Blog – Excel Maverick

October 28, 2018October 29, 2018

Data Table What-If Analysis in Excel

Using What-if analysis is performing Sensitivity analysis using excel on a set of data, where we want to see a range of possible outcomes based on certain changing parameter values. This analysis allows a decision maker to take more informed decisions.

There are three different tools available in Excel to perform what-if analysis: Scenarios, Data tables and Goal Seek. Of these, we will be looking at “Data Table” based What-If analysis in today’s article. We can perform a One Variable What-If Analysis, as well as Two Variable What-If Analysis. What this means is, that we get to see an outcome on the data set by changing a maximum of two variables, called decision variables. (Some people may call decision variable as an independent variable).
The key is that the outcome for which we are trying to perform a What-if analysis should be a formula containing the decision variable(s).

Let us see each of the above in action.

One Variable Data Table What-If Analysis

Let us take an example where we have a sample data set that consists of columns such as the Units sold, the Cost per unit, the Selling price per unit, the Variable cost, the Sale amount, the Fixed cost and finally the Net Profit. In our example, we will take the “Units sold” as our decision variable and we will perform a what-if analysis on the outcome which is the “Net Profit”.

Step 1

Lets put the data set as shown below.

Step 2

By changing the input value (aka the decision variable) which in our case is “Units Sold”, we will see how different inputs values will impact the results – the Sale Amount, the Variable Cost, and finally our main goal the Net Profit.

We will be setting up our input variable box, which will have certain arguments that will allow us to change the range of Units Sold, and this will allow the input range more flexibility. I will explain this in more detail once we have completed the scenario.

Step 3

In this step we have to setup the matrix of data table that we need. Lets understand how this matrix is built.

I have highlighted 2 areas in the matrix – Green and Blue. Green area is for the outcome, and the Blue area is for the decision variable range of values.

Input the formulas in the Green and Blue areas as shown below. This will help us build the Data Table for What-if analysis.

Matrix with Formula in Green and Blue area

Step 4

Finally, we come to the creation of Data Table.
Select the entire matrix from cells B11:E20, and then go to Data Tab, Select What-If Analysis in the Forecast section, and then select option for Data Table.

What-If Analysis > Data Table” class=”wp-image-124″/><figcaption>What-If Analysis > Data Table</figcaption></figure>

<p>You will get a “Data Table” dialog box, wherein you will need to input the Data Table arguments.</p>

<figure class= — Data Table Dialog box

There are two arguments (1) Row Input Cell, and (2) Column Input Cell. These Data table arguments are not easily understood. So lets get a better understanding of these.

These two arguments can infact be mapped to the number of decision variables we are using for what-if analysis. In our example, since we are using only one decision variable i.e. “Units Sold”, hence we will need to input only one set of argument in the Data Table dialog box. This can be either Row or Column. Now, the question is which one should we choose.

This is where the color highlighting will help us. When we prepared our matrix, we have listed the range of values of “Units Sold” in the Blue area, which is infact a column range. This color coding will help us guide which Data Table argument to choose.

In our example, Row input cell will remain blank.
Column input cell: We will put in a reference to the decision variable from our original data set i.e. Cell Reference =$B$4.

Hit OK, and Voila. Our Data Table is ready, and Matrix is populated with all the possible outcomes for our range of Units Sold.

Data Table is created based on input range of values

Coming back to Input Variable box that we had setup in Step 2 – so let’s discuss about that a bit. We had put in the following 2 arguments: “Units Sold Start at”, and “Units Sold Increment by”. Both provide additional flexibility in our Sensitivity analysis, where we can change these parameters any time, and see the Data Table refreshed based on our new set of range inputs for Units Sold.

For example, If I want to see the Units sold incremented by 200 units instead of 500, what will be the outcome. Vice Versa, If I needed to start the Units Sold at 800 Units and incremented by 200, what will be the outcome. Hope this is clear.

Data Table refreshes based on value changed in Input Variable Box

Two Variable Data Table What-If Analysis

Now continuing with the same example, suppose we wanted to see the outcome of Net Profit, based on two decision variables being “Units Sold”, as well as “Selling price per unit”.

Step 1

This time we will be setting up the Input Range, and Matrix as shown below.

Matrix for Data Table, as well as Input Variable Box

Remember, the Blue area (Column Area) is where “Units Sold” range of values are input. Green area (Row Area) is where the second decision variable – “Selling price per unit” range of values go. And the Amber highlighted cell is the outcome reference, for “Net Profit”.

Step 2

Select the entire matrix from cells B12:H21, and then go to Data Tab, Select What-If Analysis in the Forecast section, and then select option for Data Table.

When you get a “Data Table” dialog box, enter Data Table arguments as below.
In our example,
Row input cell: Cell Reference of second decision variable, =$D$4
Column input cell: Cell Reference of first decision variable, =$B$4

Hit OK, and our new Data Table is ready. The Matrix is populated with all the possible outcomes for our combination of range of Units Sold and Selling price per unit.

Data Table is populated, based on combination of input variables

Bonus

How could we make use of the above in the form a dashboard. In the below example, I have created a dashboard like scenario selection, which a user will choose, and depending on the scenario chosen be it Optimistic, Most Likely, and Pessimistic, it will update the values in the input variable box. This in turn will drive the outcome data table.

You can download the example file from the link. Also, don’t forget to visit my YouTube channel for many more video tutorials. Happy Learning !!

October 7, 2018October 29, 2018

Passing parameters to queries in Power Query

Wouldn’t it be nice to have a complex Power Query accept values from users as arguments, and wouldn’t it be more usable if we could provide a way to give these users a dropdown list for each of the argument in the Power query, where they can choose the allowed parameter values which they are already used to doing when building dashboards in excel. We can have values passed as parameters in either of the following ways – creating parameters using Manage Parameters or by creating a function. In today’s article we will have a quick look at both, including editing code written in M language.

Who can use whats described here:

Windows users using Excel 2016 or Excel 2010 to 2013 with Power Query add-in installed
Not available for MAC yet.

This was one of the top feature introduced during April of 2016 for Power BI. However, these query parameters also apply to Power Query. So, so long as you are using Power Query, feel free to use this functionality of passing parameters to queries. These query parameters allow users to make parts of their reports / data models depend on one or more parameter values.

We are going to look at a importing sample data using Power Query, sample data provided by WHO for various countries by regions, depicting data on population, life expectancy, literacy rate, child mortality rate, and so on. Data “WHO.csv” file can be downloaded from here.

Passing Parameters using Manage Parameters

Users can define new parameters by using the “Manage Parameters” dialog in the Query Editor window. Follow the example step by step to understand more about Parameters.

Step 1

In a new workbook, go to the Data tab, and choose “From Text/CSV” option in the Get & Transform Data section in the ribbon, to import the WHO data using Power Query, and stay in EDIT mode.
In the Home tab, Choose “Manage Parameters” option, and within that select “New Parameter”

Enter the following details on the Parameter dialog screen.

Parameter Name: Put any name you want. In this example, I am putting “PickRegion”.
Parameter Description: Optional, but useful to better understand the purpose and semantics of this parameter.
Optional vs. Required: Specify whether you want to keep the Parameter a required field or not.
Parameter Type: This is the Data Type. Users can define a parameter of type Text, or Date/Time or Decimal Number or True or False, for example. Data type of “Any” provides for more flexibility.
Suggested Values: There are three options available here. Any Type, List of Values and Query. For now, Choose “Any Type”.

Any Type – will allow user to put in any type of value, when specifying the parameter. This allows values to be inputted dynamically. May also be prone to errors.
List of Values – a static list of Accepted Values can be inputted, so users may pick and choose from the list created beforehand.
Query – This would enable dynamic sets of options to be displayed to the user, based on the result of a query.

Current Value: Specify any value you want to provide as Current Value. If you keep this blank, it will throw an error. I have specified “Europe”.

You will see that the Parameter is created, and a parameter frame is displayed on screen as such.

Step 2

In this step, we will use the Parameter created in step 1 as a Filter value for the Query.

On the dataset of the WHO Query, Filter on the column “Region” as shown below.

In the Filter Rows dialog box, Select Parameter as the Type of Filter in the icon, and it will default to “PickRegion” as this is the only parameter we have created so far. Hit OK

What this will do, is this will extract the data set based on “Europe” which was the region we had selected as Current Value.

You can choose Close and Load on the Home tab, to load the extract of data into Excel.

Step 3

This time, go back to Power Query, and change the Parameter Suggested Value to “List of Values” instead of “Any Type”.

When you change the value type to List of Values, you will also need to provide all the values you need to show up statically.
Also provide a Default Value and Current Value, and Hit OK.
Observe the data set change based on the selection you just gave, in my example, it was Americas. You can choose Close and Load on the Home tab, to refresh your extract of data into Excel based on new selection.

Passing Parameters using a Function

Strategy used in this method is going to be creating a data set using a Power Query, then editing that query to allow a parameter value making it into a function, and then, finally invoking that function in another query that will pass the value to the argument.

What we want to achieve is, to have a top section where user will have a choice to select a Region (this will be the value passed onto the parameter Function, based on which the WHO data in the lower section is reflected or extracted. See screenshot below:

Top section used to select parameter value, the lower section reflects data based on selection.

Step 1

In a new workbook, go to the Data tab, and choose “From Text/CSV” option in the Get & Transform Data section in the ribbon, to import the WHO data using Power Query, and stay in EDIT mode.

We will rename the Query and “Distinct_Regions”, and perform the following steps:

Select the Column “Region”, and on the ribbon in Home tab, click on “Remove Columns” and “Remove Other Columns”. You will end up with a single column table with the Regions for each row listed.
Now, on the ribbon in Home tab, click on “Remove Rows” and then choose “Remove Duplicates”.

Click on Close and Load, to bring the distinct regions list over to Excel from Power Query.
This table will be used a create a list to be used for Data Validation, which we will need for Parameter section.

Data Validation List created from unique list of Regions

Step 2

In a same workbook, go to the Data tab, and repeat the above step. Choose “From Text/CSV” option in the Get & Transform Data section in the ribbon, to import the WHO data using Power Query, and stay in EDIT mode.

This time, we will rename the Query as “GetRegionData”, and perform the following steps:
Filter on the Region Column, and select any of the Value in the List, for example “Africa” as shown below, and hit OK.

Filter on Region Column for GetRegionData.

Right Click on the Query name “GetRegionData”, and click “Advanced Editor” option. You will see the Formula that has been entered by Excel in M Language.

Now edit the code as shown below.

What we have essentially done above, is created a function that uses a parameter “Region”, and this argument is then passed on to the line of code where data gets filtered. We have replaced a static Filter Value “Africa” to a dynamic Filter argument called Region.

Once done, you will see the Enter Parameter screen as shown below.

Step 3

In the Excel, in a new worksheet, in cell A1, enter value “Enter Region”, and in Cell A2, create a Data Validation list based on Regions list in Step 1.

Create a place where user can select value

User will be expected to choose the Value in Cell A2, which will be used as parameter value that the Function will read.

Step 4

On the same worksheet, on the Data tab, choose “From Table/Range” option, which will bring up a Create Table dialog box. Ensure “My data has headers” option is checked, and hit OK.

Create Table, for the parameter region we just created

The new table will show up in Power Query editor as below.

Click on the Table, and on Add Column tab, Click on Invoke Custom Function option. Enter the values in the Function Query field, and hit OK.

When you have Invokes the Custom Function, this will bring the Filtered Data against the Region selected, as shown below.

Click on the hyper link Table, which will then list down the data set for that region. Click Close and Load, which will bring the data set into Excel.

The data is available as we wanted to achieve. In order to test whether the data set changes based on the Region selected, In cell A2, change the selection of region, and hit “Refresh All” option in the data tab. You will see the data set changes based on selection.

Personal word

This is probably the simplest of examples depicted where a parameter table is created that slows for a single value selection. Once you get the underlying concept, you can let your imagination go wild with other fantastic possibilities you could ever dream about.

The example file can be downloaded from this link. Happy Learning !! Also don’t forget to visit my Youtube channel for many more video tutorials.

October 2, 2018October 29, 2018

Pivot with Text

Gone are the days when you could pivot only with numbers!! Yes, with newer functionalities in Excel you can do pivoting with text as well. In this article, we will ponder upon how to make use of Pivot table to display text values. Pivot tables have been excellent, mainly for summarizing numbers. But, when it came to displaying text in the values area of a pivot table, there wasn’t much available until a DAX function was introduced called CONCATENATEX.

The DAX language has provided us with more possibilities than ever before. It allows for calculations that remained previously impossible. Calculated Fields and Measures are the new terminologies that one must get accustomed to when dealing with pivot tables with advanced functionalities.

Who can use whats described here:

Windows users using Excel 2016 or Excel 2010 to 2013 with Power Query add-in installed
Not available for MAC yet.

Lets first study an example, wherein one may need to have a text value show up in pivot area.

Assume a set of sales data by Sales representative and Product as below.

Traditionally, when using Pivot to analyze the data, one could have summarized the above in the following manner:

However, there was no way to include text values within the pivot area, to show up as below:

Adding a Measure that uses CONCATENATEX function is what makes everything possible. This is what we are going to see in what follows.

The syntax for CONCATENATEX is (Table Name, Expression, Delimiter)

What becomes obvious from the above, is that, it is necessary to create a pivot table based on a source table, and not a range of cells. Next, Expression refers to the Column in the Table whose text values we intend to bring into the pivot area. Lastly, Delimiter is what separates the text values, be it a dash, comma, or a hyphen.

Follow the below steps in sequence to achieve the result of having Text values in Pivot area.

Step 1

Create a Table from the data set. Download the file with the dataset for ready reference. Place your cursor on any one cell in your data set, and then Go to Insert Tab, and choose “Table”. It will bring up a “Create Table” dialog. Select option “My table has headers”, and hit OK. By default, the table will be called Table1. Click on the Table Tools Design tab in the Ribbon and assign a new name for the table as “RepSalesData”.

Step 2

Go to Insert Tab, and choose “PivotTable”. This will bring up a dialog box “Create PivotTable”. Ensure Table/Range field shows up as “RepSalesData”. Choose “New Worksheet”, and most importantly, check the option “Add this data to the Data Model”, and hit OK.

Step 3

This will show up the Pivot Table Fields section on the right side of the screen. When your pivot table is based on the Data Model, you will see a difference in the Pivot Table Fields list. The name of the table will appear at the top of the fields from that table. Right-click on the name of the table “RepSalesData” and choose Add Measure.

Pivot Table Fields, Table shows up at the top of the Field list

Step 4

You will get a new dialog screen “Measure” where the magic begins. Give a Measure name “Rep_Products”, and description. In the formula area, enter the below formula:

=CONCATENATEX(RepSalesData,RepSalesData[Product],”, “)

Click on “Check DAX Formula” button. This will check for error in the formula entered. If there are no errors, it will display so. Choose “General” as the category for the Measure newly created. Note, for a text result, the only valid choice is General, so leave the selection as General. Hit OK to create the new Measure.

Measure is defined, Category chosen as General

Step 5

The new measure will not show up in the pivot table, instead it will show up as a new field in the Pivot Table Fields list. Start to build your pivot table by dragging fields to the Rows and Columns area. Drag the new field to the Values area. On the PowerPivot Tools Table Design Tab, turn off the Grand Totals for both Rows and Columns.

Step 6

Finally, you will see the Product list showing up next to each other in the “Rep_Products” column in the Pivot table.

Product List showing as Text values in Pivot area

However, if you observe closely, the last row in the pivot table shows up with iPhone X repeated twice. What this means is that the Sales representative has sold iPhone X twice in the given data set. If your intention is to only display the unique products sold by that Sales rep, you need to nest the VALUES function within CONCATENATEX function.

Step 7

Edit the measure we created above, and replace the below formula in lieu of the earlier one.

=CONCATENATEX(VALUES(RepSalesData[Product]),RepSalesData[Product],”, “)

With the VALUES function, duplicate values are removed, and only unique values are returned.

Unique Product List showing up as Text Values in Pivot area

Will also be updating my YouTube Channel with a tutorial video of the above example, and how to get the text value show up in Pivot. Keep watching, and send me comments.