36. The difference between valuing data and valuing software

When businesses begin to think seriously about their intangible assets, a common source of confusion arises between software and data. Both are digital, both are central to modern operations, and neither appears on a traditional balance sheet in a way that reflects true worth. Yet the two are fundamentally different in nature, and applying the same valuation logic to each leads to significant errors in financial planning, deal-making, and strategic decision-making.

Software is a constructed asset. It is created through deliberate engineering effort, it has a defined function, and its value is closely tied to its utility over a predictable lifespan. Valuing software typically involves assessing development costs, licensing models, and the extent to which it supports revenue generation. Data, by contrast, is not created in the same purposeful way. It accumulates as a byproduct of activity — customer transactions, sensor readings, operational logs, and user behaviour. Its value is not intrinsic to the data itself but is contextual, depending on who holds it, what questions it can answer, and how it interacts with other datasets. A software product has diminishing returns as it ages and requires replacement; a well-managed dataset can grow in value over time as it becomes richer, more complete, and harder to replicate.

This distinction has practical consequences for how each asset is governed, protected, and presented to investors or acquirers. Software can often be rebuilt if needed — competitors can develop equivalent systems given sufficient time and budget. Data, particularly proprietary customer or operational data, is far more difficult to replicate and therefore carries a competitive advantage that software alone rarely provides. This exclusivity is a core driver of data value, and any meaningful valuation framework must account for it. Income-based methods that assess what revenue the data enables, market-based methods that compare it against similar datasets, and cost-based methods that estimate the expense of recreating it all apply differently to data than they do to software.

For companies preparing for investment, a merger, or simply better internal asset management, treating data and software as interchangeable digital assets is a costly mistake. Each demands its own valuation methodology, its own governance standards, and its own strategic roadmap. Understanding that difference is not a technical detail — it is a business-critical insight that shapes how value is recognised, protected, and communicated across the organisation.