Is ‘Practical Data Analysis’ a practical choice for small business?

You’re a small business. You’ve heard that “Data is the raw material of the 21st Century”. You want a slice of that pie, but where can you turn for practical, down to earth guidance?

So you start on Amazon and you stumble across “Practical Data Analysis” by Hector Cuesta. The title alone is promising – Practical. Data. Analysis. Let’s break that down:

Practical – /ˈpraktɪk(ə)l/ adapted or designed for actual use; useful:

Data analysis – processing data into information that can be used to make decisions.

So you start to get that hopeful, excited feeling in your gut. Could this book be a step towards becoming analytics-driven? You want to be sure, so you check out the blurb:

“For small businesses, analyzing the information contained in their data using open source technology could be game-changing. All you need is some basic programming and mathematical skills to do just that.”

Basic programming? Basic mathematics? It’s looking even more promising! In fact, now, you only have one decision to make.

Is it worth it?

To answer that, let’s start with: who is the book is for?

“This book is for developers, small business users, and analysts who want to implement data analysis and visualization for their company in a practical way. You need no prior experience with data analysis or data processing; however, basic knowledge of programming, statistics, and linear algebra is assumed.”

Next, does it do what it says on the tin?

Well, here’s where it goes wrong. The book is clearly aimed at digitally literate people with some technical and mathematical skills, which you can pick up or dust off from CodeAcademy, Coursera, or your local Learn Direct for example. It insists you don’t need data analysis knowledge to get the most out of it. The author, Hector Cuesta, has the solid, trustworthy qualifications in the technical areas that make him a promising guide. Unfortunately, judging a book by it’s cover, this gets an F.

This book is unsuitable for most small businesses. It doesn’t cover topics that are relevant to a typical small business owner. It doesn’t cover the most important part of data analysis – how to get the pay off: “discovering useful information, suggesting conclusions, and supporting decision-making“. 

Don’t get me wrong, it covers a lot of interesting, technical topics. Using a cookbook approach, Hector touches on a number of great open source tools. Therein lies the problem. It’s a technical cook book that focuses on tools, pretending to be a book about practical data analysis for small business owners who are focused on their business thriving.

The bottom line is, you can get most of this information on the internet. It will also be more up-to-date as tools change regularly. You will not learn the secrets of data analysis because you will not learn how to turn data into relevant, contextual information that will give you the insight to support your goals.

So, why should you get this book?

Are you a an analyst or developer with data analysis skills looking for a reference book of open source tools? Then this is definitely for you. As long as you remember, tools change so rapidly, the information may well be out of date.

So, is ‘Practical Data Analysis‘ a practical choice for small business? That’s a “no” from me.

Review: Data Mining For Dummies – Meta S. Brown

data-mining

I’m going to start by saying I really enjoyed this book. I’ve worked with data for nearly 20 years but there’s always something new, interesting and energising to learn.

However, my bugbear is creating a bubble for data geeks, data nerds and big data consultants. Insights are for everyone and data is one resource we can tap into for that. I am always on the lookout for books that help non-techies understand techie concepts and most importantly, what’s in it for them. This book falls squarely into that plain English, no bull approach.

Meta S. Brown aims this book squarely at domain experts – people who already have the know how that comes from working daily in their chosen fields. She then goes on to demonstrate how they can benefit from one method of manipulating data for insight: data mining.

I won’t pretend this book is perfect – it’s not. I understand some of the decisions, for example to focus on visual tools over text based ones. This is a minor niggle. The rest of the books covers in a good balance between succinctness and detail the *methodology* and *approach* to data mining.

This is key. With this grounding, the learning curve to pick up using a specific tool is reduced. And you’ll need to invest in that because that’s where this book fails to deliver. However, I’m pretty sure it would have been twice the size had Meta attempted to correct this, so as I mentioned, a minor niggle.

Bottom line: should you read this and what’s in it for you?

TLDR; Yes you should. Especially if you already do data jiggery pokery using spreadsheets, have that essential domain knowledge and want to up your game.

Even if you’re an expert data zen master, you’ll benefit from comparing Meta’s experiences, methodology and approach based on CRISP-DM to your own. A little practice-based analysis is a good thing.

Buy: Data Mining for Dummies by Meta S. Brown