Just Tell Me What I'm Doing

In my last post I took us all in the local analytic community to task for a shared lack of ambition. Having burnt some bridges there, I thought that I'd finish the job and have a go at putting a torch to the global community at the same time. Forget about moving back to France - I've picked out a nice remote location where the 'net can't find me¹.

So enough about my future. What's my beef this time?

It's our inability to agree on a common language to hold our collective body of knowledge. What do I mean? Well let's start at the top. How do you describe your profession to others? Are you a business intelligence expert? An analytics expert? Data Warehousing? Database Marketing? Performance Manager?

I bet that if I asked 10 of us in similar functions, that we would not agree on a single description. I'm not even sure that we would all recognise that we share the same profession.

You get the picture.

Why does this matter?

This inability to agree on the most basic common language presents a serious barrier to gaining a wider appreciation of what we do. How can we expect the business community to understand us when we confuse then with a mish-mash of words seemingly designed to confuse and obfuscate?

And this problem is exacerbated as we look deeper into the language we use to describe the key concepts of our profession. Have you ever tried to create a set of standard definitions for the foundation concepts of our field of knowledge? Well good luck. I have over the last couple of years but it is not something you can do easily. There is no single source of truth for the [insert here whatever term you like to describe what you do] body of knowledge. I'm going to call this analytics.

There is no single place on the web. No professional group to codify most of the key concepts we use. Wikipedia is a good start - take a look at business intelligence:

"Business Intelligence (BI) refers to skills, processes, technologies, applications and practices used to support decision making."

I can live with that. But what about plain English definitions and examples of:

• Business Analysis Models
• Business Analytics
• Business Data Model
• Conformed dimensions
• Customer Data Integration
• Data Integration
• Data Lineage
• Data Management
• Data Mining
• Data Profiling
• Data Stewards

and the list goes on. Now I'm not advocating that we need to agree on a single definition for everything, but a definition that we can all easily access, edit and make use of would be a great start.

I don't claim to have all the answers, but I am going to try and create a solution. I will start by making a modest effort to publish definitions on this blog by assembling the best information I can glean from the web, research papers, books and presentations. I will try to acknowledge the original sources wherever I can. I will naturally be drawing deeply on the work of many past and present practitioners in the Analytic field – not least Ralph Kimball, Bill Inmon, Howard Dresner, Thomas Davenport and the rest of the BI Community. I will stand on the shoulders of giants and I'm hoping that they won't complain too much.

Into this mix I will blend my own experiences and learnings. My hope is that we will begin to have a practical set of definitions that over time will contribute to the core of our body of knowledge.

In addition to blogging individual definitions, I will collect them together on the CORTEX.

Let the arguments and the plagiarism accusations begin!

.

¹ Campbell Island - a lovely place I visited in 2008. Population 0.

Comments

[Cristyna]

How crazy - more choice for web brweross, less reliable data for analytics users.Considering the most important information is sent to and can be logged by your web servers, this seems pretty pointless and will become a reason not to use Google analytics. Everyone will eventually start using other means of tracking their users and Google will no longer be able to analyse web traffic, serve the most targeted ads and make money in the same way it has before.99% of the time the data is used to enhance your browsing experience so its only the users that will suffer.

[Ted Louie]

Just an anecdote: at the holiday party hosted by my wife's company this last December--they're an enterprise software vendor just beginning to build an internal BI function--when asked what I did, only "analytics" or "databases" elicited any sort of recognition from any of the dozens of people I introduced myself to that night.

As for the other terms you list, I would NEVER use any of them in a conversation with the business, unless I wanted to end that conversation :-)

Your terms are, however, excellent questions to ask of candidates during job interviews.

[Stray__Cat]

That's actually an issue. Without an established corpus we always have to reinvent e redefine ourselves. We always need a justification to be there working because those people do some fuzzy things I do not know ...