Follow by Email

Mar 9, 2014

Analysis of HR Emails

This study has made an attempt to understand, what HR's in a day dealt with at different working hours, by analyzing their emails from a particular organization. Sample size of the study consists of 7 HRs emails who are located at two different locations of the same organization in a country and for 15 different days that are selected randomly from a quarter.

After sanitizing the data and removing unnecessary characters and punctuation's, data has been prepared and transformed in such a way to get frequency of words used by each hour. A day has been divided into seven categories viz., first hour, second hour, etc. Correspondence analysis has been chosen keeping in the mind for graphical representation of the processed data (below is the output - prepared for presentation purposes). All analysis has been carried out using open source statistical computing tool "R".


Analysis tells us that, 'First' hour of the day's had always been dealt with HR policies, queries, and aspects related to appraisal, roles and productivity (higher number of aspects (5) in first one hour). And, 'Second' hour went completely for addressing leaves and grievances of the employees. Coming to 'Third' the turn of reviews and offers, however, both second and third dealt with less number of aspects i.e. only two. Surprisingly, none of the aspects at the fourth hour. Moving to fifth hour again good number of aspects (4) have been dealt viz., payroll aspects of the employees, diversity and safety in the organization and about interviews. Sixth hour went for addressing training needs and calendar related aspects. Last hours (2.23 hours on average) has been dealt with development, retention and performance aspects.


Author has worked extensively in the HR analytics and can be reached at mavuluri.pradeep@gmail for related discussions.

Jan 13, 2014

Operational HR Analytics - Application to Attrition

As employee attrition continue to be expensive for organizations; businesses demanded increasing flexibility as interest in what can be done with Human Resource (HR) data and intensifying opportune in its preparation/planning are gaining pace, that, directs towards operational HR analytics.


As evident from above figure, reports (basic) are entirely backward looking aspect with less return on investment (ROI). Current phase is result of availability of increase in data across the organization and all-encompassing information technology progress that helps to predict what went wrong and integrate them with future outcomes. Majority of businesses today though uses predictive analytics for their organizations, return on investment had been slowed down and not effective as these analytics are not yet in the stage of adaptive application i.e. real time to the organizational operations/needs.

In current state of human capital management where dynamic workforce requirements, acute competition for talent, rapid technological advances not providing enough time to ingest for adoption and national/natural upheavals in one part affecting overall business in the other parts requires operational analytics that can foresee rising business needs of an organization. My experience with the help of open-source statistical computing environment viz. R, towards developing such environments had been yielding higher return on investment for businesses.

Author has worked extensively in the HR analytics and can be reached at mavuluri.pradeep@gmail for more details. 

Oct 19, 2013

Indian Car Brand Perception Survey Report

A recent survey during current festive season was done on seven different automotive brands and taken there preferred attributes for a purchase/like. The underlying dimensional graph characterize how customers differentiated between attributes+.
Following are key observations: 
It seems people are paying less importance to “price” factor, however, aspects like good value for money, parking space, servicing attributes work towards existing dominant brands. Coming to newer brands, attributes viz., performance, attractiveness, technology, safety, suitability, zippy (quick and energetic) driving their purchase/like. Two other brands are preferred for their sporty, venturous, exterior looks, athletic and stylish nature.
+Survey data comes from a specific Indian City.

Jul 4, 2013

Appropriate Training Objectives for Strong Analytical Foundation

Organizations likes to invest in developing their analytical talent in such a way that they can compete for global competencies, for that training objectives should be appropriately identified and cannot ignore strong foundation. Herein, they should cater for both semi-skilled (graduates of M.S.) who has awareness but lack of domain/practical experience and non-skilled (other graduates). Below are the three important objectives for strong analytical foundation.
  1. Ability to link learned analytical capabilities to real-world situations.
  2. Knowledge of basic analytical concepts, and their applications.
  3. Obtain cognizance to synthesize the constituents of an analytical project and convey the results in a clear manner to whom so ever.
Reach me at mavuluri.pradeep@gmail.com for more details. 

May 21, 2013

Time Period for Analytical Positions Recruitment.

Though having rough estimate of how much time each analytical position recruitment takes (i.e. due to experience in the filed); thought of giving a quantitative touch using a real sample. Herein I would like to thank 13 recruitment agencies that provided the requested details. Most of the sample belong to India recruitment only.

Below figure provides average time period i.e. number of months to fill a said position for analytics. Wherever possible we have provided domain (Retail/CPG & BFSI) specific time as our sample has around 30-34% of data for these domains. 


May 18, 2013

R (Web Server) Solutions - Amplifying Artichokes

Every month I see one or more new R based web server solutions coming into the market, sight seeing some of them thought of sharing one of my old architecture map manifested to the client long back in early 2009 (good to see quick spreading of scalable and customizable open source statistical computing tool in the market).

Please feel free to reach me at mavuluri.pradeep@gmail.com for more details.

May 16, 2013

Implementing some BootStrap stuff in R


“The bootstrap is a computer-based method of statistical inference that can answer many real statistical questions without formulas”

  (From "An Introduction to the Bootstrap" by Efron and Tibshirani, 1993)



May 13, 2013

Analytical Skills Development Spending.

Challenged with the acquisitiveness for adaptability and agility, analytical service organizations are turning to real-world work & emergent technologies/methods to develop next-generation analytical skills (talent). However, corporate bosses seems to playing it cautious when coming to their spending on analytical skills development.

According to our survey conducted by us across a good number of analytical services organizations in India, most expect to increase their current spending on analytical skills development (43.8%) in next six months. Around 27.2% organizations expect to spend aggressively in next six months. Only 6.7% expect spending to decrease in next six months. While 22.3% of organizations spending on analytical skills development continue to be frozen in next six months.


May 3, 2013

Taking workforce to immediate and next levels of analytical requirements


In today’s world, where the demand for analytical skilled workers is greater than the supply, employers’ needs to provide critical trainings to their workforce for two very good reasons (short-term) that enable current key deliveries for them, and (long-term) innovative and career development oriented that helps workforce to stay with them.


 





Reach me at mavuluri.pradeep@gmail.com for more details on analytical trainings.

May 2, 2013

Step Towards Making Data Analysts More Productive in Shorter Time?

Today lots of reports have been published and expected more;  highlighting analytical skills shortage, however, rare of them address, how it can be achieved in short term as in long-term supply and demand factors will match. Below are few things, which I have implemented in the past which helped in short term stint of less than three months for data analysts to become more productive:

1)  Focusing on business problems and help analysts define them appropriately. It worked superbly, since addressing or questioning required analysis helps them to quickly grasp complexity.
2)  Moving away from process of ETL.
3)  Make them learn and comfortable with automating repetitive tasks during the analytical cycle.
4)  Make them map business questions to analytical solutions using several mind map games (can be in both class-room environment or offline/online group environment)
5) Leverage them against appropriate analytical tools. 

Please feel free to reach me at mavuluri.pradeep@gmail.com to know about process more.

Knowing your data is first step for insightful analysis:

Let what ever be modeling algorithms, knowing your data is first step for insightful analysis.

 

Given today's buzz of big data, where more than 90% of the data's usefulness is questionable, it is not bad to stick to basics of the data analysis, i.e. first step, understanding received data properly.

For instance, if you receive a social media data for a single day and if it has 50 million records, knowing existent values format and frequency makes your first step for selecting appropriate modeling algorithm.