Powerpivot

Live tweeted #sqlpass’s Business Analytics VC webinar: 13 Excel Tips!

Standard

I was live tweeting during our monthly PASS Business Analytics VC meeting, Here are the tweets to learn about 13 Excel Tips!

Thanks everyone who attended, I hope it was helpful!

Here are some ways to follow the Virtual Chapter:
Website: http://bavc.sqlpass.org/
Youtube: https://www.youtube.com/channel/UCOiRAA4gBxEeVxwmEZ1qy1w
Twitter: https://twitter.com/passbavc
LinkedIn: https://www.linkedin.com/groups/PASS-Business-Analytics-Virtual-Chapter-6701113

Three Power Pivot Installation FAQ’s:

Standard

Q1: How Can I upgrade Power Pivot on my machine?

A1: Uninstall Existing version (Yes, it’s not intuitive, but you have to uninstall existing version). Download the version that you want to upgrade to. Install it.

Q2: I’ve SQL 2008 R2 on my machine. Can I install “Microsoft® SQL Server® 2012 SP1 PowerPivot for Microsoft Excel® 2010”?

A2: Yes, go ahead! “SQL server 2012 sp1” in the name is confusing 🙂 It doesn’t matter which version of SQL server you’ve on your machine – In fact, Power Pivot does not need a SQL server to be installed on your machine.

Q3: How do I check Power Pivot version?

A3: Excel > Power Pivot > Settings > You’ll see version # on the top part of the screen.

Power Pivot Settings Excel

I hope this helps!

Power Pivot DAX: Difference between two DATE values

Standard

Requirement:

Take difference between two data values.

Example:

EndDate: 11/20/2013

StartDate: 11/14/2013

DateDifference: 6

Let’s start writing some DAX!

Ok, seems simple, right? Try creating a measure DateDifference:=[EndDate]-[StartDate]

Did that work? NO? Does it return date?

Don’t worry, Here’s the solution. Try creating following DAX Measure:

[code language=”sql”]
DateDifference:=1.0*([EndDate]-[StartDate])
[/code]

That’s about it for this post. Here are some related Posts:
Calculate the difference between two dates in DAX
Q: How can I calculate difference between two dates in DAX (seconds, minutes, hours, days and months)
NETWORKDAYS() Equivalent in PowerPivot?

Example of an DAX Substring Equivalent:

Standard

DAX (Microsoft’ Data Analysis Expressions Language) does not have a Substring function but I needed something like that for the following problem:

I had domain/username as input and I needed to extract just the username part of the string.

Input format: domain/username

output format needed: username

Input column name: UserID

so here’s the DAX formula I used: RIGHT([UserID],LEN([UserID])-SEARCH(“”,[UserID]))

Note: The Formula is shown for demo purpose only, It may not work directly before making appropriate changes to the formula like making sure the column name is right & If the double quotes show problems, try deleting it & typing them back again.

Note that I combined some of the available DAX text functions to achieve what I was looking for. There might be other way to do this and I would be happy to learn about it too. Meanwhile, I hope if you reading this, this might give a good starting point while researching your DAX problem.

How to add additional Columns to an Existing Table in a Power Pivot Model?

Standard

Problem:

There’s a Power Pivot Model. It imports some number of columns from a table. Now, there’s a need to import an additional column to this table. How do you do that?

Solution:

Here are the steps:

1. Open Power Pivot Model.

2. Go to Design Tab.

3. Click on Table Properties:

Power Pivot Table Properties

4. From here, You can add additional columns:

Add Columns Table Power Pivot5. click on save and that should add the column to your model:

Power Pivot Edit Connections TableFAQ:

#1: I can’t see what you see inside Design Tab. What do I do?

Make sure you have the latest Power Pivot version. You can read this: http://parasdoshi.com/2013/06/04/excel-2010-how-to-check-the-installed-version-of-powerpivot-and-whats-the-latest-version/

#2: How about changing the data source & pointing to new cube/database? It’s very helpful when you switch between QA, Dev or PROD servers. Here are the steps:

you can do that too! Go to Design > Existing connections > Select Connection > Edit > Make changes > Test it! > Save > “Refresh”. Any Problems? No? Great. That’s Done!

Related Posts:
How to add a column in an imported Table while developing SQL Server 2012 Analysis services Tabular Model

Using Excel as BI Tool? Consider 64-Bit version.

Standard

Excel as Business Intelligence Tool. What do I mean?

For the purpose of this blog post, If you’re creating Excel based reports against Power Pivot Model, SQL Server Analysis Services data or creating reports by combining data from other data sources – then you are using Excel as a Business Intelligence Tool.

What’s the problem?

when you’re using Excel as a Front end tool for creating Business Intelligence reports – there’s a chance that you’ll be bringing in more data than excel could handle which would create “performance problems”.

How can 64-bit help?

Performance power offered by Excel 64-bit > 32-bit version.

Why? Because of the advantages that 64-bit computing has over 32-bit computing

How can YOU decide whether you need 64-bit version? (FOR BI FOLKS ONLY)

(prerequisite: your OS should be 64-bit)

If you don’t have Excel & planning to get it. Consider 64-bit!

If you already have Excel & it’s 32-bit – Do you see slow performance when you create your reports against Power Pivot model or SSAS? Do you have more than 4 GB RAM in your machine? YES? Try 64-bit.

What’s the down-side of 64-bit?

There are compatibility issues with 32-bit Excel add-ins. Read more here: Choose the 32-bit or 64-bit version of Office

Microsoft does not recommend using 64-bit excel.

What version of Excel support 64-bit?

Excel 2010 & onwards.

Case Study

In my current project, a Business Analyst who was creating reports against a SSAS (SQL Server Analysis services) cube & was seeing 1-2 min delays in “data refresh” each time filter values used to change. Clearly, She was not happy!

I looked at her computer configuration. She was using:

  • Excel 2010 32 bit
  • 8 GB RAM
  • 64 Bit edition of Windows 7

so I recommended 64-bit version of Excel. After the version was upgraded, we tried again! This decreased the time delays by 25x. The data was now getting refreshed in 2-4 seconds!

Custom Calculations for Invoice & Returns using PowerPivot DAX formula

Standard

This is from the thread in the SQL Server PowerPivot for Excel forum.

Problem Description

Here’s the sample data:

STATUSWEEKSALESCUSTOMER
INVOICEW01$150.00  A
RETURNW02$120.00  B
INVOICEW02$120.00  B
INVOICEW02$130.00  C
INVOICEW02$150.00  D
INVOICEW03$130.00  E
INVOICEW03$120.00  F
RETURNW01$150.00  A
INVOICEW04$100.00  G
INVOICEW05$150.00  H
RETURNW03$130.00  E
RETURNW02$120.00  B
RETURNW06$100.00  I
INVOICEW06$100.00  I
RETURNW05$150.00  H

What the user wanted was an output like this:
Excel Invoices and returns calculations

Without PowerPivo this is how the user was doing it:
“Create one pivot table filtered by INVOICE (WEEK in Columns, CUSTOMER in Rows) and second table filtered by RETURN (WEEK in Columns, CUSTOMER in Rows). Then manually calculate INVOICED pivot – RETURN pivot.”

Solution

Let’s see how DAX formula in PowerPivot can help the user so that it eliminates the “manual” calculation.

So Here are the steps:

Step 1

Create two calculated measures:

Invoiced:=CALCULATE(SUM([SALES]),TABLENAME[STATUS]=”INVOICE”)

Returned:=CALCULATE(sum(DATA[SALES]),TABLENAME[STATUS]=”RETURN”)

 

Step 2

Create one more calculated measure:

Invoiced-Returned:=[Invoiced]-[Returned]
Now from the usability standpoint, Hide measures created in step 1

Here’s the screenshot of the PowerPivot Model:
Power Pivot Excel DAX Invoice

Step 3

Let’s view this using PivotTables:
Excel Power Pivot excel and DAX

Conclusion

In this post, we saw how to create custom calculation to handle invoices and returns using PowerPivot DAX formula’s.

Paras Doshi

PowerPivot DAX: Moving Average

Standard

A quick blog post about sharing PowerPivot DAX formula to calculate moving average. Please use this as building block for your scenario:

Here’s the Data:

DATEAMOUNT
6/1/201350
6/2/201340
6/3/201330
6/4/201320
6/5/201310
6/6/201310
6/7/201320
6/8/201330
6/9/201340
6/10/201350

(usually, the date would in a date table. For the purpose of sharing building block of this formula, I kept it this way to keep it simple)

Formula (Calculated Measure):

[code language=”text”]
Moving Average Last 3 days:=CALCULATE(AVERAGE(TableName[Amt]),DATESINPERIOD(TableName[date],PREVIOUSDAY(TableName[date]),-3,day))
[/code]

Results via a Pivot Table: powerpivot dax moving average

Note that the formula is a building block and you’ll have to make changes as per your requirement and data model. Feel free to leave a comment if you need some assistance from my side. And also consider using the PowerPivot forum to reach out to community: MSDN Forum – PowerPivot for Excel

Exploring, filtering and shaping web-based public data using Data Explorer Excel add-in:

Standard

Data Explorer let’s you “Explore” (search) for web-based public data. This is a great way to combine data that you may have in your data-sources with public data sources for data analysis purposes. Sometimes your data might not tell you the reason behind the observed trends, so when that happens – you can try to see if a public data-set might give you the much-needed context. Let me give you an Example before we start hands-on w/ data explorer so that you have better understanding of importance of public datasets. Here’s a sample that I found here. So, Here’s a demo:

An auto company is seeing sales trends of Hybrid cars and SUV’s from the sales data-sources. But what is the reason behind that? company data does not show that. Someone hypothesizes that it might be because of gas prices. So they test out the hypothesis by combining gas prices information available via public data. And turns out gas prices might be the driving force of sales trends! SEE:

if the gas prices increase, then the sale of SUV go down and the sale of Hybrids go up:

data analysis combine data with public datasets

You know that public data can be helpful! So how can you search for public data-sets? Well, You can manually search online, ask someone, browse through public data repositories like azure data market (and other data markets), there’s also a public data search engine! OR you can directly search for them via Data Explorer.

Here are the steps:

1) Excel 2010/2013 > Data Explorer Tab > Online Search > type “Tallest Buildings”

excel public data search data explorer2) I selected one of the data-sets that said “Tallest completed building…. ”

excel data from internet

3) Now let’s do some filtering and shaping. Here are the requirements:

– Hide columns: Image, notes & key

– clean columns that has heights data

– Show only city name in location

OK, let’s get to this one by one!

4) Hiding Columns:

Click on Filter & Shape button from the Query Settings:

excel data shaping cleaning

Select Image Column > Right Click > Hide:

excel hide remove columns

Repeat the steps for notes & key column.

Click on DONE

5) clean column that has heights data.

Click on Filter & Shape to open the query editor

A) let’s rename it. Select column > Right Click > Rename to Height > press ENTER

B) let’s remove the values in brackets. Select Column > right click > split column > By delimiter > At each occurrence of the delimiter > Custom and enter “(” > OK

excel split a columnThis should transform the data like this:

excel data explorer split a column

Hide height.2 and rename the height.1 to height

Click on DONE

6) Let’s just have city names in the location column

click on Filter & shape to load query editor:

A) select location > right click > split column > by delimiter > Custom – Enter: ° in the text box like this:

an excel split by delimiter dataclick on OK

Hide Location.2, Location.3, Location.4 & Location.5

Select Location.1 > Right Click > Split Column > by Number of characters > Number of characters: 2 > Once, as far right as possible > OK

cleaning data in excel shaping filtering

Hide Location.1.2 and rename Location.1.1 to Location

One last thing! making sure that the data type of height is numbers.

Select height > change type > number

Also,

Select floors > change type > number

click on DONE. Here’s our filtered and shaped data!

filter data excel shape clean

7) LET”S VISUALIZE IT!

For the purpose of visualization I copied first 20 rows to a separate excel sheet and created a chart:

z excel data visualization

That’s about it for this post. Here are some related Posts on Data Explorer:
Unpivoting data using the data explorer preview for Excel 2010/2013
Merging/Joining datasets in Excel using Data Explorer add-in
Remove Duplicates in Excel Tables using Data Explorer Add-in
Web Scraping Tables using Excel add-in Data Explorer preview:

Your comments are very welcome!

Upgrading PowerPivot Excel Add-in

Standard

Here are the steps visually:

upgrade powerpivot excelHere are the detailed steps:

Uninstall existing version:

  1. Navigate to Control Panel > Programs > Programs and Features > Uninstall a program
  2. Locate the PowerPivot add-in installed on your machineun install power pivot excel
  3. Uninstall this program

Install new version:

1. Search for “PowerPivot download”. As of now, the URL is: http://www.microsoft.com/en-us/download/details.aspx?id=29074

2. Download the add-in

Now, how do select 32 – bit vs. 64 – bit? While 64-bit is recommended, you should have the 64-bit office to do that. Either case the version of the office installed should match the version of the PowerPivot installed. So how do you check that?

Excel > File > Help

power pivot 32 64 bit excel3. Since I’ve 64 bit office, I’ll install the 64-bit version of PowerPivot. For that to happen, I downloaded the 64-bit version of PowerPivot.

4. And installed it.

Install download power pivot excel

That’s about it for this post. you can also double-check the version of the installed version of the PowerPivot: How to check the installed version of PowerPivot and what’s the latest version?

Related articles