By 2020 the amount of data saved and collected will be about 35 trillion gigabytes, reports World Economic Forum. Big Data is getting classified as a new asset.
Predictive Analytics technologies make sense of all the data. They show trends, patterns and help organizations understand habits. But, this is only the first step. Those who take the next step harness the predictions to actually influence the outcome of transactions, becoming the fearless leaders of their industries. That's the power of persuasion predictive analytics can foster.
Predictive Analytics World Founder Eric Siegel's book - Predictive Analytics: The Power To Know Who Will Click, Buy, Lie Or Die - inspired me to share few examples in the book of the power of turning predictions into actionable insight. These are simple examples for CFOs and finance executives who are eager to learn more about Big Data and Analytics. These technologies are being applied to marketing, sales,
Examples of predictive analytics applications
- Company: Netflix
- PA Application: Movie choice prediction
- What's predicted: Which movie will the customer would like to watch next
- What's done about it: Present personalized best choice movie options for viewers to increase customer satisfaction and retention
- Company: Amazon
- PA Application: Shopping behavior prediction
- What's predicted: What goods the customer likes to buy
- What's done about it: Present personalized catalog for buyers to enhance shopping experience and maximize revenue opportunity in each customer interaction
- Company: Target
- PA Application: Pregnancy prediction
- What's predicted: Which female customers will have a baby in coming months
- What's done about it: Market relevant offers for soon to be parents of a newborn
- Company: Hewlett Packard
- PA Application: FlightRisk — employee retention
- What's predicted: Which employees will quit
- What's done about it: Managers use the predictions to make specific human resources decisions with those they supervise.
- Company: Chase Bank
- PA Application: Mortgage
risk prediction - What's predicted: Which customers will prepay and terminate the relationship (refinance with another bank)
- What's done about it: Mortgages are valued accordingly to decide whether to sell them to other banks.
There are many examples of the rising usage of predictive analytics by companies all around us. Pandora, Spotify, Uber, Match.com are just a few of the other companies based on predictive analytics.
How finance departments can use predictive analytics
If you are not using predictive analytics, here's an idea for how your finance department can start.
- Company: Your company
- PA Application: Customer payment risk prediction
- What's predicted: Which customer will pay, delay, dispute or default
- What's can be done about it: Make accurate predictions of cash flow, personalize collections strategies to maximize cash flow, tailor credit limits to maximize revenue growth and mitigate risk
The power to know which customers pay, delay, dispute or default can help us make smarter micro-level decisions to manage each customer in a specific way and thereby also help drive accurate macro-level decisions such as - borrowing decisions for working capital
The science of predictive analytics applications
Predictive analytics applications driving operational decisions at a micro-transaction level will become the new normal. The cumulative effect of smarter transaction level decisions can greatly reduce risk and improve financial performance of organizations.
So, how do predictive analytics applications work? In a nutshell, these applications first build a predictive model based on an initial data set to give a predictive score for each input. The next step is making these predictive models self-learning, as the data grows the scoring models continuously learn to make predictions more and more accurately. The last step is to tie these predictive scores to specific operational actions and drive smarter micro-transaction level decisions and guide customers to the next transaction.