In an economy where decisions regarding the cloud, AI, cyber security or the location of infrastructure can tie up capital for years, the quality of information becomes one of the factors limiting the cost of error.
Data is increasingly becoming an integral part of management infrastructure. In 2025, 33 per cent of the surveyed companies in the European Union carried out their own analytics, whilst among large firms the figure reached 78.8 per cent. 69 per cent of the largest companies used business intelligence tools. In Poland, 25.9 per cent of companies carried out data analytics, 6.6 percentage points more than two years earlier.
This is not solely due to the development of BI. Business decisions require an ever-increasing number of variables. Cost, demand and revenue are not enough when a company is selecting an infrastructure provider, assessing the market for a new investment, planning a technology migration or looking for a location for a data centre. Factors such as energy prices, connection options, network availability, the regulatory environment, demographics, the labour market, investment plans and environmental risks all come into play.
The more interdependencies there are, the less one can rely on intuition alone.
‘Public’ does not mean ‘easily accessible’
Much of the necessary information already exists. Government bodies, regulators, statistical offices, infrastructure operators and European institutions publish vast datasets. The problem is that formal availability is not the same as business availability.
Data may be in different formats, relate to different time periods, use different territorial units, or employ definitions that cannot be directly compared. Sometimes, it is only by combining several public datasets that information emerges which was not previously present in any of them.
The European Commission regards this issue as a key factor in the economy’s competitiveness. The Data Union Strategy, adopted at the end of 2025, aims to increase businesses’ access to high-quality data. The Data Labs currently being established are intended to combine public and private resources and to deal, amongst other things, with their curation, annotation and preparation for further use. The Commission describes European data spaces as a source of structured, reliable resources which, only after appropriate preparation, can be utilised in business and AI.
This is a significant change. Public data does not lose its value simply because, in theory, anyone can download it. The advantage may lie precisely in the fact that someone is able to find it more quickly, verify it, standardise it and combine it with other information.
Decision first, data second
However, the sheer volume of data does not determine the quality of the analysis. It only becomes meaningful in the context of the question it is intended to answer.
If a company is considering the location for a major technology investment, a comparison of land prices alone is of little help. It may also need to analyse the availability of energy, telecoms infrastructure, environmental risks, the labour market, planned network investments and potential constraints on development.
Each variable may come from a reliable source. However, it is only when they are combined that they answer the business question: where is the risk lowest and the potential greatest?
That is why good analytics should start not with the question ‘what can we show?’, but ‘what do we need to know before making this decision?’.
This distinction separates reporting from a decision-making tool.
Data quality becomes even more important with AI
The stakes are rising as analysis becomes automated. In 2025, 20 per cent of businesses in the EU were using AI technology, compared with 13.5 per cent the previous year. Models can scan a much larger number of variables, detect correlations and generate recommendations faster than traditional analytics.
However, they cannot turn poor-quality data into good data.
An unclear definition will remain unclear. Incomparable datasets will remain incomparable. Out-of-date information can only be used more quickly to produce a convincing but erroneous recommendation. It is no coincidence that the Commission, when describing the development of the European AI ecosystem, highlights the limited availability of reliable, high-quality data and the need to improve its interoperability and accessibility.
The value of analytics is therefore shifting across the entire chain: from source and definition, through structure and comparability, to the context of a specific decision.
A dashboard can simplify this process and allow decision-makers to quickly identify interdependencies. However, it need not be just another dashboard showing what happened in the previous quarter. It can answer far more valuable questions: where the potential lies, what distinguishes the best option from the next best, which factors influence the outcome, and where risks are mounting.
Data is a resource. Its business value becomes apparent when it enables a significant decision to be made with less uncertainty than competitors.
