Predictive analytics: Forecast what customers will do next.  

Huge piles of data are worthless without analysis. But you lack the army of data scientists required to do it. Predictive analytics will help you to transform vast data into actionable insights that solve people’s problems. Recognise and respond to patterns, identify and address anomalies. Find your most-valuable customers and learn what makes them tick with predictive analytics tools. 

What is predictive analytics?

Predictive analytics analyses pre-existing data to work out what might happen next. Statisticians have used it for decades, but businesses are increasingly adopting predictive analytics to predict what customers may do in the future. By analysing historical data, algorithms can gain insights into potential future behaviours - whether someone is likely to convert, for example.

Predictive analytics techniques use technologies such as…

  • Predictive modelling: Statistical technique that predicts future behaviour by analysing historical data and creating a model for predictions.
  • Data mining: Analysing large amounts of data to identify patterns and similarities using pattern recognition and mathematics.
  • Machine learning: Computer analyses data and uses insight to make predictions or suggestions. Uses technologies such as natural language processing.

Ready for predictive analytics? Adobe can help. Request a demo or arrange a call-back today

How predictive analytics works?

Predictive analytics analyses historical and current data to predict what might happen next. Mathematical calculations are used to identify patterns in the data and make predictions. Machine learning, statistics and data mining are among the technologies used to do this. This process is called predictive modelling.

Predictive analytics models:

  • Regression: Statistical analysis that compares variables to explore the relationship between them and identify patterns.
  • Decision trees: Family-tree style diagram that plots a pathway of decisions - highlights how different choices create different journeys.
  • Neural networks: Machine learning designed to mimic the human brain. It has to learn the information itself. Designed for complex tasks.

History of predictive analytics?

Predictive analytics has its roots in the mid-20th century when the first computers arrived. But it wasn’t until the 1970s that analytics technology began to be used in business. In the early 2000s pioneering statisticians developed analytics that enabled computers to make predictions based on the data it had been fed. This brings us to today, where huge advances in computing power and technologies such as AI have made machine-powered predictive modelling possible.

Statistics for predictive analytics

  • Predictive analytics global market projected to reach approximately $10.95 billion by 2022 - Zion 
  • Sophisticated analytics could increase earnings in global banking by $1 trillion - McKinsey.

Ready for predictive analytics? Adobe can help. Request a demo or arrange a call-back today

Benefits of predictive analytics

Predictive analytics bring many benefits, giving you a better view of your customers and their journey in super-fast time - without the need for in-house analysts and statisticians  army of data scientists. With the right predictive analytics tools, you can…

  • Uncover insight in your data: You’ve got all the data but no time - quickly assess huge amounts of data that may otherwise be wasted.
  • Analyse customer behaviour: See how people use your site to identify customer interactions both bad and good.
  • Identify high value customers: Learn which audiences to keep in touch to for the greatest chance of conversion.
  • Predict future behaviour: Learn whether your customer is likely to convert or drop out of the journey, with powerful predictive modelling.
  • Retarget customers fast: See where customers fall out of the journey with real-time visualisations showing customer behaviour.


“With machine learning, we can see beyond basic activities, monitor multiple channels and audience segments, reveal actionable insights buried in data and act instantly to help ensure that we’re delivering the best experiences possible.”

Roberto Gennaro  |  Chief Digital Officer,

Why is predictive analytics important?

Predictive analytics techniques are important because they can help you to get ahead of the competition. Technology has always been at the heart of customer experience - from barcode scanners for faster checkouts, to shopping apps that placed the high street in the palm of their hand. But today’s customer experience is all about the personalised digital experience: Customers want every interaction to show you understand them - and brands are responding:

1 in 4 larger businesses identify data-driven marketing that focuses on the individual as their most exciting opportunity.


of large businesses say making better use of data for audience segmentation and targeting is a top priority.

Predictive analytics is central to understanding and delivering that. It can help you to target the right customer with the right message at the right time

“Digital analytics has evolved from static, rear-view mirror reporting to becoming the brain behind truly orchestrated one-to-one marketing at scale, through prescriptive and predictive analytics.”

2019 Digital Trends report  |  Econsultancy in collaboration with Adobe

Ready for predictive analytics? Adobe can help. Request a demo or arrange a call-back today.

How predictive analytics helps businesses

Predictive analytics may be pioneering technology, but it is helping traditional business industries to provide better services for their clients and customers.

Predictive analytics in HR

Data is crucial to HR. Traditionally, data analysis in HR has been rooted in the past - things that have already happened. But predictive analytics can help people make HR-related decisions for tomorrow by predicting potential outcomes of those decisions. Dominic Hammond, people analytics and insights leader at PwC, told HR that predictive analytics could help to reduce risk around hiring new staff.

Predictive analytics in marketing

Predictive analytics is an established part of many brands’ marketing strategy. In marketing, it is used to make predictions about customer behaviour - such as whether someone is likely to make a sale. Closely linked to customer experience, predictive analytics helps brands to meet the needs of their customers by predicting this need and serving them the right content.

Predictive analytics in insurance

Insurance has always been rooted in data. Now, that historical data is helping insurers to predict what a customer might do next when making a claim - and crucially helping to establish whether that claim is genuine or fraudulent. Sophisticated data analytics is helping insurers move away from crude rating factors, suggests Craig Skinner, Insurance Data & Analytics Leader, PwC.

Predictive analytics in retail

Today’s customers expect “authentic, compelling experiences” says Nate Smith at Adobe. Analytics is key to better understanding your customers, says Nate - “how they behave and how to best make product and merchandising decisions”. Predictive analytics can help retailers to improve their customer experience, make more informed decisions around stock inventories and much more.

Predictive analytics in education

Universities can introduce better-informed decision making by using predictive analytics. Empowered by data, experts in higher education can predict things like the applicants that will be successful at their institution and those that won’t . Analytics can also help forecast student numbers in the coming years and more.

Predictive analytics in healthcare

At a time when technology is transforming healthcare, NHS Digital talks about using “cutting-edge analytics” to tackle healthcare challenges. Predictive analytics is among the technologies being used by healthcare providers to work out the likelihood of, for example, a patient returning to hospital . Meanwhile a recent Deloitte Insights report says the healthcare sector can be a “key beneficiary of predictive analytics”.

Predictive analytics in banking

Banking is among the industries best placed to benefit from analytics, according to McKinsey . Though many banks use traditional analytics, early-adopters of machine-learning powered predictive analytics can get ahead of the pack. More sophisticated analytics could increase earnings in global banking by $1 trillion - McKinsey.

Ready for predictive analytics? Adobe can help. Request a demo or arrange a call-back today 

Predictive analytics strategy

You need to have the right strategy - and predictive analytics techniques - to understand what you are doing and why, in the right order. At Adobe, the analytics strategy is broken down into four key areas.

Four stages of predictive analytics strategy

Collect and measure

  • Gather the data: Collect your data sources - from online transactions to point-of-sale. Sift using statistical modelling and data mining.
  • Observe user journeys: See how people use your site in real-time with live data feeds, so you can respond quickly.

Predictive analytics tools include: Visualisations, predictive analytics machine learning

Explore and understand

  • Analyse your data: You’ve got lots of data - historical and live, structured and unstructured. It’s too much to analyse manually. Machine learning makes it easy.
  • Recognise patterns: This data analysis will throw up patterns and trends you can learn from, using both historical and real-time data.

Predictive analytics tools include: Advanced segmentation, propensity scoring, predictive modelling, predictive analytics machine learning

Predict and model

  • Predict what’s next: From these patterns you can learn more about the people using your site - eventually predicting what they may do next.
  • Identify anomalies: Notice and learn from changes in predictions - for example, when far fewer people sign up to your newsletter from a specific page.

Predictive analytics tools include: Anomaly detection, intelligent alerts, contribution analysis, predictive analytics machine learning

Share and act

  • Create high value audiences: Engagement with personalised content reveals your high-value customers to create distinct segments.
  • Personalise content: These insights enable you to deliver the right content, to the right person, at the right time.

Predictive analytics tools include: Advanced segmentation, propensity scoring, predictive modelling, predictive analytics machine learning 

Predictive analytics tools

To measure the effectiveness of your strategy across these four key stages, you need the right predictive analytics tools. Powered by technologies such as machine learning, data mining and predictive modelling, predictive analytics tools include:



See how people use your website in real-time. Live data feeds form interactive graphics that help you to identify what’s happening and how to react. Visualisations can be helpful for retargeting.


Advanced segmentation

Go beyond traditional audience segmentation. Segment IQ breaks down your various audiences to pull out the key behavioural differences and similarities. Such learnings help refine your audiences to increase the likelihood of conversion.


Propensity scoring

Identify high value customers by scoring people on their likelihood to perform a specific action, such as convert or open an email. Learn which customer behaviours are most likely to lead to conversion and personalise content accordingly.


Predictive modelling

With customers’ next steps forecast, machine learning plots the best route to conversion based on the devices people are using and where, among other factors.


Anomaly detection

Things go wrong - find out when they do, quickly. Powered by AI, anomaly detection combs huge volumes of data to identify problems without having to spend lots of time doing it manually. See when anomalies occur.


Intelligent alerts

You’ve got a lot of data - too much to watch 24/7. Get notifications when something noteworthy happens - for example, smashing your traffic targets or suffering a drop in your newsletter sign ups.


Contribution analysis

Get to the bottom of issues without manually trawling through vast datasets. Learn why people are dropping out of the customer journey by identifying and analysing the contributory factors.

Ready to reboot your customer journey? Adobe can help. Request a demo or arrange a call-back today 

Adobe case studies - Predictive analytics examples

Travel brand Hostelworld used analytics to learn how people were using its site. It wanted to understand what its young travellers wanted - the adventures, locations and experiences - to deliver personalised experiences for every customer. 

We want to be with our customers “every step of the way”, says Otto Rosenberger, CMO at Hostelworld Group. “Whether they are looking to book accommodation, find fun activities or connect with fellow travellers.”

The results for Hostelworld with Adobe

  • One billion emails every year with high click-through rates
  • 61% of bookings coming from repeat Hostelworld customers


“Adobe Analytics is the foundation of our digital marketing strategy. It helps us learn more about our customers so that we can build more personalisation to encourage greater community interaction.”

— Otto Rosenberger  |  CMO at Hostelworld Group

Hostelworld and Adobe - learn more


What is the definition of predictive analytics?

There isn’t a set definition of predictive analytics. But it’s summed up as data analysis that predicts what your customer might do next. That may relate to their likelihood to convert, open an email, download an e-book or more. It uses technologies such as data mining, machine learning and statistical modelling to forecast the future.

How does predictive analytics help your business?

Predictive analytics can benefit your business by giving you a more complete view of how people use your site and what they may do next. Using machine learning to analyse vast datasets, predictive analytics helps you to identify your most valuable customers and what makes them most likely to convert. Empowered by this knowledge, you can deliver personalised content.

Predictive analytics and data mining - how does it work

Data mining is one of the main technologies used by predictive analytics. Data mining uses maths and pattern recognition to analyse vast datasets. In predictive analytics, data mining is used to predict what may happen next based on what has happened historically.

Related Content

Get out of your email marketing rut.

How to Create Emails Your Customers Actually Want to Open.

Let's talk about what Adobe Analytics can do for your business.