Fast and reliable application development and delivery is critical.This report exposes areas of breakage and disruption as many IT organizations continue to struggle to find cohesive approaches to application optimization.
Published Date: October 31, 2022
Prescriptive analytics is thought to be the most sophisticated type of data analytics, designed not just to provide data and insights but to suggest a course of action. Prescriptive analytics uses machine learning (ML) and mathematical algorithms, basing its output on past and current performance, available resources and likely scenarios to determine the best decisions for the organization.
It’s also another valuable tool in the big data kit. Organizations will increasingly need to find ways to take advantage of their data, especially as it grows to the zettabyte level. Prescriptive analytics can help even smaller organizations and businesses get the most out of their data and automate crucial business processes.
Some experts consider prescriptive analytics to be the logical outgrowth of predictive analytics, whereas some consider them to be separate and complementary disciplines. Regardless, prescriptive analytics is related to predictive analytics in the sense that they are both driven by machine learning algorithms, gather and synthesize large amounts of information and produce outcomes designed to aid in decision making across an enormous range of applications.
In other words, predictive analytics can tell you what might happen next, whereas prescriptive analytics will tell you what to do about it, with very high confidence. One challenge to this idea, however, is that prescriptive analytics can make recommendations about future action, but isn’t necessarily 100% certain if the choice is correct. In this article we’ll discuss what prescriptive analytics is, how it’s used for optimization and various use cases, how it differs from other types of analytics, the benefits and practice of prescriptive analytics and some suggestions for implementing it.
How is prescriptive analytics used?
The term prescriptive analytics describes a methodology and a way of using data and machine learning to arrive at the best course of action and make better decisions given a particular set of parameters. That being said, there are as many possible uses of prescriptive analytics as there are ways to use data in decision making. Prescriptive analytics can be embedded in a business intelligence program to guide its operation, or the results of a prescriptive analytics decision can be presented to human operators as data analysis in a dashboard.
Here are a few examples of prescriptive analytics.
Healthcare: Predictive analytics in a healthcare environment can tell the organization if there is a spike in emergency room visits, particular symptoms or other relevant factors. Prescriptive analytics takes the next step to suggest what can be done based on the data, including:
- Determining a patient’s risk factors based on existing conditions and prescribing a treatment plan.
- Identifying correlations among symptoms, comparing the efficacy of different treatments and recommending the best possible care and improved user experience.
- Predicting the likelihood of increased hospital and provider visits based on trend data and suggesting the necessary steps to prepare.
Banking and financial services: Predictive analytics is used in banking and financial services to forecast possible outcomes. With prescriptive analytics, more of the process can be automated to:
- Prevent credit card fraud by not only flagging but denying unusual transactions.
- Score credit for a loan or credit applications, and automatically route the loan for approval.
- Predict customer churn, allowing banks to reach out automatically if a customer is likely to switch institutions.
Sales and marketing: Prescriptive analytics can be used in sales and marketing to identify which customers and prospects are most likely to respond to particular offers and establish related business rules. Examples include:
- Deciding on offers and discounts based on customer characteristics.
- Scoring inbound marketing leads based on how likely the customer is to make a purchase and recommending the next outreach.
- Choosing the optimal mix of engagement and frequency of contact.
- Identifying coming trends and providing recommendations.
- Providing more personalized customer experiences based on individual location, buying patterns and other factors.
- Making more accurate predictions, allowing sales and marketing teams to create and achieve more accurate forecasts.
Security and IT operations: Prescriptive analytics has an especially valuable role in helping to preserve an organization’s cybersecurity as well as ensuring efficient IT operations. Prescriptive analytics can:
- Analyze network activity as well as external risk factors and act to prevent cyber attacks.
- Identify network components at risk of failure and schedule them for maintenance or repair.
- Automatically neutralize cybersecurity threats and perform root cause analysis.
- Perform routine and repetitive tasks, freeing human operators for more critical tasks.

Prescriptive analytics can be embedded in business intelligence programs in multiple industries.
Why is prescriptive analytics important?
According to a 2021 study by market intelligence firm IDC, global data creation and replication will experience a compound annual growth rate (CAGR) of 23% between 2020 and 2025. In 2020 alone, 64.2 zettabytes (ZB) of data was created or replicated — that’s four times the amount of data created in 2017.
The Internet of Things (IoT) is a key driver. In 2006, there were around 2 billion connected devices in the world, according to Intel. As of 2020, the number was estimated to be greater than 200 billion. Each one of those devices creates data that can be used to provide better customer service, optimize networks, target marketing messages more effectively, increase data security and dozens of additional uses.
The speed of computer processing combined with ever-increasing storage capacity means that the volume of data available for a particular task continues to grow, as well as the speed with which it can be accessed. But this growth and speed won’t matter without the ability to accurately search and filter the data and arrive at an outcome or business decision.
Prescriptive analysis:
- Bridges the gap between potential and recommended outcomes.
- Turns data into not only actionable insights but practical strategies.
- Cuts through the fog of data and presents a clear path forward.
What are prescriptive analytics techniques?
Prescriptive analytics, much like predictive analytics, relies on a combination of techniques and tools such as algorithms, machine learning and data modeling. These techniques are applied to the data sets selected by users and data scientists, including historical data, transactional data and real-time data feeds. Depending on the type of algorithm selected, the system will look for specific types of correlations among the data and then perform the desired function.
How is AI used in prescriptive analytics?
Artificial intelligence, especially AI in the form of machine learning (ML), is a key component of prescriptive analytics and could in fact be described as one of the key differences between prescriptive analytics and other forms of analytics. ML provides the predictive component to predictive analytics, taking large amounts of data and correlating it to understand, based on historical data, what is most likely to happen next. Understanding the likely outcomes is key to prescriptive analytics’ ability to suggest specific actions designed to achieve specific goals.
In practice, ML is applied to the data by using algorithms, the series of mathematical steps, like a recipe, executed to achieve a result or solution. Models define the way the algorithms are applied to solve a particular problem. The model is the framework that defines the questions, and the variables considered in answering them. The algorithms are the steps used to weigh variables and arrive at answers.
What is the difference between prescriptive analytics and predictive analytics?
Predictive analytics generally provides a range of potential outcomes based on the available data, whereas prescriptive analytics works by suggesting potential actions to take in order to achieve specific goals.
Specifically, predictive analytics works by applying mathematical models to large amounts of mined data to identify patterns in previous behavior and to predict future outcomes.
The combination of data mining, machine learning and statistical algorithms provides the “predictive” element, allowing predictive analytics tools to go beyond simple correlation. Predictive analytics is important because it allows businesses and organizations to make critical decisions based on actual data — predicting probable outcomes — at a scale that was previously impossible.
Prescriptive analytics goes beyond predictive analytics to suggest courses of action that can be taken to affect the outcome of data-driven predictions. In other words, prescriptive analytics can be used in decision making, planning and taking action based on prevailing data in order to achieve a specific goal or desired effect.
For example, in a cybersecurity context, a solution based on predictive analytics could be used to analyze network traffic, identify anomalous behavior and send an alert when that behavior matches the pattern of a specific threat. With prescriptive analytics, the software would not only identify a potential threat but would suggest actions to shut it down.

Predictive analytics provides a range of potential outcomes based on the available data, whereas prescriptive analytics works by suggesting potential actions to take in order to achieve specific goals.
What is the difference between descriptive analytics and prescriptive analytics?
Descriptive analytics is basic statistical analysis that uses raw data to provide a summary of historical behavior, describing what happened when. It is a static representation of a point in time, presented in a data visualization format, like a spreadsheet or dashboard. Descriptive analytics tells you what happened without any additional predictive (what will happen next) or prescriptive (how can we affect the outcome) elements. Descriptive analytics is valuable for providing a description of an event that has occurred or is still occurring and the data associated with that event.
What are the benefits of prescriptive analytics?
Every business and organization generates data as part of its day-to-day operations. Prescriptive analytics goes several steps beyond business analytics, providing decision makers with potential recommended actions to take to achieve the desired results.
Create plans based on real data: A prescriptive analytics model relies upon an organization’s historical data to understand and predict what will happen next, extending the value of business analytics that allow decision makers to create plans that are more likely to succeed.
See real-time and forecast data side-by-side: Prescriptive analytics allows decision makers to see not only historical data but also the predicted business outcomes of potential paths, giving them increased intelligence and more context for both short- and long-term operational decisions.
Reduced potential for human error: Human beings will always be necessary in organizational decision making, but by allowing machines to parse the data, understand correlations and present potential paths, the possibility for simple human error — as well as human bias — is greatly reduced.
Free people from repetitive tasks: A lot of data-driven activity requires human beings to perform repetitive tasks, impacting job performance as well as job satisfaction. A prescriptive analytics practice can free human beings to do more important and more satisfying tasks.
What should you look for in a prescriptive analytics tool?
The tools that are best suited to getting started with prescriptive analytics use artificial intelligence and machine learning and other advanced analytics tools to build predictive models. They require the ability to ingest new data at scale and “clean” the data, i.e. make sure it conforms to a particular configuration. Prescriptive activities require a significant amount of processing power and are built on prescriptive data science frameworks. The prescriptive tool also requires the ability, using machine learning, to train itself based on the outcomes of its activities. Designing the training algorithms requires human operators with deep subject matter expertise in order to determine best practices and what actions the prescriptive solution should perform based on the outcome of its predictions.
How do you implement prescriptive analytics?
Getting started with prescriptive analytics requires that you either have in-house data analytics expertise, or are working with an outside vendor who can bring that expertise to you. No matter how you go about it, you’ll likely face a significant amount of discovery work to determine your desired goals, as well as customization and programming to build the fundamental elements of the system.
Having a powerful tool that can run AI/ML models is the first step. The second is having the available staff and expertise to interpret the outcomes and recommend the right course of action. The third step includes applying the right tools that can allow you to combine and recommend or automate responses ( e.g. orchestration and automation, SOAR).
Ultimately, the best way to implement prescriptive analytics is to follow the same path you would use to evaluate, plan and implement any major software development or operational change in your organization.
- Understand what you can and can’t solve: Predictive analytics has multiple benefits, but it has limitations and it can’t replace the skills, judgment and experience of human professionals. Predictive analytics only works when there is enough data to provide useful output.
- Define the most critical problems to solve: You won’t achieve a usable outcome unless you know exactly what problems you are trying to solve. While it may be possible to apply predictive analytics indiscriminately to large datasets and hope to identify problems in the output, it’s far more effective to define the problem in the most precise way possible.
- Identify gaps in skills and technology: Software solutions make the practice of predictive analytics easier, but they still require expertise to use them. It’s critically important to have the people, infrastructure and tools necessary to identify and prepare the data you’ll need in your analysis.
- Conduct a pilot project: Pick a problem that you know other people agree is important. Determine the outcome you would like to achieve, and the metrics you will use to prove it. Do you want to reduce process time? By how much? Will you measure it in seconds, or as a percentage? Do you have the baseline data you need? Your pilot will be much more effective if you can state the outcome quickly and quantify the value. (i.e., “We reduced process time by 32%, which resulted in an average savings of 18 hours per week per employee” is more effective than “We optimized our process time significantly.”)
What is the future of prescriptive analytics?
So far prescriptive analytics is not as widely used as predictive analytics, but analyst firms believe that will change. With more data available, as well as more computing power and more advanced machine learning capabilities, the promise of prescriptive analytics is vast. Predictive analytics has established itself as a trustworthy discipline for understanding what might or will happen under a specific set of circumstances. It’s logical that prescriptive analytics will follow a similar trajectory. The ability to predict likely outcomes and automatically perform the appropriate function given both historical and immediate parameters could power significant advances in autonomous systems in fields such as healthcare, finance and cybersecurity.
The data explosion is real. While it can sometimes sound like a significant challenge, the most forward-looking organizations see it as an opportunity. Nearly every problem in business, information technology, information security and most areas of business operation is, in one way or another, a data problem.
Human beings cannot possibly manage this flow of information, making data management and analytics key capabilities for any organization. One of the chief complaints about the explosion of data is that it’s hard to know what to do with the data once you wrangle it.
Prescriptive analytics is designed to solve precisely that problem, allowing every organization to better use their data, providing understanding into the actions they can take based on related insights. As the volume, velocity and variety of data continues to grow, the need for and value of prescriptive analytics will grow along with it. Prescriptive analytics is a discipline in its infancy, but with unlimited potential.

Splunk Predictions 2023: Leadership Trends and Emerging Technologies
Splunk’s leaders weigh in on the most important technology trends facing the enterprise.