Here is a sample dataset of learners who have cleared different stages of multiple courses on offer within an Organisation:

Learner

Stage completed

Course

Bill

Stage 1

Public Speaking

Bill

Stage 2

Public Speaking

Bill

Stage 3

Public Speaking

Susan

Stage 1

Effective Communication

Bob

Stage 1

Public Speaking

Bob

Stage 2

Public Speaking

Sheila

Stage 1

Effective Communication

Sheila

Stage 2

Effective Communication

Sheila

Stage 3

Effective Communication

Frank

Stage 1

Effective Communication

Frank

Stage 2

Effective Communication

Henry

Stage 1

Public Speaking

Henry

Stage 2

Public Speaking

Bill

Stage 1

Effective Communication

Bill

Stage 2

Effective Communication

From this sample dataset, one may want to know how many participants have completed each stage of these multiple courses. The expected result is shown below:

Row Labels

Stage 1

Stage 2

Stage 3

Effective Communication

1

2

1

Public Speaking

2

1

Grand Total

1

3

2

In this workbook, I have shared 2 solutions - one using formulas and the other using the Power Query & PowerPivot.

Refer to a simple 5 column representative inventory dataset of a Glass manufacturer:

Model

Length (MM)

Wide (MM)

Thk (MM)

CAT

HX9-G-ARD

1071

273

3.5

A

MYP-G-3RD

580

535

3.2

B

EPO-G-3RD

580

535

3.2

A

MYG-G-3R

966

350

3.2

A

MYN-G-3RD

649

530

3.2

A

GM SPIN-G-3FD

882

395

3.2

A

MY8-G-AR

880

400

3.5

B

GM2-G-AR

880

400

3.5

A

From this inventory data, one has to furnish customer orders based on specific dimensions demanded by them. A typical Customer request would be to supply glass sheets as per the following dimensions

Length (MM)

Wide (MM)

Thk (MM)

CAT

780

542

3.5

A

The firm may or may not have glass sheets of this specific size. The objective is to identify glass sheets, from the inventory on hand, which match customer specifications. If there is no exact match, then one must be able to obtain all inventory items which have the same Thk (MM) and CAT as the customer specified dimensions but the Length and thickness should be more than equal to the customer specified dimensions. The length and width can then be trimmed to match the exact customer dimensions. Furthermore, the result returned should:

List only the Top 30 glass sheets available in inventory; and

List those Top 30 glass sheets in ascending order of wastage (wastage caused when the glass sheet is trimmed to match the customer specified dimensions)

You may refer to my solution in this workbook. I have shared two solutions - one using Excel formulas and the other using Power Query a.k.a. Get and Transform in Excel 2016. Please read the Comments in cells F1, J9 and J16 of the "Solutions" worksheet. The difference between the 2 solutions is:

Formula driven solution - This is in range J10:AM14 of the Solutions worksheet. This is a semi dynamic solution (as compared to the Power Query solution). To get the models in ascending order of wastage, one will have to create an Area column in the base data and sort that column in ascending order.

Power Query solution - This is in range J17:AM21 of the Solutions worksheet. This is a dynamic solution. Just change the customer specified dimensions in range G2:J2 of the Data and Query worksheet. Thereafter just right click on any cell in the range below and select refresh.

A simple Pivot Table (with a slicer) created from this dataset looks like this

The objective is to determine the Top 3 users of each week for each slicer selection. Unfortunately, there is no way to sort multiple columns of a Pivot Table all at once. Once may either sort by the Grand Total column or by the individual week wise columns. Since we do not want to sort by the Grand Total column, the only way out is to sort the individual week wise columns. The expected result should look like this:

I have solved this problem by using CUBE formulas. You may refer to my solution in this workbook.

Here is a small sample of a Project matrix which shows tasks to be accomplished for various projects. There can only be upto 6 tasks per project.

Project Name

Task1

Task2

Task3

Task4

Task5

Task6

Project1

Painting

Chef

Gardener

Project2

Tiling

Digging

Engineering

Project3

Mechanic

Engineering

Here is a competency matrix showing the competencies of employees on different tasks. 1 indicates that the employee is competent to perform that task.

Task

Tom

Jane

Mary

Paddy

Lynda

Painting

1

1

1

1

1

Tiling

1

1

1

1

1

Plastering

1

1

1

1

1

Digging

1

0

1

1

1

Mechanic

1

1

1

0

1

Detective

1

1

1

1

1

Engineering

1

1

0

1

1

Boxer

1

0

1

1

1

Chef

1

1

1

1

1

Gardener

1

1

0

1

1

Banker

1

1

1

1

0

From these two tables, one may want to generate another table showing which employees can be assigned to which project (only those employees should be assigned to a project who can complete all tasks). So the ideal solution is to create another column (8th column) in the Project matrix table above which should have a drop down (Data > Data Validation) for every project showing which employees are competent for that project.

Here's an illustration:

Assuming that the Project matrix is in range A1:G4 (headers are in row 1)

In cell H2 (for Project1), the drop down should show Jane, Lynda, Paddy and Tom. Mary should not appear there because she cannot perform one of the 3 tasks required to complete the project i.e. Gardener.

In cell H3 (for Project2), the drop down should show Lynda, Paddy and Tom. Jane and Mary should not appear there because they cannot perform the Digging and Engineering tasks respectively.

The solution is dynamic for the following:

Projects added to the Project matrix Table; and

Tasks added (upto 6 only) or edited in the Project matric Table; and

Employees added to the Competency matrix Table; and

Tasks added to the Competency matrix Table

I have solved this problem by using:

Power Query; and

Formulas in Data > Data Validation.

You may download my solution workbook from here or here.

Assume a single row of data with numbers and blanks appearing at random intervals. The objective is to sum the largest 5 of last 10 numbers in that row. Solving this problem entails multiple steps:

Identify the last 10 numbers in that row i.e. starting from the right hand side, identify the last 10 numbers

Identify the largest 5 of those 10 numbers

Sum those largest 5 numbers

Here are the steps

Suppose the numbers and blanks are in range A2:V2

Type 10 in cell X1

Enter this array formula (Ctrl+Shift+Enter) in cell X2

This post is in continuation of an earlier post where I applied Excel's Business Intelligence tools (PowerPivot, Power Query and PowerView) to analyse the Sales data of an E-Commerce Company. So, just for starters, in that post, I have basically sliced and diced the Sales dataset of an E-Commerce Company from multiple perspectives/facets to know the performance of this Company by Categories, Products, SKU's, Shipping Cities, Modes of payment etc.

After analysing the data, I have also visualised that data using PowerView. From the link shared above, you may download the workbook, watch the YouTube video and see a PowerBI desktop custom visual ("Sankey Diagram"). In this post, I have taken the same dataset and showcased/discussed the following:

1. How one can discover insights from this data with minimal effort using a Custom PowerBI desktop visual called "Sand Dance"; and
2. How one can query the dataset using "Natural Language" on a web browser (using www.powerbi.com); and
3. How one can query the dataset using "Natural Language" using Cortana (Microsoft's personal digital assistant in Windows 10).

For aspects 2 and 3 above, here are a few "Natural Language queries" which returned the correct result:

1. Show total revenue and growth in total revenue over previous month where order status is delivered by month in ascending order of month order as a Table
2. Show total revenue by category as a column chart
3. Show total revenue by order period as a pie chart in descending order of total revenue
4. Show total revenue by order period as a pie chart in descending order of total revenue where day of week is Sunday
5. Show Business generated from new categories by month where order period is mid day, payment type is COD sorted by month order in ascending order as a table
6. Show total revenue where portion of month is first half of month

Enough talking!!. You may view all three aspects mentioned above in this YouTube video

You may download the Powerbi desktop workbook from here and play around with the Sand Dance visual yourself. The PowerBI.com service also allows one to Publish reports to the Web (which can be viewed and interacted with by anyone). This is currently in preview stage and may become a payable service later. You may view and interact with the Sand Dance visual here:

Consider this simple two column table showing text entries in column A and the corresponding numbers in column. Assume this data is in range A2:B11 (headings are in A1:B1).

text

Value

A

1

B

2

C

3

D

4

E

5

F

6

G

7

H

8

I

9

J

0

The objective is to generate the numeric code for text code of any length entered in a certain cell. For example, a user will type a certain text code, say ABEJ and the expected result should be 1250. For JABF, the result should be 0126. The text entry and text length are both user determined.

With ABEJ, typed in cell D2, enter this array formula in cell E2

Assume a simple two column dataset with dates in column A and numbers in column B. The dates in column A are from January 1, 2013 to December 31, 2016 and numbers in column B are for the period January 1, 2013 to December 31, 2015 (there are no numbers for January 1, 2016 to December 31, 2016).

The objective is to "Compute an average for each day of calendar year 2016. The average should be for the occurrence of that day in the previous 3 years". Here's an example:

1. January 1, 2016 was a Friday (the first Friday of 2016) and is in cell A1097
2. In cell B1097, the average should be computed as: Average of the "First Friday of each of the previous 3 years"
3. January 8, 2016 was a Friday (the second Friday of 2016) and is in cell A1104
4. In cell B1104, the average should be computed as: Average of the "Second Friday of each of the previous 3 years"

I have solved this problem with the help of the PowerPivot. You may refer to my solution in this workbook.

Assume a 4 column dataset (a small sample) as follows:

City of Origin

City of destination

Mode of Transport

Passengers travelled

New Delhi

Pune

Air

123

New Delhi

Mumbai

Air

213

New Delhi

Kolkata

Air

125

Chandigarh

Jammu

Bus

785

Chandigarh

Amritsar

Train

567

Given this dataset, one may want answers to the following questions:

1. Of all those passengers who originated their journey (City of Origin) from Chandigarh, how many terminated their journey (City of destination) in New Delhi via different modes of transport; and
2. Of all those passengers who terminated their journey (City of destination) in Jammu, how many arrived in Amritsar (City of Origin) via different modes of transport; and
3. Of all those passengers who travelled by Bus, how many travelled from City A (City of Origin) to City X,Y,Z (City of destination)

While one can analyse/slice and dice this data using Pivot Tables, one cannot visualize this data very clearly (even after creating a Pivot chart). I have attempted to visualize this data using a software called PowerBI desktop (a free for download and use Business Intelligence software from Microsoft which rolls all of Excel's BI tools into 1 - PowerPivot, Power Query, Power Map and Power View).

You may download the source Excel workbook and the Power BI desktop workbook from this link.

In this workbook, I have Sales data of an E-Commerce Company for 3 months. The typical columns in the base data are:

1. Order Date/Time
2. City to which orders were shipped
3. Order Number
4. Payment Type i.e. Cash on delivery, Net Banking, EMI's
5. Order Status i.e. Delivered or cancelled
6. SKU's which the ordered items fall into
7. Products which the ordered SKU's fall into
8. Categories which the ordered products fall into

Given this simple tabular representation, one may want to analyse and visualize this dataset from multiple perspectives based on user selections, such as

"What was the revenue earned from the Top 5products in the A100 category in April for orders shipped to New Delhi?"

In this query framed above, the end user should have the leeway to select any/all of the underlined facets. So one can either choose revenue earned or Number of orders. Likewise, one can either select Top 5 products or Top 15 products/Top 5 SKU's etc.

With relative ease, one should also be able to "Perform an affiliate analysis" showing which categories are ordered together (to study affiliations). Please review this post for an independent discussion on "Affinity Analysis".

Furthermore, one should be able to perform a free form timeline search such as - "I would like to study growth in Total revenue of March 2-8 2015 over Feb 1-4 2015"

You may download the workbook from the link shared above.

You may watch similar videos showcasing the capabilities of Business Intelligence in MS Excel:

Here's a video showing the capabilities of this Sales data model

You may also watch this short video to see how I visualized the revenue flow from Categories to Shipping cities during different Order periods using Custom visuals available in PowerBI desktop.

Please feel free to download the PowerBI desktop workbook of the video shown above from here.

For a detailed overview of Sankey diagrams (a Custom visual available in PowerBI desktop), you may refer to my Blog article here.

Another great Custom visual (Sand Dance) which allows data discovery has been shown at this link. At that link, you will also be able to see how I queried the underlying dataset using "Natural Language".