What Is Data, Really?

People talk about data as if it were obvious. Businesses claim they are becoming data-driven. Governments say policies must be based on data. Technology firms collect vast amounts of it. Investors describe it as the “new oil,” a phrase repeated so often that it has become part of business folklore. Yet despite the constant attention, many discussions skip a basic question: what actually counts as data?

To understand data as a business construct and system, it helps to strip away the hype. Data is not intelligence by itself, and it is not the same thing as knowledge. At its simplest, data is recorded information that can be stored, organised, and analysed. Something happens in the real world, and a record of that event is captured in a form that can be used later.

That record might describe a sale in a shop, a patient’s blood pressure reading, a parcel delivery scan, a football player’s sprint speed, or the time someone clicks a video online. These records may appear small and ordinary, but when collected systematically they become the raw material that modern organisations use to understand and coordinate their activities.

A helpful way to think about this is to separate three related ideas: data, information, and insight. Data is the raw record of something that occurred. Information is what happens when those records are organised into meaningful patterns. Insight appears when someone interprets those patterns and takes action.

Imagine a supermarket chain. The till records that 47 bottles of orange juice were sold on Tuesday in a particular branch. That number is data. If sales are grouped by product, day, and store location, the company now has information about demand patterns. If managers notice that orange juice sales consistently rise before school holidays and increase stock in advance, that becomes insight. Many organisations possess enormous quantities of data but struggle to move beyond the first stage.

Once we look at real businesses, the range of things that count as data becomes clear. A delivery company records the time each parcel leaves a depot, the route taken, and the moment it is scanned at the customer’s door. A hospital records diagnoses, prescriptions, test results, and recovery times. A streaming platform records what viewers start watching, when they pause, and whether they finish a film or abandon it after ten minutes. A city transport authority records passenger numbers, train arrival times, congestion points, and equipment failures. In each case, the organisation is creating structured records of activity so that the system can be monitored and improved.

One of the most important things to understand is that data is rarely found in pure form. It is constructed. Someone decides what to measure, how to measure it, and how frequently to record it. Someone defines categories and labels events. Someone chooses what counts as relevant and what can be ignored. Those decisions shape the dataset long before anyone begins analysing it.

Consider customer satisfaction. One company might measure it through surveys asking customers to rate their experience from one to five. Another might measure repeat purchases and product returns. A third might track social media complaints. All three claim to be measuring satisfaction, yet the data they produce will look completely different because the underlying definitions differ. This is why critical thinking matters when interpreting data. A dataset is not simply a mirror of reality; it is the result of design choices, incentives, and assumptions.

Businesses invest heavily in data systems because recorded information reduces uncertainty. A retailer uses sales data to forecast demand and plan inventory. An airline analyses booking patterns to adjust ticket prices. A bank monitors transaction data to detect fraud. A manufacturer collects machine data to predict when equipment might fail. Data does not guarantee good decisions, but it allows organisations to replace guesswork with structured evidence.

Because of this role, many firms now treat data as an asset. Unlike physical assets such as buildings or vehicles, data can often be reused repeatedly. A retailer’s transaction records might support stock forecasting, marketing analysis, supplier negotiations, and customer loyalty programmes at the same time. Yet not all data is valuable. Some datasets are incomplete, inconsistent, poorly defined, or impossible to act upon. A company can possess enormous volumes of data while still lacking clarity about its operations.

For data to become useful, several conditions must be met. The records must be accurate and reliable. They must be structured in ways that allow analysis. People must understand what the fields and categories actually mean. And the results must be linked to decisions that improve outcomes. Without these elements, data becomes little more than digital clutter.

Inside modern organisations, data often travels through its own supply chain. It is created or captured at one point, transferred through software systems, cleaned and standardised, stored in databases, analysed by specialists, and finally consumed by managers or algorithms. A customer clicking on an online store, for example, generates a record that flows through tracking systems, data platforms, analytical tools, and reporting dashboards. Each stage transforms the raw event into something more structured and usable.

Control over data can also shape power within markets. Companies that gather detailed records of customer behaviour often gain strategic advantages. A digital marketplace that sees buyer demand, seller prices, delivery performance, and refund rates has a level of visibility that individual sellers do not. That information can be used to refine pricing, promote certain products, or redesign the platform itself. Data therefore does not simply describe economic activity; it can influence how that activity unfolds.

It is equally important to recognise the limits of data. Recorded information can reveal patterns, but it does not automatically explain why those patterns exist. Correlation is often mistaken for causation. A dashboard may show that sales declined in one region, but it may not capture the human reasons behind the change. A hospital dataset may show treatment outcomes but cannot always capture the social factors influencing recovery. Numbers can create a powerful impression of precision even when important context is missing.

Some businesses have gone further by building entire products around data itself. Credit rating agencies convert financial histories into scores used by lenders. Market research firms sell consumer insights to manufacturers. Digital advertising platforms use behavioural data to target marketing campaigns. In these cases, data becomes part of the business model rather than just an internal tool.

The rise of digital technology has expanded the scale at which this happens. Sensors monitor machines in factories. Apps record user behaviour on smartphones. Logistics networks track vehicles and shipments across continents. Hospitals store medical records digitally. Governments collect large statistical datasets about populations. As more aspects of the world become measurable and storable, organisations increasingly depend on data to operate effectively.

Seen clearly, data is neither mysterious nor magical. It is a structured record of activity produced within a system. Something happens, it gets recorded, the record is organised, and the organisation uses it to guide action. That process can improve coordination, reduce uncertainty, and reveal patterns that would otherwise remain hidden.

But data also reflects the assumptions and incentives of those who create it. The categories chosen, the measurements taken, and the events recorded all shape the picture that emerges. Understanding data therefore requires both technical knowledge and critical thinking.

In modern business, data has become the operating language through which many organisations understand the world. Yet its value does not come from sheer quantity. It comes from clarity of definition, thoughtful interpretation, and the ability to turn records of the past into better decisions about the future.