The Average is Always Wrong: A real-world guide to putting data at the heart of your business
By Ian Shepherd
()
About this ebook
Buzzwords abound – ‘data science’, ‘machine learning’, ‘artificial intelligence’. But what does any of it really mean, and most importantly what does it mean for your business?
Long-established businesses in many industries find themselves competing with new entrants built entirely on data and analytics. This ground-breaking new book levels the playing field in dramatic fashion.
The Average is Always Wrong is a completely pragmatic and hands-on guide to harnessing data to transform your business for the better.
Experienced CEO and CMO Ian Shepherd takes you behind the jargon and puts together a powerful change programme anyone can enact in their business right now, to reap the rewards of simple but sophisticated uses of data.
Filled with practical examples and case studies, readers will come away with a powerful understanding of the real value of data and the analytical techniques that can drive profit growth.
Ian Shepherd
Ian Shepherd is a CEO and CMO who has held senior roles in a range of world-class consumer brands over the last 25 years including Sky, Vodafone, Game and Odeon. Ian has launched loyalty programmes, built new digital revenue streams for traditional retailers, and turned declining market share into stellar growth – all based on a keen practical understanding of the consumer and of the power of data and customer insight. Now a non-executive director and adviser to a range of retail and technology businesses, Ian lives in Oxford with his family.
Related to The Average is Always Wrong
Related ebooks
The Analytic Detective: Decipher Your Company’s Data Clues and Become Irreplaceable Rating: 0 out of 5 stars0 ratingsChoose Your Customer: How to Compete Against the Digital Giants and Thrive Rating: 0 out of 5 stars0 ratingsDigital (R)evolution: Strategies to Accelerate Business Transformation Rating: 0 out of 5 stars0 ratingsBig Data Analytics for Creative Marketers: Money Spinner Rating: 3 out of 5 stars3/5What Is Big Data Rating: 0 out of 5 stars0 ratingsThe Big Data-Driven Business: How to Use Big Data to Win Customers, Beat Competitors, and Boost Profits Rating: 0 out of 5 stars0 ratingsNumbersense: How to Use Big Data to Your Advantage Rating: 3 out of 5 stars3/5Analytics Stories: Using Data to Make Good Things Happen Rating: 0 out of 5 stars0 ratingsMonetizing Your Data: A Guide to Turning Data into Profit-Driving Strategies and Solutions Rating: 0 out of 5 stars0 ratingsThe Essential Book of Business and Life Quotations Rating: 0 out of 5 stars0 ratingsThe 24-Hour Customer: New Rules for Winning in a Time-Starved, Always-Connected Economy Rating: 5 out of 5 stars5/5Profit Driven Business Analytics: A Practitioner's Guide to Transforming Big Data into Added Value Rating: 0 out of 5 stars0 ratingsThe Technologized Investor: Innovation through Reorientation Rating: 0 out of 5 stars0 ratingsPredictive Analytics: The Power to Predict Who Will Click, Buy, Lie, or Die Rating: 4 out of 5 stars4/5The Data Driven Leader: A Powerful Approach to Delivering Measurable Business Impact Through People Analytics Rating: 0 out of 5 stars0 ratingsChasing Digital: A Playbook for the New Economy Rating: 0 out of 5 stars0 ratingsThe Quest for Prosperity: How Developing Economies Can Take Off - Updated Edition Rating: 5 out of 5 stars5/5Big Data: Does Size Matter? Rating: 4 out of 5 stars4/5Analytics: The Agile Way Rating: 5 out of 5 stars5/5Leading with IT: Lessons from Singapore's First CIO Rating: 0 out of 5 stars0 ratingsThe Billion Dollar Byte: Turn Big Data into Good Profits, the Datapreneur Way Rating: 0 out of 5 stars0 ratingsCompliance to Commercial: The QUIET approach to Finance Business Partnering Rating: 0 out of 5 stars0 ratingsThe Algorithmic Leader: How to Be Smart When Machines Are Smarter Than You Rating: 3 out of 5 stars3/5Big Data, Data Mining, and Machine Learning: Value Creation for Business Leaders and Practitioners Rating: 3 out of 5 stars3/5Market share analysis A Clear and Concise Reference Rating: 3 out of 5 stars3/5Becoming Data Literate: Building a great business, culture and leadership through data and analytics Rating: 0 out of 5 stars0 ratingsBusiness Value in an Ocean of Data: Data Mining from a User Perspective Rating: 0 out of 5 stars0 ratingsQualitative Analysis Financial A Complete Guide - 2019 Edition Rating: 0 out of 5 stars0 ratingsCredit Risk Analytics: Measurement Techniques, Applications, and Examples in SAS Rating: 0 out of 5 stars0 ratingsLay an Egg and Make Chicken Soup: The Holistic Innovation Process from Concept to Market Expansion Rating: 0 out of 5 stars0 ratings
Reviews for The Average is Always Wrong
0 ratings0 reviews
Book preview
The Average is Always Wrong - Ian Shepherd
The Average Is Always Wrong
A real-world guide to putting data at the heart of your business
Ian Shepherd
Contents
Acknowledgements
Introduction
The data debate
Building a data-led business
Using this book
Why you need this book
Part One: Data analysis that drives profit
Chapter 1: A few useful concepts
The language of data
Why the average is always wrong
Getting behind the average
When statistics mean something, and when they don’t
The business significance of significance
Conclusion
Chapter 2: The mistake we’ve all made
Conclusion
Chapter 3: Getting into the detail
The dangerous trap of the straight line
What trap did we just fall into?
An alternative to the straight line – introducing clusters
Back to business
Chapter 4: Not everyone is the same
The characteristics of a segment
Segments are not always the answer
Stuck in the middle
What do you know about your customers?
The value of segmentation today
Conclusion
Chapter 5: The science of prediction
Prediction in action – a bad-debt model
The tree algorithm in action
Prediction – art and science
Choosing a model – accuracy vs clarity
Deciding what model to build
Some important notes of caution
Cost-benefit analysis and the lift curve
Conclusion
Chapter 6: Umbrellas don’t make it rain
Correlation and causality unpicked
Spotting the difference between correlation and causation
The illusion of the survivor
The confidence delusion
Conclusion
Chapter 7: What are the chances of that?
The business of probability
How independent are your data points?
Overlapping variables in a real business scenario
Bias, Bayes and targeted marketing
Conclusion
Chapter 8: Data science in the real world
The pull exercise – what questions do you have?
The push exercise – what data do you have?
A watchout
The characteristics of useful data
Data, databases and that cost-benefit challenge
Conclusion
Part Two: Valuable data and where to find it
Chapter 9: Start with the customer
The untapped power of investing in retention
Who is your most valuable customer?
Data about who customers are
Data about what customers do
Building your data model
The customer-data audit
Introducing customer lifetime value
Conclusion
Chapter 10: To loyalty, and beyond
The power of being small
The pureplay advantage
The loyalty illusion
The real loyalty reward
The right question and the wrong question to ask about your loyalty scheme
The cost of loyalty
And the benefits of loyalty
Loyalty cards are not the only answer
Conclusion
Chapter 11: Stock, stores and business performance
Data unlocks profit in every corner of a business
The power of negative space
Conclusion
Chapter 12: The outside-in view
The power of why
Insight and analytics
Beware of Goodhart
Why measure customer satisfaction at all?
Survivor bias revisited
The power of market share
Conclusion
Part Three: Building the data-centric business
Chapter 13: Culture clash and the comfortable silo
The corrosive power of fear
Denial and the lure of ‘what’s always worked’
The comforting silo and the email factory
Six routes through the denial stage
The role of desktop data tools
Conclusion
Chapter 14: The central process of the data-centric business
PPDAC
Getting the question right
Cost/benefit and PPDAC
The critical interplay between analysis and insight
Conclusion
Chapter 15: Make vs buy
Hiring directly – the make decision
Outsourcing – the buy decision
Beyond make and buy – the organisational-design question
Four design principles for the data-centric business
Conclusion
Chapter 16: The joy of change
1. Becoming a test and learn business
2. Long- and short-term planning
3. Change can be harder to land than we think
4. People
5. The power of listening to customers
6. To really change a business, you need to be willing to sacrifice things that make you money
Conclusion
Conclusion
Publishing details
Praise for The Average is Always Wrong
This book is about how to survive or perish. It is as stark as that. Ian’s book is brimming with examples of how leaders can not only survive – change today and they can thrive
.
– Matt Truman, CEO and Co-Founder, True
Ian has taken a complex topic and demystified it for leaders across the organisation – an essential read.
– Robert Kent, Chief Data Officer, Pets at Home
Required reading for all consumer business leaders who want to understand how to gather and use data to their competitive advantage.
– Jon Florsheim, Senior Advisor, AMP Capital
Business is awash with data; yet most companies haven’t a clue how to use it. This book will ensure you do.
– Danny Russell, Customer Insight Consultant
For Bridget
Acknowledgements
No business book springs fully formed from the mind of its author without the input, guidance and support of a lot of people. I’ve been fortunate over the years to work with, and for, some first-rate leaders and pioneers of data-centric business, and I am profoundly grateful for the opportunity I’ve had to observe them in action and learn from them.
More specifically, while writing The Average is Always Wrong I took the opportunity to share ideas, gather case studies and get hugely valuable input from some absolutely class-leading experts in data, analytics and customer insight. For their support, enthusiasm and wise counsel, I will be forever indebted to Clare Iles, Jon Rudoe, Danny Russell and Steve Delo. In particular, I’m grateful to Steve for proofreading much of the early draft and providing incredibly helpful comments. Of course, if you find errors in this book, they remain mine, but if you find clever insights, I probably pinched them from one of these four terrific experts. I am also tremendously grateful to the terrific team at Harriman House for their support in pulling this book together. From initial editorial conversations right through the editing and publication processes, they have been a delight to work with.
Finally, the structure of this book and the arguments it puts forward have gained hugely from forensic and insightful input from my wife, Bridget, a start-up founder and no mean data analytics expert herself. I am forever in her debt.
Introduction
Hang on, why is the average always wrong?
In a marketing presentation you might hear that ‘the average customer visits us 2.3 times a month’. That’s a comfortingly precise number, and it is easy to move on to a discussion about how you might increase it to 2.4 times a month without giving too much thought to the underlying data.
But what is an average anyway? What does the figure mean, how should we interpret what an ‘average customer’ really is, and what might this simple overall figure be hiding?
Consider two scenarios:
In the first scenario, most customers really do come between two and three times a month. Your business is an occasional visit for almost all your customers. By diving into research and customer feedback about how people use your products, you can chase that increase from 2.3 to a higher figure.
But in the second scenario, it turns out that 10% of your customers visit you 20 times a month, and the remaining 90% only visit once every three months. That generates exactly the same 2.3 overall average frequency, but obviously reveals a very different story. In this reality, you’ll want to know a lot more about why some customers visit so frequently – are they different to the majority? Or is it that the customers who visit your business so rarely are actually equally avid buyers, but from your competitors?
Getting a bit further under the skin of those simple overall figures is of massive value to business leaders. It is not the average of a set of data which is interesting, but instead the variances and discrepancies to that summary figure. When someone tells you that your average customer spends £100 with you each year, shops three times each year and has been a customer for 2.5 years, they are giving you only a tiny part of the story.
In reality, almost none of your customers behave like that. What’s interesting are the differences between customers. The great power of modern data science is the ability to use complex algorithms to understand the real richness of data and what it tells you, rather than just being satisfied with simple summary statistics.
The data debate
Underneath all those averages in presentations is the underlying data. Data is a topic that consistently dominates discussion of the fortunes of consumer businesses around the world. How much of it does a business have? How good is its loyalty programme and what benefits does that programme deliver? Is the business deploying the latest cutting-edge machine learning and artificial intelligence to increase its profits? How customer-centred is the business, and how does it use ‘big data’ to achieve that?
There is a good reason for all that focus on data. The rise of the pure play online business, powered by the latest technology and unencumbered by costly stores or legacy investments, poses a huge challenge for established consumer businesses the world over. And, as if all the other advantages these new online businesses had weren’t enough, they are awash with data. Since it isn’t possible to buy from an online business without providing them with your email address and quite possibly your actual address too, an online business acquires a customer database almost by default. No surprise, then, that they have used that data advantage ruthlessly – building predictive models and generating insights that enable them to squeeze just a bit more spending out of each customer than their mortal competitors can.
Those of us in leadership roles in consumer businesses faced with questions from shareholders and analysts about how we use and extract value from customer (and other) data, could be forgiven for a bit of world-weary cynicism in our responses. After all, there is always something new we could be investing in, and some of those press releases about ‘artificial intelligence’ just sound a bit far-fetched, don’t they?
Data and analytics only get easier to dismiss for busy leadership teams because of the increasingly obvious and frequently discussed ‘dark side’ of analytics. With big data sets being used to influence (and even arguably rig) local and national elections, and with global internet giants gathering an ever more terrifying amount of information about us, surely data is a topic best avoided?
But consider these real-world examples of data and analytics in action:
The business which uses computer-driven analysis of the words in emails from customers to pre-emptively sort the urgent ones (‘my delivery hasn’t arrived’) from the less urgent ones to ensure that the time of the customer services team is best used.
The retailer that uses careful modelling of customer spending patterns to predict when a valuable customer has fallen out of love with the brand (or may defect to a rival) so that their communications to that customer can be tailored appropriately.
The retailer that is able to build a predictive sales model to make sure each individual store holds exactly the right amount of stock of a new product.
The cinema business that has implemented real-time dynamic pricing based on consumer demand for individual screenings of films, generating a significant uplift in yield-per-screening as a result.
Building a data-led business
All these examples, and the many more that we will explore throughout this book, are built on data. They represent real, practical and profitable consequences that flow from becoming a data-oriented and customer-oriented business. Far from being the preserve of Silicon Valley gurus with PhDs and supercomputers, a smart use of the data we have from around our businesses can make a big difference to our profitability and cash flow. And in a world where businesses of all kinds, online and off, are being challenged more and more by those data-centric new entrants, no business can afford to pass up that opportunity.
So, what does it take to turn a consumer business into a data-led one? That’s the big question facing many leadership teams, and it can be an imposing and uncomfortable one. Many business leaders in retail, for example, have grown up with a sharp focus on supplier relationships, buying and merchandising, but with relatively little real customer data flowing back into decision making. Generations of leaders of those businesses have risen to the top based on their product knowledge, their negotiating ability, their operational and organisational skills – but have never before needed to consider how to mine and extract value from large amounts of consumer data.
For many management teams, the ‘data discussion’ can feel like a strange and alien one. As a result, however, some businesses shy away from investing in the area at all, while others, knowing they need to do something, end up hiring experts or consultants but effectively leave them on the side-lines, letting them ‘do their data stuff’ while the real business carries on as it always has.
Neither of those strategies are acceptable anymore, as the spectre of more and more competition looms in an era when consumers need better and better reasons to part with their money.
Using this book
It is time, then, for business leaders to embrace data and analytics and what those concepts mean for them and their teams.
That is the purpose of this book. My intention is not to turn the lay reader into a fully fledged data nerd, building neural networks before breakfast and running statistical significance tests with glee. That’s neither possible, nor necessary.
Instead, my objective is to give business leaders an insight into what is possible with data, to showcase some great examples of how analytical techniques can drive profits and, most importantly, to create a simple process for management teams to go through to make their businesses more customer- and data-centred.
As you go through the process of becoming a ‘data-centric business’, you will end up hiring or working with data scientists who can do amazing and hugely complicated things with large data sets. Reading this book will not give you the skills they have, but it will enable you to ask the questions and make the investments that can unlock the power of data. It will also equip you to take the hardest step towards becoming a data-centred business: the cultural change you will need to create right across your organisation to build a community of business leaders who are on that mission with you.
We are going to approach the challenge of building data into the heart of our business by breaking it into three steps:
Part one is about analysing data: what can we do with rich customer and business data when we have it? What does best practice look like in analysing data and turning it into profit? What do some of the buzzwords and ‘of the moment’ analytical techniques really mean? How can the business leader who is not well versed in data (and may not have thought about statistics since school!) equip themselves to lead a data-centred business effectively?
In part two we turn to gathering data: for many retail and hospitality businesses, customer data does not come as easily as it does to an online competitor, but it is essential to unlock the power of analysis we’ve explored in part one. What can we do to ensure we know as much as possible about our customers, our stores and the stock in our business, as well as ensure that that data is safe, accessible and useful? We’ll review some obvious and not-so-obvious sources of useful data that can transform your ability to run your business. We’ll even explore the data sources that can give you a real edge over those online new entrants.
Finally, in part three we really get down to business as we consider how to build a data-centric business. It’s one thing to have people generating clever insights, but how do we turn that into real profit? That requires teams all around the company to alter how they do business and can require a big culture change. We’ll adopt a pragmatic approach to putting data at the heart of everything the business does. Generating value from data also requires us to reconnect with what our data points really represent – actual customers, products and suppliers. Turning analysis and insight into profit requires us to take a big step from the theoretical world of numbers towards the practical world of the customer experience that we create in stores and online.
My intention is to go through each of these three steps in a way that is relevant and useful for the business leader in the real world. We’ll touch on technology, but not in a way that only your CTO could follow. We’ll even touch on the mathematics of data analysis, but in a way that is designed to leave no-one behind. We’ll take the fear and anxiety out of the topic of data and make it an integral and natural part of your business strategy.
As you read through the book, you’ll find specially marked-out sections which explain some of the terminology and concepts that are important for our journey. If memories of school maths classes bring you out in a rash, you can skip these sections first time through – the story the book tells makes sense without them. At a later reading, however, I’d encourage you to dip a toe in the