Skip to main content
cancel
Showing results for 
Search instead for 
Did you mean: 

The ultimate Microsoft Fabric, Power BI, Azure AI & SQL learning event! Join us in Las Vegas from March 26-28, 2024. Use code MSCUST for a $100 discount. Register Now

Greg_Deckler

How to Get Your Question Answered Quickly

Introduction

This article contains numerous suggestions and recommendations that will help you get your particular question about Power BI, PowerAutomate and PowerApps answered quickly and correctly versus seemingly ignored. 

 

First off all, before anything, be sure to check the community Code of Conduct.

 

Urgent Issues

First and foremost, the community is a "best effort" kind of support forum. There are no true guarantees or service level agreements (SLA's) regarding when you might get an answer. You may not get an answer. Don't get me wrong, the community is an incrediblely robust and an amazing resource but if you are having an urgent issue that needs addressed immediately, you need to get actual Microsoft support involved. To do this, open a support ticket here:

 

Power BI - https://support.powerbi.com
PowerAutomate - https://flow.microsoft.com/support
Power Apps - https://support.powerapps.com

 

Do a Little Research

Before posting your question, use the Search functionality to search for an existing answer to your issue. These forums have been around for awhile and there is a good chance that somebody else has had your issue and that it was solved by the community. This is the fastest way to get your question answered because the answer already exists. I literally have macros defined for about 10 different questions that get asked again and again on the forums that provide a nice standard answer with links to the solution. So, yes, it happens quite a bit. Search is your friend. 

 

Post in the Correct Forum

This may seem elementary, but choose your forum wisely. Some top answerers regularly peruse the "All Topics" view of the forums, but not everyone. Some of the top answerers are more specialized. For example, I do not answer much in the Developer forums because I do not do a ton of custom, hard-core Development. But other answerers pretty much only do Development and that is the only forum they follow. Also, posting a question in the wrong forum will create confusion on the part of the answerer. For example, posting a question in the Power BI Service forum for a Power BI Desktop question will have the answerer thinking how the issue is solved in the Service, not the Desktop. 

 

Don't Cross Post

Cross posting is considered rude and wasteful and it has been so since the dawn of forums. Don't do it. Again, some answerers use the "All Topics" view of the forums and then we see 5 posts with the same title and question. Which one do we answer? Bottom line, some community members will not answer a cross posted topic on principle alone. If you are not sure where to post your question, use the "General" forum.

 

Be Nice

The vast majority of the community are volunteers with day jobs. Being rude or complaining about Power BI, PowerAutomate, PowerApps or Microsoft will not make answerers endeavor to help you. And when the answerer misunderstands your question and gives you the wrong answer, be nice and polite and inform them that they misunderstood and re-explain what you were going for. Being terse and such will, again, not encourage answerers to endeavor to help you. I have not seen it on this site, but I have known some members in other similar communities that kept lists of other members that they would just rather not deal with.

 

Post Sample Data

UPDATE: @ImkeF wrote a fantastic article for the best way to post data to the forums: https://community.powerbi.com/t5/Community-Blog/How-to-provide-sample-data-in-the-Power-BI-Forum/ba-...

 

I cannot stress this one enough. That's why it is in red. If you have a question about something that just isn't working right, post as much information as you can, starting with your data. Scrub it if you have to, but post it. It makes things SO much easier to understand and replicate the problem. Plus, answerers will be more inclined to answer your question if they can quickly copy the data and start figuring out the solution. Even better, the answerer can give you back an answer using your very own column names instead of "SomeTable" and "SomeColumn". This means that you can just copy and paste the answer (if it involves a formula) into your own model and not have to monkey with it. PowerApps, PowerAutomate and Power BI all deal with data at some level so make sure to post examples of the data you are working with.

 

Attachments

You know what is even better than posting sample data as text? Posting the actual file that you are working with! Obviously this cannot be done in all cases due to privacy issues but if you can recreate your issue with a sample file and attach the sample file to your post, this is a fantastic way to speed up the process of getting to a resolution.

 

When Posting Data, Put it in a Nice Format

The easier that you make it for answerers to copy your sample data directly from your post and paste it into Excel or an "Enter Data" query, the more likely and faster that you going to get responses. Think about it, if you paste your data in an unfriendly format or as a picture, the answerer is going to have to key in all of the data by hand to recreate your model. Thus, the odds of you getting an answer in a timely manner drastically decreases.

 

The most tried and true method that I have found is to copy and paste data from Excel into the post. You will get an error posting the first time about invalid HTML and such, but just hit the post button again and your data will be in a nice table within the post that can be simply copied and pasted into an "Enter Data" query or back into Excel. The next best thing is to format your data as a comma separated file (CSV). Again, very simple to copy and paste into Notepad and save as a CSV file that can then be easily referenced in a CSV query. Oh, and one last thing, include your headers!

 

Include Your Relationships

Here is where posting a picture is the perfect solution. Take a screen shot of the relationship pane and upload the image to your post. Failing that, use proper notation to explain your relationships, such as:

 

SomeTable          SomeOtherTable

SomeColumn 1:* SomeOther Column

 

SomeTable          SomeOtherTable

SomeColumn 1:1 SomeOtherColumn

 

The first denotes a one-to-many relationship, the other a one-to-one relationship. Use * for the many, use 1 for the 1.

 

Here is an excellent way to post your data model visually courtesy of @makdaddy :

image.png

 

Include Your Formulas

It is very helpful to explain what you are trying to do as this provides context to the problem, however, don't forget to post the formulas that you are having problems with. And, for bonus points, put them in a code block. That's the page icon with the "</>" in the middle of it next to the smiley icon when creating a post. This helps draw the answerer's eye to the formula and makes it easy to know exactly what to copy and paste into their own test model.

 

If you are posting code, run your code through a beautifier like:

 

Include Your Expected Result

If you are looking to create a particular visual, table or form that calculates data a particular way, by all means, include an example of the visual or the expected results of a calculation in your post. Even if you have to mock it up in Paint or something or simple text, something is better than nothing. This will help the answerer understand exactly the result that you are trying to achieve instead of going down the wrong path. It is even better if the expected results that you post are for the sample data that you post. This means that an answerer can take your data, create the solution and then cross check that back with what you are expecting. 

 

Don't Bother Posting Correlation ID Errors

correlation.png

 

This may seem a little harsh, but posting a picture like the one above and asking what is wrong is essentially a waste of everyone's time. Why? Because none of us speak Correlation ID or GUID either. There's simply no information upon which we can act to give you an answer. If you post one of these, the answer that will come back will likely be something along the lines of:

 

"I suggest that you contact support."

 

The exception to this advice is if you are asking a question about something that worked yesterday and now does not (for example) and the way to preface that image is something like:

 

"Hey, yesterday when I did <this>, everything was fine, but now every time I try to do <this>, I get this error. Is anyone else seeing this kind of error (see below) <insert screen shot>"

 

Or the classic:

"Hey I'm trying to do <this> and keep getting the error below, has anyone gotten this to work? <insert screen shot>"

 

Again, nobody can speak Correlation ID or GUID, but we CAN attempt to replicate the particular issue you are having and help confirm or deny whether or not it is a general issue or just you. 🙂

 

The people that CAN speak Correlation ID are the fine folks at Microsoft since they have access to the logs. You can go to https://support.powerbi.com, https://support.powerapps.com or https://flow.microsoft.com/support and open a support ticket.

 

SQL Code

Please don't just post your SQL Code and say "How do I do this in ..." OK, think about this, this is like me pasting in some assembly code and asking you "How do you do this in C#?" If you don't read assembly code, it will be difficult to answer the question. Just because SQL and DAX are both querying languages does not mean that they work the same or that people that understand one understand the other. If you must post SQL code, include a plain language explanation of what you are trying to accomplish. Because, this doesn't help...

 

How do I do this in C#?

 

 

 

 

 

segment .data
   count dw  0
   value db  15
segment .text
   inc [count]
   dec [value]
   mov ebx, count
   inc word [ebx]
   mov esi, value
   dec byte [esi]

 

 

 

 

 

 

Issues

If you have found a true bug or issue, you can make a post to the forum to let others know (and that would be very nice of you to do so). But if you want to get it fixed, you should post it in the Issues/Ideas (it varies) area of the community. Specifically, you should post it here:

 

Power BI - https://community.powerbi.com/t5/Issues/idb-p/Issues

PowerAutomate - https://powerusers.microsoft.com/t5/I-Found-A-Bug/bd-p/BugFound

PowerApps - https://ideas.powerapps.com

 

Ideas

If you have a new idea for functionality or how something should work, you can certainly start a thread on the community to solicit feedback on your idea, but you should also post the idea in the ideas section of the website and solicit others to vote on your idea. The development teams use the Ideas area of the communities to build their development pipeline and a significant amount of weight is placed on Ideas that have lots of votes versus those that do not. The ideas areas of the communities are here:

 

Power BI - https://ideas.powerbi.com

PowerAutomate - https://powerusers.microsoft.com/t5/Flow-Ideas/idb-p/FlowIdeas

PowerApps - https://powerusers.microsoft.com/t5/PowerApps-Ideas/idb-p/PowerAppsIdeas

 

Be Concise

Brevity is a wonderful thing in forum postings. The best questions are those that provide a brief overview of what the poster is trying to achieve and then post their data, relationships, formulas and expected output. If you post in this way, if there is an answer to be had, you will get an answer.

 

Example

Below is an example of an exemplary post that is sure to be answered. The example is for Power BI but the same tenants hold true for a question in any of the forums. 

 

I’m having trouble “thinking in DAX” WRT describing proportions.

 

I have two tables.

  1. A table that defines a taxonomy of categories (Dictionary)[Category] and unique issues (Dictionary)[All Issues].
  2. A table of questionnaire responses. Each response can be parsed across multiple issues.
  • Among many other columns, the table includes one named (Verbatims)[Unpivoted Issues].
  • It also includes a unique respondent ID (Verbatims)[UUID].

Dictionary

Category

Issue

GENERAL

General - OTHER Issues

GENERAL

General – Vague issues

Startup

Startup - Other Issues

Ports

USB Port - Physical damage

Internet/Connectivity

Wi-Fi - Vague Issue

 

Verbatims

UUID

Unpivoted_Issues

5a6b5a614e593f722a24736c39

General - OTHER Issues

3e2b46627b51763b556b5b2d2e

General – Vague issues

654e434370456f4c2d6455496c

General - OTHER Issues

654e434370456f4c2d6455496c

USB Port - Physical damage

654e434370456f4c2d6455496c

Startup - Other Issues

2d3e5e557c4e4373592754275f

Wi-Fi - Vague Issue

6a7b3255424430656a6e7a5329

USB Port - Physical damage

6a7b3255424430656a6e7a5329

General - OTHER Issues

4b394832746520534c2165794f

USB Port - Physical damage

 

I need to produce visualizations that show the proportions of issues within categories by unique respondents.

 

I didn’t have trouble coming up with a visualization for % of issues by unique respondents or count of issues within their categories but I can’t seem to get my arms around this one.

 

What I am expecting are results like:

Category Count of Issues % of issues by Distinct UUID
GENERAL 4 67%
Ports 3 50%
Startup 1 17%
Internet/Connectivity 1 17%

 

Where: Distinctcount(Verbatims)[UUID] = 6

 

Conclusion

I can't promise you that you will get your question answered to your satisfaction and yes, putting a question into the format of the example above takes a little bit of work. However, it is often the case that when you sit down to frame out the question properly you end up finding the answer on your own, and that's way better, faster and far less frustrating than writing a poorly worded question and sitting around helplessly while it gets ignored! Bottom line, putting in a little extra effort into the framing of the question will go a long way to getting your question answered or issue resolved as quickly as possible and that's really what the community is all about. 

 

Comments