Thinking with Data: How to Turn Information into Insights

by Max Shron

Many analysts are too concerned with tools and techniques for cleansing, modeling, and visualizing datasets and not concerned enough with asking the right questions. In this practical guide, data strategy consultant Max Shron shows you how to put the why before the how , through an often-overlooked set of analytical skills. Thinking with Data helps you learn techniques for turning data into knowledge you can use. You’ll learn a framework for defining your project, including the data you want to collect, and how you intend to approach, organize, and analyze the results. You’ll also learn patterns of reasoning that will help you unveil the real problem that needs to be solved.

Genres: Nonfiction, Business, Computer Science, Technical, Technology

91 pages, Paperback

First published January 20, 2014

Reviews

[Louis · Rating 4 out of 5]

Thinking with data focuses, not on how to do data analysis, but on the questions that one should be asking. It does so in two ways, first through providing an overall framework to looking at situations, then working through a series of topics using examples to serve as plausible paths of decision making. In a fairly short book, it covers the framework, determining purpose, threats to validity, experimental design, and a few extended examples that illustrates both concepts and deviations. It is a useful quick big picture book that is useful for those whose focus has been on the methods of data analysis or for those who do not have a quantitative background but are faced with data questions and need to be able to work with data analysts.

The first part is probably the most rewarding. Max gives a framework of how to frame a data problem. Context (who is interested in the problem, what are their overall goals and why, what is the goal of the project), Need (the specific need that could be solved through the use of the data model), Vision (an understanding of what the results of data analysis would be like), and Outcome (an understanding of how the data analysis results would be used). The end of this framework would be a story that you can tell

Next is a discussion of how the details of the problem could be fleshed out. The content is probably familiar to anyone who has had to work with stakeholders. The valuable portion here are the vignettes of working through this process on projects. In particular the fact that the vignettes are not projects that necessarily go smoothly, so it does not have the idealized feel that many published vignettes do.

Next is a discussion of presenting the results. The focus here is that the results are not the output of the data analysis, but the use of the data analytics methods to construct and argument. And that argument is going to be presented to people who have backgrounds, prior beliefs, prejudices, and sometimes reasons to argue against your findings.

How to address these disputes is through conducting experiments and testing alternative hypothesis. So a section of the book is on defining causality and designing experiments (interventions) to handle different types of alternative hypotheses.

What makes this useful is the framework and the vignettes. It is good for a quick introduction to this area. As others have noted, it is not tightly organized, so after the first chapter with the framework, it is not useful as a reference, but it helps in focusing how to think.

I teach classes on working with data, and one area that is difficult to get across is the concept that there is a unified whole in the topic, not only a bunch of separated techniques. I plan on using much of what is in this book to help provide that unified whole my classes.

Disclaimer: I received a free electronic version of this book as part of the OReilly Bloggers program.

[Bianca · Rating 4 out of 5]

Timely for me. It puts emphasis on asking the right questions and thinking through the project first before diving deep into data work. Straightforward and doesn't cover much technical stuff, which are actually good points for this book.

[Daniel Wright · Rating 1 out of 5]

'How to turn information into insights? Well that's easy! You just give me and my consultancy firm unfeasible amounts of cash and we'll tell you a whole load of things you probably could have guessed using just a dash of common sense anyway! It's great! If you don't believe me, I've written this insipid little book about it! Come on, some of it's not even about the relevant subject and shows an utterly superficial understanding of the humanities - you must be convinced now!'

I am rather cruel. This book is actually of some interest to anyone who is wondering what the current fad for 'data science' is all about. Not being much of a person for fads, I did not find it all that interesting.

I'm sure Shron's consultancy does actually provide worthwhile services as well. But since this book is in some ways his sales pitch, read it but don't pay for it - you can get it for free online (I did have a link to it, put it seems I'm not allowed - I suggest you use Google).

[Donn Lee · Rating 3 out of 5]

It's an "ok" book that needs a more coherent storyline. There's no theme that ties the elements together, and sometimes it feels as if you're reading a stream of consciousness (or encyclopedia). The sort of thing somebody might tell you over a lunch or in the hallway while walking to the bathroom.

There are some valuable nuggets in there, especially with how it relates to business understanding, so maybe this would be more geared toward those who are more technical and trying to get a grip on the business aspects of data science, but doesn't work particularly well the other way (business people trying to get a grip on the technical aspects).

[Carolina Bento · Rating 5 out of 5]

This is a must-read book for anyone that works with data on a daily basis!
The framework presented in this book is very straightforward and focuses on the essential topics that you should think about before starting any project. It can definitely help you find clarity about what's important in your projects.

[Yuriy · Rating 4 out of 5]

Excellent book to get structure with working on data science projects. Convo and scaffolding, techniques I would like to apply regularly.

[Shibaprasad Bhattacharya · Rating 4 out of 5]

A good book for novice Data Scientists/Analysts. It will guide you in asking the right questions that are pertinent to Data Science.

[Paul Laughlin · Rating 4 out of 5]

Thinking with Data helps you turn information into insights

One of the perennial challenges that I hear data leaders face is how to train their analysts how to turn information into insights. Too many underwhelm stakeholders by just presenting data, facts, results without relevant insights. The book I am reviewing today, entitled “Thinking with Data”, should help those leaders & their analysts.

I was delighted to receive this book as a kind gift from Dante (thanks again) following a call. I’m delighted that he shared it with me as there are too few focused on this topic. In “Thinking with Data” the author, Max Shron, has created a handy guide that could be used by almost any analyst. His practical experience from many years as a data strategy consultant shines through.

That said, if you buy this book you might be underwhelmed when it first arrives. Given the price point, you might have expected something longer. At only 75 pages, it looks like barely more than a pamphlet. But don’t be fooled. A surprising amount of value is packed in those few pages. Plus this means it’s quick for any time-poor leader or analyst to read.

What does Thinking with Data include?
Let’s start by giving you an overview of the content you can find in this book. It is arranged into 6 pithy chapters, followed by a shortlist of recommended books for further reading. These can be further grouped into these topics:

Scoping (understanding Why? before starting on How?)
What Next? (domain knowledge and verifying understanding of challenge)
Creating Arguments (turning observations into knowledge):
Types of arguments
Patterns of reasoning
Special case (Causual reasoning)
How to put all of this together (case studies/deep dives)
Hopefully, the language above helps you hear the practical focus of this book. It really is a short treasure trove in which any analyst will find at least one gold nugget. But beyond the practical methods & tips, there is theoretical underpinning. The author (Max Shron) does a great job of demonstrating the polymath nature of this craft. He shows that best practice draws on disciplines far beyond maths & stats. Including thinking models from English, Psychology, Political Science & many other humanities.

Scoping well, to turn information into insights
As Max highlights, many analysts have the bad habit of wanting to rush into action too soon. The temptation to start extracting data or using your analytics package/language can be too great. In the rush for progress, quality thinking at the outset can be missed. All too often this will come back to bite you. Ill-defined problems. Misunderstood needs. Unrealistic expectations. The source of most of those woes is a lack of time spent ‘scoping’ to understand both the real need and limits on what is to be done.

When I train analysts in the people skills they need to be effective, we cover Socratic Questioning to get to the implicit or underlying need. I really like how Max builds on that skill set with a simple mnemonic to guide such questioning. He uses this CoNVO model:

Co = Context (organisational priorities, interests, issues, how will this work further progress?)
N = Needs (needs that could be met by data, the Why?, key questions, action to be informed)
V = Vision (what will it look like to meet that need with data? envision impact & progress)
O = Outcome (what is to be produced, who will use it, what will change as a result?)
Such a focus can be very helpful. I also really support Max’s frequent call to write things down. Like the benefits of a Vision Script for leaders, writing down your understanding of each part in a short paragraph helps to clarify thinking. He couples this with a number of recommended forms to use. Mockups and Argument Sketches to capture a Vision. A clear written story to express all the above on one page.

Structuring how you will approach analytical work
Many analysts work in an ad-hoc manner. Some, who might be considered to represent best practice, use a consistent workflow or methodology (CRISP-DM etc). However, very few have a structured approach to thinking first. Planning out an appropriate critical thinking approach to suit this problem. That is some of the gold dust that Max shares in chapters 2-5.

He starts by guiding the analyst to dig deeper into understanding the problem domain. Sense checking your understanding of what is needed and data sources or stakeholders who can help you learn what is needed. Max shares how to use the following techniques at this stage:

Interviews (knowledge elicitation from experts)
Rapid Investigation (order of magnitude estimates, easy graphs, BI for sense checking etc)
Kitchen Sink Investigation (like a detective asking everything & train of thought interrogation)
Working Backward (start with intended output, work back step by step what’s needed)
Mockups (sketching ideas of outputs and processes to get there)
Roleplaying (lay the part of the final consumer & think out loud)
Lots of good ideas there, but Max goes quite a bit deeper by then exploring how you can construct arguments via your data analysis. These next chapters are a useful guide on taking approaches including: Confirming audience & prior beliefs; Identifying claims that need to be made; presenting evidence, justification of claims & predicting rebuttals.

Learning in practice and from practical examples
Analysts will best learn these skills & techniques in practice. But a helpful first step is case studies or walkthrough examples. Max peppers this short book with such examples. It is helpful to follow the metamorphosis of some of these ‘Deep Dive’ sections through each chapter. He also illustrates bringing everything together in the final chapter through two longer such case studies.

Before that final chapter, Max brings to life some helpful patterns of reasoning that should guide analysts through this process. Drawing from both the legal & scientific disciplines he walks us through identifying & classifying different points of dispute to address. This brings to life how an analyst will need to take slightly different approaches to address disputes about facts, definitions, values & policies. He also provides extra advice for special types of arguments (optimisation, bounding cases, cost/benefit analysis & proving causation).

It is a credit to such a short book that I was reminded of relevant principles from much more in-depth studies. You can see the application of lessons I learnt in critical thinking from “Calling Bullshit“. There are also some of the perspectives recommended in “How to make the world add up“. Plus the chapter on causality was a useful primer to my current reading, a much deeper exploration in “The Book of Why” by Judea Pearl (review coming next month).

How are you turning information into insights?

So, I hope you can tell that I’m a fan of this book & despite the higher price tag for a small book would recommend it for analytics teams. But even more important is that you practice these techniques in the real world. Then share your experience. Recognise that as well as honing your technical skills, analysts need to develop their thinking skills and ability to question & plan well from the start.

What about you, dear reader? What has worked for you? Have you found other resources or approaches that have helped you better generate relevant insights for your business? If so, please let me know. I am keen to share more to help all analysts develop these skills. True transformation of organisations to be data-led needs much more than the mainstream focus on technology & technical skills. So, expect more on this topic in future. Happy thinking & I hope you are seeking success in delivering insights that drive valuable changes.

[Ernestasia Siahaan · Rating 3 out of 5]

Data professionals (data analysts/scientists) are storytellers - we work out meaningful stories from data. I think this book captures that definition/purpose of data professionals. It describes the framework of thinking to work out a (business) meaningful story from data.

It starts out with describing a famework for nailing the “WHY” of a project before diving into the “HOW”. It then continues with examples of building arguments, and from there defining relationships between the points in our arguments. It mentions only slightly upon (statistical or technical) methods to work on those arguments and relationships.

Coming from a research background, I find a lot of the content covered in this book to be more of a quick refresher on structuring the way we approach a (data) project. If you want an overview of how to structure one’s approach to a data project, or you just want a qualitative description of a data project from beginning-to-end, this book is for you, and provides plenty examples to bring its message across.

[Faye Zheng · Rating 4 out of 5]

I use and teach these principles every day on my job. Essential reading for anyone who does analytic work. This book is not about data science, it is about the 95% of data science that should be spent on problem formulation, critical thinking, evidence-based arguments, and deep examinations of value and outcome. Read it in 4 hours and it contains no more information than necessary while providing a thorough, well-organized framework.

[Anna · Rating 2 out of 5]

To be fair, it is what it says it is: a book about data strategy and how to ask questions and form arguments. In attempting to move away from focusing on tools and technologies, this book goes rather far to the other extreme. I was looking for something closer to the middle ground between overly-focused on tools and overly-focused on “soft” skills. That said, I can see where this book has its value.

[Roberto · Rating 2 out of 5]

I am not sure why I didn’t like it. Maybe it is too oriented to a Data scientist reader. And because of that, even though the ideas were simple, the reading is complex.

This is the first book I write in, and I recommend this.

[ezequiel orbe · Rating 2 out of 5]

It's a "meeh" book..

[Princessadivana · Rating 4 out of 5]

Как ставить правильные вопросы при работе с данными, различные техники анализа + case studies

[Alex · Rating 3 out of 5]

Concise book full of useful (if often "obvious") information. Interestingly, this book was more geared towards advice that I found to be useful from a general consulting perspective, rather than a "data science" perspective per se. Adds some color to areas that were not covered in such depth in Provost & Fawcett's "Data Science for Business," which overall remains the comprehensively best book on the subject I've yet read. Shron's book is much quicker to get through, though, and does point towards some other interesting avenues of further reading which appear to be less in the technical vein of many other data science books.

[Greg · Rating 2 out of 5]

Too informal and not much new content here for anyone that has done basic project management. Feels like a short work meant to cash in on the data science/big data book frenzy. Data Science for Business, while much lengthier, is time better spent.

[Sean · Rating 4 out of 5]

Focuses on asking the right questions more than how to ask them. A bit abstract in spots, but overall useful.

[Ingus Rūķis · Rating 4 out of 5]

Loved the first chapter of the book, it applies not only to data science but to other types of projects as well.

[Melissa · Rating 4 out of 5]

Sound advice and great prose, but sometimes the organization is too opaque.

[Amber · Rating 3 out of 5]

some good concepts to think about but not a ton of novel information

[Jim Razinha · Rating 2 out of 5]

Well...that was a wasted hour and a half. Little value in this. Nothing new and no paradigm shifting of the old. Even though it was quite short, it could have been distilled into a tri-fold tract.

[Menaka Sankaralingam · Rating 2 out of 5]

Not a very useful book in my opinion.
For a complete review check http://plusminusnmore.rapo.in/thinking-data-max-shron/

[Azucena Coronel · Rating 1 out of 5]

Very basic, it is like a compilation of definitons and an attempt to apply those definitions to small use cases. Didn't found value

[Andrew · Rating 3 out of 5]

For whatever reason, I had a tough time making it through this short book. I did enjoy the section on causality, which you don't see very often in books like this