Welcome to Analytic ABC blog,

The inspiration for this blog comes from discussion I started on LinkedIn back in late 2009 that drew a lot of attention. Answer to the question of “Training Analytics – Is it possible?” appears to be more challenging that I have imagined.

The latter part of the question has been subject to a great philosophical discussion, you can find the excerpts from this discussion in the philosophy section of this blog.

The former part of the question has been addressed by many in the discussion and this blog intends to help users do research in any area of analytics. It can be used in two ways:

1. I often use SEARCH bar above, powered by Google, to answer question around analytic approach, techniques, tools and training. Its great because it only searches thru content that is linked to this site, thus greatly improving relevance of the search output. It also embraces the power of social media since everyone can help to develop this blog as a tool.
2. There are 16 modules that I will continue to enhance on the ongoing basis. Each module has a vague definition around different analytic topic and specific level of complexity. The goal is to segregate content for anyone progressing in applied analytics knowledge so that level of difficulty steadily increases. After discussions with colleagues and my own research it appears that there are at least 4 levels and 4 dimensions that should be explored separately.

Here are the short descriptions of those modules:

Data Skillset:

At level 1: This is beginning of one’s adventure with data manipulations but still powerful tools can be leveraged. I intend to focus on topics around MS Excel and spreadsheet based tools here.

At level 2: Use of MS Access and other object driven software type tools (SAS EG, Hyperion-Brio, etc) and basic familiarity with SQL language.

At level 3: Use of object oriented tools with combination of programming language like SQL. Understanding of database concepts (schemas, key assignments, etc)

At level 4: Building data process flows and ability to create database solutions for analytics. Develop robust quality assurance concepts. Lastly, programming in analytic software to shape one's own point of view on best practices.

The above is roughly the progress I see an analyst must reach to tackle any analytic problem using analytical methods best fitting the resolution (as opposed to methods limited to data manipulation skills s/he has)

Method:

Method section of the blog will be a reference guide to virtual world of many smart people that apply and write about method application in the following areas:

At level 1: General methodologies are aligned with logic, critical thinking, problem resolution, and mathematical concepts. What a person needs at the beginning is structure around how the problems are solved through analytics in the applied settings. Project development methodology is therefore a crucial component of every analyst's toolkit and I will focus my efforts around best practices on this topic.

At level 2: Next level is about applying descriptive statistics in the business settings. Concepts from college level statistics courses: design of experiments, comparison of averages and college level mathematics, especially calculus are often used in organizational setting at the large scale.

At level 3: Here I aim to provide references to the rest of standard (college +) type of analytics. I will link to professionals that apply those concepts in the real world. Most references will revolve around predictive modeling used in wide variety of business applications.

At level 4: We go beyond curriculum of any college level statistics book into intersections of programming, statistics and social studies. I will focus on themes around machine learning and complexity modeling and frankly whatever else I discover out there as I reach beyond my current levels of knowledge.

Visualization:

Conceptually, analytic visualization will start around creating reporting solutions and representing data and progress to representation of information and insights in a visual manner, including concepts on visualization as an analytic technique at level 4.

At level 1: I will focus my efforts around basic charting techniques using MS Excel and other basic software. The general purpose is using data visualization techniques for reporting of aggregated data.

At level 2: This chapter will evolve from using mostly spreadsheet to use of object based tools for creating dashboard solutions.

At level 3: At this point I will focus at the more advanced data visualization concepts and use of charts to represent the insights and "story telling". I will focus on two primary subject here: one, around information visualization and two around use of more advanced charts such histograms.

At level 4: Here I will focus on advanced visualization techniques such complexity modeling, and other uses of advanced visualization techniques.

Organizational Alignment:

4 modules of this section are organized based on following principles:

At level 1: Focus on the organization from analyst perspective and analytic positioning in terms of organizations suitability to individuals' career goals.

At level 2: I will focus on a type of organizations that typically use dashboarding solutions as their primary analytical need.

At level 3: This will focus on organizations that have very particular "specialized" type of analytics as their core business need.

At level 4: One of the most interesting topics around analytic solutions in organizations is the ability of firms to adapt and embrace those solutions at the enterprise level. This chapter will focus on some of my research in this broad topic.