Data Assets and Technology Capital

Intangible Asset Masterclass — Lesson 6 of 10

Data has been called "the new oil" so frequently that the phrase has become a cliche. But the comparison is misleading. Oil is consumed when used. Data is not. Oil is fungible. Data is not. Oil's value decreases with extraction. Data's value often increases with accumulation and combination. The economics of data assets are fundamentally different from any physical resource — and understanding these economics is essential for identifying and valuing the technology capital that increasingly drives enterprise value.

This lesson examines four categories of technology capital: proprietary data and databases, algorithms and AI models, platform technology and network effects, and digital infrastructure. For each, we cover the identification criteria, protection strategies, valuation approaches, and the strategic implications for businesses that own them and investors who evaluate them.

★ Key Takeaway

Data assets and technology capital are the fastest-growing category of intangible value, yet they are among the hardest to identify and measure using traditional frameworks. A database's value depends not on how much data it contains, but on the uniqueness of that data, the insights it enables, and the cost a competitor would incur to replicate it. Companies that treat data as a strategic asset — with formal governance, quality standards, and monetisation strategies — create significantly more value than those that treat it as a byproduct of operations.

The Data Asset Landscape

181ZB global data volume projected by 2025 (IDC)
$274B global big data and analytics market (2023)
68% of enterprise data goes unused (Forrester)

The volume of data generated globally is staggering, but volume alone does not create intangible asset value. Forrester Research estimates that 68% of enterprise data goes unused — collected and stored but never analysed, monetised, or even accessed after creation. The intangible value of data depends entirely on whether it can be converted into economic benefits.

Proprietary Data and Databases

Under IFRS 3, databases are classified as technology-based intangible assets. They are identifiable because they can be separated from the entity and sold or licensed independently. But the value of a database varies enormously depending on four characteristics.

Data Asset Value Framework

Characteristic Higher Value Lower Value
Uniqueness Data that cannot be replicated or purchased from third parties Commodity data available from multiple sources
Completeness Comprehensive coverage of the target domain Partial, fragmented, or inconsistent records
Freshness Continuously updated with real-time or near-real-time data Static or infrequently updated
Actionability Directly enables revenue-generating decisions or products Raw data requiring extensive processing before use
✔ Example

Bloomberg Terminal's value rests on a proprietary database of financial data covering 35 million instruments, fed by 5,000 data providers, updated in real-time. This dataset cannot be replicated — the relationships with data providers, the cleaning and normalisation processes, and the 40 years of historical depth represent a formidable intangible asset. Bloomberg generates approximately $12 billion in annual revenue, with the Terminal's data asset being the foundational value driver. An acquirer would classify this as a technology-based identifiable intangible asset worth billions.

Customer Databases as Intangible Assets

Customer databases sit at the intersection of customer-related and technology-based intangible assets under IFRS 3. A customer database typically includes contact information, purchase history, behavioural data, preferences, and interaction records. Its value as an intangible asset depends on the database's size, quality, recency, and the revenue that can be attributed to the relationships it represents.

In practice, customer databases are often the most valuable data asset a business owns — particularly for businesses with recurring revenue models where the database represents the installed base from which future revenue will flow.

Algorithms and AI Models

Algorithms and AI models represent a rapidly growing category of technology capital. A trained machine learning model — encompassing the architecture, training data, learned weights, and fine-tuning — can be an identifiable intangible asset if it is separable or arises from contractual rights.

Algorithm as Process

  • Set of defined computational steps
  • Deterministic output for given input
  • Protectable as trade secret or patent
  • Value: efficiency gain or capability enablement
  • Examples: sorting algorithms, routing optimisation, pricing engines

AI Model as Asset

  • Trained on data; learned parameters
  • Probabilistic output; may vary
  • Protectable as trade secret (training data + architecture)
  • Value: prediction accuracy and unique training data
  • Examples: recommendation engines, fraud detection, image recognition

Valuing AI Models

AI model valuation is an emerging discipline with no established consensus methodology. In practice, valuers consider several factors.

Factor Description Impact on Value
Training data uniqueness Whether the training data can be independently assembled Higher if proprietary data creates a barrier to replication
Model performance Accuracy, precision, recall relative to alternatives Higher if measurably superior to available alternatives
Compute investment The cost of training and maintaining the model Floor value — replacement cost sets a minimum
Revenue attribution How much revenue the model directly or indirectly generates Primary value driver for commercial models
Transferability Whether the model can be deployed in new contexts without retraining Higher if the model generalises beyond its original application
ℹ Note

The most common valuation approach for AI models in M&A is a combination of replacement cost (what would it cost to recreate the model, including data collection, labelling, training, and iteration) and income approach (what revenue does the model enable). Pure cost approach often understates value because it ignores the competitive advantage; pure income approach risks double-counting with other assets.

Platform Technology and Network Effects

Platform businesses — those that create value by facilitating interactions between two or more groups of participants — derive a significant portion of their value from network effects. A network effect exists when the value of a platform increases as more participants join.

Types of Network Effects

Type Mechanism Examples
Direct (same-side) More users make the platform more valuable for existing users Social networks (Facebook, LinkedIn), messaging (WhatsApp)
Indirect (cross-side) More users on one side attract more users on the other side Marketplaces (Amazon, Uber), app stores (iOS, Android)
Data network effects More users generate more data, improving the product for all users Search engines (Google), recommendation systems (Netflix, Spotify)

Network effects are intangible assets — they create competitive moats, pricing power, and switching costs. However, they are not separately identifiable under IFRS 3 because they cannot be sold independently of the platform itself. In an acquisition, network effects contribute to goodwill rather than identifiable intangible assets.

The platform technology that enables network effects, however, may be separately identifiable. The software, APIs, matching algorithms, and data infrastructure that constitute the platform are technology-based intangible assets that can be valued using standard approaches.

Data Flywheel Value

The most powerful data assets create self-reinforcing cycles — data flywheels — where more users generate more data, which improves the product, which attracts more users. Google's search algorithm improves with every query. Tesla's autonomous driving system improves with every mile driven. These flywheels represent extraordinary intangible value because they are virtually impossible for competitors to replicate without matching the user base and data volume. While the flywheel effect itself is goodwill (not separable), the underlying data and algorithms are identifiable technology-based assets.

Digital Infrastructure as Intangible Capital

Beyond data and algorithms, the configured digital infrastructure that runs a business represents significant organisational capital. ERP systems customised over years, CRM platforms integrated with proprietary workflows, cloud architectures designed for specific workloads, and API ecosystems connecting internal and external systems — all represent accumulated intangible investment.

Infrastructure Asset Categories

Category Examples IFRS 3 Classification
Configured enterprise systems ERP customisations, CRM workflows, HR platform configurations Technology-based (if separable) or goodwill
Integration architecture API layers, data pipelines, middleware configurations Technology-based (if documented and separable)
Cloud infrastructure as code IaC templates, CI/CD pipelines, monitoring configurations Technology-based (if proprietary and documented)
Development tools and frameworks Internal libraries, testing frameworks, deployment tools Technology-based (if reusable beyond original context)
⚠ Warning

Most businesses undervalue their configured digital infrastructure because the customisation costs were expensed incrementally over many years. But recreating a complex ERP configuration from scratch — including the business rules, approval workflows, reporting structures, and integration points accumulated over 5-10 years — can cost millions. This replacement cost establishes a floor value for the infrastructure as an intangible asset.

Data Governance and Protection

Data assets require active governance to maintain their value. Unlike physical assets that depreciate predictably, data assets can lose value suddenly — through quality degradation, regulatory changes, security breaches, or competitive replication.

1. Data inventory and classification

Catalogue all data assets by type, source, ownership, sensitivity, and business criticality. You cannot manage or value what you have not inventoried.

2. Quality management

Implement automated quality checks — completeness, accuracy, consistency, timeliness — with defined thresholds and remediation processes. Data quality degrades without active maintenance.

3. Access control and security

Define who can access, modify, and export each data asset. Implement audit trails. Data assets lose their trade secret protection if access is not actively controlled.

4. Regulatory compliance

Map data assets against applicable regulations — GDPR, CCPA, sector-specific requirements. Non-compliance can destroy data asset value overnight through enforcement actions or required deletion.

Practical Valuation: Data Asset Assessment

For a practical assessment of your data assets, apply this evaluation framework across each identified data asset.

Dimension Questions to Ask Scoring
Uniqueness Can a competitor build an equivalent dataset? How long would it take? 1 (commodity) to 5 (irreplaceable)
Revenue linkage What revenue directly depends on this data? Would revenue decline if the data were lost? 1 (no direct link) to 5 (core revenue driver)
Quality Is the data complete, accurate, current, and consistent? 1 (poor) to 5 (excellent)
Protection Is the data adequately secured, access-controlled, and compliant? 1 (minimal) to 5 (comprehensive)
Scalability Does the data become more valuable as volume grows? Does it enable network effects? 1 (no scale benefit) to 5 (strong flywheel)

Assets scoring above 20/25 represent high-value data assets that merit dedicated management, investment, and protection. Those scoring below 10 should be evaluated for cost-effectiveness — the storage and maintenance costs may exceed the asset's economic value.

What Comes Next

Having examined all major categories of intangible assets across Lessons 1-6, we now turn to the techniques for placing a financial value on them. In Lesson 7: Valuation Methods — RFR, MPEEM, and With-and-Without, we cover the three primary valuation methodologies used in professional practice, with worked examples showing how each method is applied to different asset types.

Ivan Gowan is CEO of Opagio, the growth platform that helps businesses and investors measure, manage, and grow intangible assets. Before founding Opagio, Ivan held senior technology and leadership roles across financial services and digital platforms for 25 years. Meet the team.