider the pseudo SQL that gets generated. (SUM(`NewValue`) - SUM(`OldValue)) / SUM(`OldValue`) select (sum(newValue) - sum(oldValue)) /…

It's unclear whether you need help building a card or writing a beast mode. what didn't work about calculating the previous month? how is your data currently structured? can you get the correct results in a table card?

Yeah this is a tough one... In a nutshell any ETL tool (except Fusion) will have to transfer your data into the transformation engine, a SQL database or Magic's ... data processing environment, before you can transform it. Hence why your ETLs take so long. Your goal should be to use VIEWS which go directly to the…

EDIT:: video response (20 min) https://youtu.be/9uNv1_0XXao EDIT:: oops sorry ... just saw that this was a magicETL dataflow. read my SQL example and i'll post the Magic version below. OOOOH a fun brain teaser! so step 1) build a query that just extracts the last time i completed the 3 programs. we GROUP BY userID and…

Domo does not have MySQL v8, so you actually won't find support for windowed functions in MySQL dataflows :( You can do it using variables Basic examples https://stackoverflow.com/questions/2563918/create-a-cumulative-sum-column-in-mysql With Partitions.…

Testing for blanks inETL can be such a PITA! Maybe start with a Filter and try a couple of Filter permutations until you've locked down exactly what the data contains. You can try to use the length function to see if the blank is truly blank or space.

I can't answer if 'people are doing fewer surveys' but there are some established ways to measure the effectiveness of your survey. A/B testing is possibly one of the more better known and easily understood methods. https://www.optimizely.com/optimization-glossary/ab-testing/ Basically you have the same survey that you…

not to be ... that guy ... but have you opened a support ticket? given that this is a beta feature ... i have no idea what the response time on that should be. maybe check in with your CSM.

If you haven't worked with Windowed Functions before, I give a longer (15min) answer on YouTube for a related use case. https://www.youtube.com/watch?v=ZPf41Fjn1H8

No need to do ETL!! Create a Beast Mode with (swap the correct spellings for column names). You'll have to have the feature switch that allows you to use aggregate functions enabled (talk to your CSM). sum(sum(`Opens`)) over (Partition by month(`Date`) ORDER BY `Date`) / sum(sum(`Sent`)) over (partition by MONTH(`Date`)…

Sorry late for joining late in the game... Question, do you know for a fact that you have EndDates for all your Data? If you don't have an EndDate for all your data, then any type of JOIN should filter it out (hence I think why you ended up putting Date on the LEFT side. I think what I would do is create a EndDate_clean on…

If I understand your question, you're trying to create a parameterized queries. that doesn't really exist in Domo, so you can take a MySQL dataflow esque approach. Basically create a webform with the following columns and JOIN it to every row of your transaction dataset. (then change your parameter values using the webform…

I'm pretty sure you want the .appendChild method https://www.geeksforgeeks.org/html-dom-appendchild-method/ function geeks() { var node = document.createElement("LI"); var textnode = document.createTextNode("Web Technology"); node.appendChild(textnode); document.getElementById("gfg").appendChild(node); }