To say that today’s business landscape is unpredictable and rapidly changing would be an understatement. In the UK alone, 29% percent of businesses cite economic uncertainty as the most common concern. In this climate, success doesn’t just mean making the right decisions. It means making the right decisions fast and with confidence.
Accuracy and speed in data-driven decision making are exactly what predictive and prescriptive analytics enable.
Predictive or prescriptive analytics?
There are many similarities between these two forms of data analytics. Both methods use Machine Learning (ML) and Artificial Intelligence (AI) solutions to analyse vast amounts of data so leaders can make informed, data-backed decisions at light speed. There are some key differences, however.
Predictive analytics has been around longer, and describes analysis driven by big data to answer the question: “What might happen?” In other words, algorithms and AI tools are trained to make predictions based on existing data patterns.
Prescriptive analytics goes one step further, telling decision-makers not only what might happen, but advising them on what they should do. This additional step makes a big difference. It accelerates timely decision-making, which can make all the difference in a fast-paced and dynamic economy. That is why prescriptive analytics, in particular, has gained traction in recent years and is set for even wider adoption in 2025 and beyond.
Impact on the modern economy
Whether users realise it or not, advanced data analytics already impact their everyday lives in more ways than they imagine. For instance, the Harvard Business Review lists a number of applications, spanning from detecting malfunctions in manufacturing to behavioural targeting in marketing. Below are use cases that gained the most traction over the recent years.
Healthcare
Healthcare professionals have long used predictive analytics to forecast disease outbreaks. Today, pharmaceutical companies analyse historical data to predict the outcomes of drug development processes and clinical trials, helping them improve operational efficiency and bring drugs to market faster and at lower cost. For example, Pfizer has developed a predictive model for identifying patients with ATTR-CM (a rare heart condition) with 87% accuracy using data from electronic health records and medical claims.
Cybersecurity
In last year’s PwC Pulse survey, 75% of surveyed executives said cybersecurity poses a risk to their business. Hackers are continuously developing new techniques. In response, data scientists are using prescriptive analytics to actively preempt attacks. They leverage security logs, network traffic, threat intelligence feeds, and other data sources to identify behaviour patterns. They then look out for anomalies that could indicate a threat. Importantly, prescriptive analytics enables them to execute security measures proactively, putting them one step ahead of hackers.
Human Resources
Predictive analytics has been a mainstay in HR for over a decade. A textbook example profiled by the Wall Street Journal is Credit Suisse, whose HR analytics team analysed over 40 variables to identify 10-11 key parameters that accurately predicted flight risk, saving the bank approximately $70 million annually by enabling managers to engage and retain high-performing employees likely to leave. According to Gartner, predictive analytics can be effectively used not only to improve retention but also to better understand emerging trends in talent supply.
Retail and E-Commerce
Retailers and e-commerce marketplaces are using prescriptive analytics to personalise customer experiences. Amazon’s ML-powered recommendation engine not only predicts what users might buy next but also suggests discounts, delivery optimisations, and inventory adjustments to maximise conversions. This level of analytics has become crucial for customer retention and revenue growth.
“Handbrakes” on the data economy’s growth
These examples, however, are just the tip of the iceberg. Predictive and prescriptive analytics are transforming supply chain management, travel and transport, logistics, heavy industry, e-commerce marketing: the list goes on. The use of prescriptive analytics, in particular, is forecast to increase significantly. According to Adobe, 65% of senior executives see “leveraging AI and predictive analytics” as primary contributors to business growth in 2025.
Nevertheless, for this increase to happen, potential “handbrakes” on the successful use of data analytics need to be removed. One of the challenges is the human factor. Due to data illiteracy, many leaders still lack the knowledge and skills to interpret predictive and prescriptive data correctly.
Both access to data and the tools used to manage and analyse it need to be democratised for employees in different departments to be able to utilise organisational information independently. This can lead to more efficient resource allocation, reduction of underutilised “dark data,” and better decision-making through diverse perspectives.
Beyond people, there is another “handbrake” that is potentially even more important — a lack of quality data. According to Gartner, US organisations lose $12.9 million per year on average due to poor data quality. Another study by IBM estimated that $3.1 trillion of the United States’ GDP might be lost due to bad data annually, with 1 in 3 business leaders reporting that they don’t trust their own data. You don’t need a lot of expertise to realise that you will only get successful data-based decision-making if the data itself is reliable, timely, and relevant.
There are a number of data quality issues companies run into. For example, one common problem is depending solely on historical data for predictive and prescriptive analytics, which leaves companies addressing yesterday’s challenges rather than tomorrow’s.
Another common issue is internal data being siloed within departments. This can lead to an incomplete picture that produces skewed results and insights. More generally, while internal data can be very valuable (as Google’s HR example shows) for answering specific questions, businesses also need to incorporate external data if they are to really benefit from advanced data analytics.
Web Scraping: an invisible game-changer for data-driven insights
This brings us to web scraping — a method of accessing and collecting publicly available data in an automated way. By using APIs or other scraping solutions, businesses can collect unstructured data from various sources — e-commerce sites, news media, forums, travel engines, job boards, you name it — and parse it into analyzable datasets.
Web scraping allows organisations to collect granular data. This is incredibly valuable for both predictive and prescriptive analytics. Multifaceted external data can complement internal datasets and shed light on such factors as consumer sentiment, market trends, pricing fluctuations, and competitor strategies.
Web intelligence collection is an invisible industry that provides a competitive edge for different economic sectors that millions of people use daily. For instance, in real estate, accurate knowledge of how the market is developing is essential in order to accurately advise sellers and buyers. In travel and tourism, companies like Trivago are leveraging web scraping to give end users access to accurate, up-to-date information on deals and packages.
For cybersecurity professionals, web intelligence is a vital method providing valuable insights into emerging threats, vulnerabilities, and the tactics, techniques, and procedures (TTPs) used by cybercriminals. In e-commerce, players large and small use web scraping to dynamically adjust pricing and gather insights into customer interest and sentiment. In other words, unbeknownst to them, data scraping helps millions of users get better prices, better services, and better experiences every day.
The newest, AI and ML-powered scraping solutions make it possible to collect data in real-time. Syncing timely external signals with internal data analytics solutions is crucial for prescriptive analytics. When you are being presented not only with forecasts but with concrete plans of action, you need to know that following this course of action is the right response to conditions right now. What if pricing has fluctuated? What if new technologies have rendered previous approaches obsolete? Not factoring in the latest developments could be fatal in terms of decision-making.
Here’s a concrete example. Imagine you are shipping a time-sensitive, ready-to-use product like perishable food or ready-mixed concrete. Using external data on potential travel disruptions can be essential in optimising your logistics so that your product arrives in time and in perfect condition. Real-time data is essential for companies working with fast-moving consumer goods or perishables, where a constant balancing act is needed between managing stock levels and coordinating logistics to ensure fluctuating demand is always met.
Enabling data-driven decision-making
The bottom line is that predictive and prescriptive analytics are fast becoming indispensable. If a few years ago they were “nice-to-haves” that gave companies a competitive edge, today they are essential tools to keep up with rapidly changing market conditions and global competition. If your business has yet to fully embrace advanced data analytics, it is important to do so. And it is important to follow the right process.
That means first and foremost, defining your business objectives and main problem areas. Set crystal clear organisational goals that you want predictive and prescriptive analytics to help you with and drive value. Then, collect the data, combining internal datasets with external, alternative data, if needed. Remember that ensuring compliance with data privacy regulations is key.
Finally, your team of data scientists will need to ‘clean’ and organise this data before performing analytics that drive data-based insights. With this process set up, evaluate your results, then rinse and repeat — you will have to continuously optimise your data strategy. Use prescriptive analytics to iterate and improve decision-making over time.
