Analytics-Based Competition

What is analytics-based competition?

Advancement of computing capabilities, mostly powered by cloud-based systems along with huge improvements in data analytics and artificial intelligence, has brought a new game to many industries: data-driven competition.

Data analytics has been around for a while, but it is only now that the intersection of these powerful technologies enable affordable integrations that it can have meaningful impact on a company’s competitive capabilities.

And because of its relatively low penetration in some industries, business-driven analytics can still deliver an edge to those willing to explore and exploit its potential.

When we talk about data-driven competition, we refer to the use of data analytics to improve your earnings and increase profitability. As we explained in the book, there are basically two ways to retain a profitable position: differentiation and lower costs. Data analytics can help you with both.

For example, IBM is using data analytics to improve diagnostics in the healthcare space, helping some practices achieve superior accuracy over human-only diagnostic centers.

In one case, IBM’s Watson computer analyzed 20 million research papers on cancer and correctly diagnosed a 60-year-old woman with a rare form of leukemia.

The woman had been receiving other treatments without meaningful results, but Watson’s diagnostic pointed to the specific treatment she needed, helping doctors finally improve the patient’s health.

A more widely known case of using analytics to gain a market advantage is Alphabet’s web search engine Google, which uses thousands of parameters and signals to show the best search results based on the individual’s particular interests and other parameters like location and the quality of web pages.

That accuracy and personalization is one of the main reasons why Google became the dominant web search service, outperforming rival engines like AltaVista and Yahoo! who were the pioneers in the space.

On a different note, data analytics is also helping companies lower costs across multiple industries. One of the most notorious examples is how Walmart, a giant retailer that moves around $32 billion a year in inventory across 70 countries, has developed an advanced supply chain analysis platform which has now been extended to vendors and suppliers and is helping the company keep logistics costs down and retain its low-cost leadership position in retail.

And once again Google, through its machine learning company DeepMind, has developed powerful algorithms that together have reduced energy consumption in the company’s data centers by over 40 percent.

Data analytics can also be used to uncover customer preferences and understand user behavior to increase your sales and improve customer retention. Amazon.com for example uses data analytics and sophisticated algorithms to make personalized product recommendations to buyers based on a series of factors including the user’s purchase history, pages visited and products placed in the shopping cart.

In a similar application, Netflix’s recommender algorithm which presents their users with accurate recommendations based on their historic behavior and crowdsourced factors is just another example of how data analytics and machine learning can help make personalization (understanding customers better) part of a business’s business model.

In general, data analytics systems are divided into three types based on their functionality:

1. Descriptive systems: Evaluate data ex-post and produce insights about past behavior, providing relevant metrics and performance indicators that can be used to make decisions later. For example, Microsoft has saved millions of dollars a year in energy consumption by using software and sensors in a descriptive analytics application that helps them see where energy is being used, identifying opportunities for improvement.
2. Predictive systems: Use historical data to find patterns and predict future behavior. For example, many banking institutions are using predictive systems to measure the probability of delinquency and default on credit customers.
3. Prescriptive systems: Use historical data and sophisticated algorithms to recommend actions and behaviors. For example, UPS’s ORION platform evaluates the company’s deliveries to be made every day and provides drivers with the specific routes that they must follow, generating the company millions of dollars every year in fuel savings.

Data analytics is the new secret competitive weapon that many companies are using to find and seize profitable positions in their markets. There’s a lot of value that might be hidden in your data that is just sitting there, waiting to be created.

A good way to start thinking about potential data analytics applications for your organization is by using a table like the one shown below, where you can brainstorm with your team the different ways in which you could create a competitive edge.

All you need to do is to try to complete each of the empty spaces with applications for your company.

Goal/Application	Descriptive	Predictive	Prescriptive
Differentiate or increase sales	Banks provide users applications that provide a breakdown of their expenses so they can see where their money is going.	Netflix’s algorithm provides accurate recommendations to users.	IBM has been using analytics and machine learning to improve its M&A activity.
Lower costs or improved efficiency	Microsoft has been using data analytics to break down energy consumption by device and optimize costs in data centers.	Banking institutions have been using analytics to predict defaults and delinquency.	UPS’s ORION system provides turn-by-turn instructions to drivers saving millions in fuel costs.

Because these applications will need data, this exercise must be complemented with a data gap analysis, where you first identify the kind of data that is currently available, and then point at the additional data that would need to be collected in order to make those applications perform as expected.

An alternative reverse-engineering approach to identifying potential data analytics applications would be to identify the data that is currently being collected, then see what additional data could be collected, and then figure out which applications you could create with that amount of information.

Data analytics tools are designed to extend human capabilities to help make better decisions, and business strategy is not out of their scope.

References:

Wu, Sun. Strategy for Executives, this book can now be downloaded for free here.

Davenport, Thomas; Harris, Jeanne G. Competing on Analytics: Updated, with a New Introduction: The New Science of Winning. Harvard Business Review Press. Kindle Edition.