Book review: BIG DATA
Using smart big data analytics and metrics to make better decisions and improve performance
Genres:
- Information management
- Data Science
- Business Statistics
Review posted on:
9.1.2023
The number of pages:
256 pages
Book rating:
3/5
Year the book was published:
First edition published 2015
Who should read this book:
- People starting out in data analytics.
- People wondering if “this data fad” is worth their time and if it is, where and how they should start.
Why did I pick up this book and what did I expect to get out of it:
The first book that I read by Bernard Marr was Big Data in Practice which I disliked very much as it provided no value to me. But then when looking up which book I should read about data analytics or big data this book was highly recommended and had great reviews. I was skeptical, but the back cover of the book got me intrigued as the author mentions a “SMART” model in which each step covers something I have learned to be important when doing data analysis. Plus the text mentioned that I “will get to learn from clear explanations and countless examples how successful organizations use the model to get ahead”. Well, that got me to pick up the book and at least give it a go for a few pages. “Spoiler alert” – I finished the book.
The one thing that I was looking to get out of this book when reading the reviews, the back cover, and the table of content was a deeper understanding of the “SMART” model and each of its steps backed up by depictive and concise examples backed by technical insights.
My thoughts about the book:
Bernard Marr did a good job breaking down the main tasks a data analyst should have his focus on when working with a company’s data and reports/dashboards. I liked how his “SMART” model comes full circle that way. On the other hand, I found his examples very simplified and not providing too much in-depth technical insight into how to implement each step of the model to get the most out of it. One of the issues I had with the book is that it’s a book for beginners in data analytics, or for management who needs to learn a little bit more about what data analytics and big data are. The book has no technical insight on big data or analytics, which is good if you are looking for a book that only covers the basics, on the other hand when being promised to get explanations and “tutorials” on how to implement practices such as leading firms do, you kind of expect some technical insight. Don’t you?
Also, the title could be different as the content of the book actually doesn’t have to do so much with big data as it does with data analytics. Yes, the author explains what big data is and provides some case studies as also benefits and traps of big data but overall the “SMART” model seems more appropriate for data analysts as a roadmap on how to better do their job be it having big data or not. Overall the book provides value if you want to know the basics of data analytics and what big data can also “bring to the table” but keep in mind that this book is from 2015 and that is one of its shortcomings – it lacks a lot about big data that has come to light since then, also there are many new case studies that could be shared for each step of the “SMART” model. Maybe the author will do a “fresh and updated edition” of the book one day.
If you decide to pick up this book let me know what are your thoughts on it.
A short summary of the book:
The author starts out with how the “world” is getting smarter with all the smart gadgets which collect data and how companies in different fields use that data to improve their users’ lives. Then in the first chapter, the author paints the picture of what are the pitfalls of big data where you can get lost, and how his “SMART” model saves you from those pitfalls. In the second chapter, he explains why you need to know your (company) strategy and that is because you need to know which questions to answer to get valuable insights from your data. And by doing so you know which data you need to get those answers. Without this step, it may happen that you collect and store the wrong data and can’t get correct insights on how to improve your business. The author also tries to explain the SMART strategy board by using a pear tree analogy (what are the roots, trunk, and major branches).
In the third chapter, the author at first explains where and about what companies like Facebook, Google, and other institutes such as banks and insurance collect and process data. You also go over the types of data. Then he goes on to explain metrics and data for strategic advantage. The fourth chapter covers the five key formats of data, the different ways of analysis for each format of data, as also combined analytics which opens even more possibilities for valuable insights. But the author also mentions the dangers of doing so without the consent of the users/the observed. He points out that we the users give companies data for free and we may even not be aware that we do so. Do you know what your “Likes” on social media say about you? If not, be sure that those companies (social media) do and if not yet they will soon be selling that information to other companies you have probably never been in contact with.
In the fifth chapter the author explains more about reporting results, what are the pitfalls, what are the key ingredients of a successful visualization that you have to have in mind when creating a report or a dashboard. One of the biggest mistakes that companies might make is that they want to use the current or next “hot” BI tool which might cause the company more costs (due to training of employees who will use the new “hot” BI tool) and may take more time to prepare the reports or dashboard which might cause the company to miss an opportunity due to not having the data on time to make the right decision. In the last chapter, the author comes full circle with the topic of transforming your business based on collected, analyzed and presented data which gave you certain insights into your business, into your customers, and how they interact with your company and product or service. You also need to have an open mind and “eyes wide open” for any new opportunities that you have not thought of but the data clearly shows that it is there. Customers or users will not always use your product or service as you intended them to so, and by researching how, when, and why they use it you might find a new niche. Also, the processes and the equipment you use can be used in a different way or optimized by going over the numbers in different formats (spreadsheets or visuals).
My notes from the book:
- The basic idea of Big Data is that everything we do is increasingly leaving a digital trace (data). With access to ever-increasing volumes of data and our ever-increasing technology capability to mine that data for commercial insight is the new driving force.
- Just because we can measure, monitor and access everything doesn't mean we should. The danger is that we get lost in a sea of data that delivers no value whatsoever. Don't concern yourself with all the metrics and data that currently exist. Only focus on the metrics and data that will help you answer your specific "SMART" QUESTIONS.
- Most businesses are already data-rich, but insight poor. Are you one of them?
- Instead of starting with the data, start with your business objectives and what you are specifically trying to achieve. This will automatically point you toward questions that you need to answer which will narrow data requirements into manageable areas.
- Data discovery is a process of looking at data with no questions or agenda to see what the data shows you about your business. If you have huge amounts of data that can be mined and analyzed then it's definitely worth spending 10% of your analytics efforts on data discovery.
- Data is a strategic asset but it's only valuable if it's used constructively and appropriately to deliver results. The value of data is not the data itself but it's what you do with the data.
- The human brain is wired to see patterns and connections so visualization taps into a natural process and speeds up comprehension.
- The danger with models/algorithms and their predictions is that each of us can be pigeonholed by all sorts of organizations and businesses based on probability, not reality.
- Even if a business uses data visualization there is still room for confusion if the way data is visualized is constantly changing. By institutionalizing data visualization you ensure that the information itself is presented in a common way across the whole company. For each sector of the company specify the key variables, how they should be displayed visually, and in some cases even the relationships between the variables and forecasts based on the relationships. This way executives from different branches spend less time trying to understand data and much more time putting the data to use and making better decisions.
-
The ingredients of a successful data visualization:
1. Identify your target audience - who is going to see it and what do they already know about the issue being discussed? What do they need and want to know? What will they do with the information?
2. Customize the data visualization - data visualizations should always be customized to the recipient and only include what they need to know.
3. Give the data visualizations a clear label or title - don't use cryptic or clever titles.
4. Link the data visualization to your strategy - if the data visualization is answering any specific or strategic questions include the question in the opening narrative.
5. Choose your graphics wisely - don't feel the need to fill all the space in the report or dashboard. Don't use too many different types of graphs, charts, or graphics.
6. Use headings to make the important points stand out.
7. Add a short narrative where appropriate. - What many of us don't realize is that by clicking the like button on social media we openly share information about ourselves that can then be used to predict other more personal attributes that we would never dream of sharing so openly.
- The real value is not in the large volumes of data but in what we can now do with it because of new technology.
- When predicting personality traits and attributes permission is "no longer needed" as we can ascertain personality traits and predict behavior based on publicly available data without anyone ever knowing about it.
- A study conducted by researchers at Cambridge University and Microsoft Research Labs showed how the patterns of Facebook "Likes" can be used automatically to very accurately predict a range of highly sensitive personal attributes.
-
Types of data by generation (creation) method:
1. CREATED DATA - You ask people question and put a mechanism in place to capture their answers. For example, market research surveys, focus groups, loyalty programs...
2. PROVOKED DATA - It wouldn't exist unless you invited people to express their views. For example customers rate or reviews.
3. TRANSACTION DATA - Is generated every time a customer buys something.
4. COMPILED DATA - it comes from giant databases from different sources often using name and address as the common identifier
5. EXPERIMENTAL DATA - This is a hybrid of created and transacted data. It involves designing experiments in which different customers receive different marketing treatments.
6. CAPTURED DATA - Data that is gathered passively from an individual's behavior. Such as search terms or location data from your phone. Mostly this is data most people are unaware is taken about them.
7. USER-GENERATED DATA -Data that is consciously generated by users, for example, social media posts,...
