Category Archives: Analytics

Transforming Shadow Analytics

This content is relevant only if you are a CIO (or IT decision maker). Here is the video of the session I conducted at CIO Power List event on 4th May, 2018, at Conrad, Pune. Shadow Analytics has been around ever since “shadows” – also called end users – are around. Everyone knows about. Some people tried to eliminate it. Nobody succeeded.

This 30 minute video explains how to use Shadow Analytics as an opportunity to empower rather than restrict users and improve effective utilization of data.

Demos Included

The demos included in this Shadow Analytics video are:
Flash Fill, Insights, Explain the increase and Q&A.

Insights option in Insert menu of Excel

What is Shadow Analytics?

It is all kinds of data capture, clean-up, manipulation and report generation performed by end users without IT intervention.

If you generate a report from a business system (which is built or managed by IT), it is alright. But if you copy paste data from multiple such reports into Excel and then generate a new report, it becomes “Shadow Analytics”.

As you can imagine, it is difficult to eliminate it. Irrespective of how much time and effort you have spent on creating the most flexible ad-hoc reporting systems, it is impossible to provide every possible variation that users want. Therefore, Shadow Analytics has always been there and is likely to survive in the foreseeable future.

Problems associated with Shadow Analytics

Primarily two problems. It is extremely error prone and time consuming. There are lots of related problems. The root cause is that data is handled in a casual manner without regard for its recency and in a completely undocumented manner.

This can lead to wrong decisions, delayed decisions, increased operational risk and enormous wastage of precious time.


It is impossible to handle and correct the data sources and deliver data to users in a manner which is so easy that they stop doing the manual capture and clean-up altogether.

Once clean, accurate and updated data is available as input, creating reports can be done by end users in a more informed and productive manner.

Left hug Computer Right hug

How to create Custom Data Labels in Excel Charts

I received a query from a customer about dynamic data labels for charts. Instead of replying directly, I thought of writing this article. This will help all of you in refining your charts. The idea is to create a chart which explains the fluctuations using text based explanations. The best part is, the explanation can be a part of the data itself.
Watch the two minute video and read details below. This works with Excel 2013 onwards. I have also included the solution for older versions, which is not as elegant, but it works.

Continue reading How to create Custom Data Labels in Excel Charts

How to calculate YOY growth in Excel Pivot Table

Estimated reading time 3 min. Works with ALL versions of Excel.
The data is a list of sales transactions, two columns – amount and date.
We have 5000 transactions over many years. We want to know how the business grew year on year. Here are the steps…

SNAGHTML48223f                  image

Continue reading How to calculate YOY growth in Excel Pivot Table

Quality Management 4: Histogram (any version of Excel)

Histogram is used to visualize the frequency with which data occurs. This is a good way of understanding data more than just sum and average. It is a good idea to look at each data set you get as a histogram. Here is how you do it in Excel.

Quality Management: Histogram in Excel

Continue reading Quality Management 4: Histogram (any version of Excel)

Instant benefit: Try Distinct Count wherever you are using Count

SUM and COUNT are the most common methods of summarizing data. It is easily done in Pivot Table or any other analytical tool. What is equally important is DISTINCT COUNT. But it is not commonly used. Why not? Firstly, due to lack of awareness and secondly, due to lack of that feature in Pivot Table. Let us solve both problems in the next 10 minutes.

distinct count or unique count

Continue reading Instant benefit: Try Distinct Count wherever you are using Count

Act Now: Discover one new and useful thing from familiar data

We get data and make reports repeatedly. Often we forget to look at the same data in different ways. Due to this unbelievable amount of useful information is lost.

Act Now is a new idea I am trying. It asks you to do some activity and post the results.

Act Now for post results

Photo credit: Machine Project / Foter / CC BY-NC-SA

Continue reading Act Now: Discover one new and useful thing from familiar data

What did I learn today: Pivot Table Transpose

Consider a pivot table which has many fields in row as well as column area. Now, for whatever reason, you have to transpose the pivot table. Whatever is in the rows has to go into columns and vice versa. We cannot use Paste Special Transpose with Pivots.

The only choice seems to be manually dragging and dropping fields across row and column areas. Not only is this cumbersome, but it can also lead to mistakes. Don’t worry. I just found a smarter way.

Add a Pivot Chart. Never mind which type. Choose Pie because it takes least amount of effort graphically and it happily ignores child series of data. Now click inside the chart. Choose Design tab and click Switch Rows / Column. It instantly transposes the row and column fields. Delete the chart. Job done.

Pivot table transpose  image  image

This works with Power Pivots as well. For large pivot tables, you may get the maximum series limit reached error for charts. Ignore that error and continue – because in this case, the chart is just a temporary means of achieving transpose operation.

In-depth: Excel Recommended Charts

Now that we know about Excel Recommended Charts, let us explore it in-depth. This is an implementation of artificial intelligence or machine learning at your fingertips. Don’t underestimate it… exploit it.

Excel recommended charts

Photo credit: Elsie esq. / Foter / CC BY and salendron / Foter / CC BY-NC-SA

Continue reading In-depth: Excel Recommended Charts