Data Assets and Data Financing

Intangible Finance — Lesson 4 of 10

Data has been described as "the new oil" so frequently that the phrase has become a cliche. But unlike oil, data is non-rivalrous (it can be used simultaneously by multiple parties without being depleted), it appreciates with scale and combination (larger datasets are disproportionately more valuable), and its extraction cost is marginal once the collection infrastructure is in place. These properties make data a fundamentally different type of asset — and they require fundamentally different financing structures.

Proprietary data is already one of the most valuable intangible assets in the modern economy. Technology companies are valued largely on their data assets. Healthcare companies compete on clinical data. Financial institutions differentiate through alternative data. Yet data finance — the use of proprietary data as collateral, the securitisation of data-derived revenue streams, and the structured monetisation of data assets — is still in its earliest stages.

★ Key Takeaway

Data financing is the newest frontier of intangible finance, and it presents both the greatest opportunity and the greatest structural challenges. Unlike patents or trademarks, data assets lack standardised legal ownership frameworks, established valuation methodologies, and clear enforcement mechanisms. However, the economic value is undeniable — companies like Palantir, Snowflake, and Bloomberg have built multi-billion-dollar businesses on proprietary data. The organisations that develop rigorous frameworks for valuing, protecting, and financing their data assets will have a significant competitive advantage as this market matures.

The Data Asset Landscape

Not all data is created equal as a financial asset. The value of data depends on its exclusivity, its relevance, and the infrastructure required to collect, maintain, and analyse it.

$274B estimated global data monetisation market (2025)
64% of enterprises report data as a strategic asset
<10% have formally valued their data holdings

Data Asset Categories

Category Description Financing Potential Example
Proprietary operational data Data generated through business operations (transactions, logistics, customer interactions) High — exclusive, hard to replicate, generates measurable value Retailer transaction data; logistics route optimisation data
Collected market data Data aggregated from external sources through proprietary collection infrastructure Medium-High — value depends on uniqueness of collection and curation Bloomberg terminal data; alternative data providers
User-generated data Data created by platform users through their activity Medium — subject to privacy regulation and user consent Social media engagement data; usage analytics
Scientific and research data Clinical trials, experimental results, research datasets Medium — value depends on exclusivity and IP protection Pharmaceutical clinical trial data; genomic databases
Synthetic data Artificially generated data that mirrors real-world distributions Emerging — growing use in AI training and testing AI training datasets; financial stress-test data

Data Valuation Approaches

Valuing data assets is more challenging than valuing patents or trademarks because data lacks the standardised legal frameworks (patent claims, trademark registrations) that provide objective reference points. Three principal approaches have emerged.

Cost Approach

The cost approach values data at the cost of recreating or replacing it. This includes: collection infrastructure costs, data cleaning and normalisation, historical accumulation period, and specialist personnel costs.

✔ Example

A healthcare analytics company has spent 8 years building a proprietary dataset of 50 million anonymised patient records, collected through partnerships with 200 hospital systems. The cost approach would value this dataset by estimating the total investment required to recreate it: the cost of establishing hospital partnerships ($15M), data integration infrastructure ($8M), legal and compliance costs ($5M), and 8 years of accumulated data ($4M/year in ongoing costs). The replacement cost estimate: approximately $60 million. However, this approach understates the true value — because it ignores the network effects and the competitive moat created by exclusive partnerships that cannot be easily replicated.

Income Approach

The income approach values data based on the revenue or cost savings it generates. This is conceptually similar to the Relief from Royalty method used for patents: what would the organisation pay to licence equivalent data from a third party?

Input Method Challenge
Revenue attributable to data Isolate revenue streams dependent on proprietary data Data often enables rather than directly generates revenue
Licensing revenue Value of data licensing agreements Many companies do not licence their data externally
Cost savings Operational efficiencies created by data analytics Difficult to isolate data's contribution from other factors

Market Approach

The market approach uses comparable transactions to benchmark data values. Data marketplace transactions (Snowflake Marketplace, AWS Data Exchange, Dawex), M&A transactions where data was a primary acquisition driver, and data licensing agreements all provide market evidence.

The Valuation Gap

The fundamental challenge of data valuation is that data's value is context-dependent. A dataset worth $50 million to a company with the analytics infrastructure and business context to exploit it may be worth $5 million to a buyer who lacks those capabilities. This context dependency makes data valuation inherently more subjective than patent or trademark valuation — and it is the primary reason why data-backed lending remains in its early stages. Lenders need confidence in liquidation value, and context-dependent assets have uncertain liquidation value.

Data Financing Structures

Despite the valuation challenges, several data financing structures have emerged and are gaining traction.

Data-Backed Credit Facilities

The most straightforward structure mirrors IP-backed lending: the data asset is pledged as collateral for a credit facility. The lender obtains a security interest over the data and the infrastructure required to access it.

Data-Backed Facility Terms

Parameter Typical Range Commentary
LTV ratio 10-30% More conservative than patent-backed lending due to valuation uncertainty
Tenor 1-3 years Shorter than IP facilities; reflects data obsolescence risk
Revaluation Semi-annual or quarterly More frequent revaluation required due to data dynamics
Data custody Escrow or third-party hosting required Ensures lender access to data in default scenario
Regulatory compliance GDPR/CCPA compliance certification required Non-compliance can destroy data's commercial value

Data Trusts and Pooling Structures

A data trust is a legal structure where data from multiple contributors is pooled and managed by an independent trustee. The trust can licence data access to third parties, generating revenue that flows back to contributors. For financing purposes, the trust structure creates a legally separable entity with identifiable assets and revenue — which is significantly easier to finance than data held within a corporate entity.

Revenue Securitisation of Data Products

Where data generates a predictable revenue stream — through data licensing, API access fees, or data-as-a-service subscriptions — that revenue stream can be securitised. The structure mirrors royalty securitisation: future data revenue is sold to an SPV, which issues notes backed by the expected cash flows.

ℹ Note

Data financing structures must navigate complex regulatory requirements. Under GDPR (in the EU and UK) and CCPA (in California), personal data cannot be "sold" in the traditional sense — it can only be processed with appropriate legal basis. This does not prevent data financing, but it requires careful structuring to ensure that any transfer of data (whether as collateral or through securitisation) complies with data protection requirements. Non-compliance risk is itself a material consideration for lenders — a data asset that becomes non-compliant loses its commercial value.

Regulatory Landscape

The regulatory environment for data finance is evolving rapidly and varies significantly across jurisdictions.

Global Regulatory Comparison

Jurisdiction Framework Data as Property? Financing Implications
European Union GDPR + Data Act (2025) No property right in data; rights are access and processing rights Data cannot be "owned" but can be commercially controlled; licencing structures preferred
United Kingdom UK GDPR + Smart Data schemes Similar to EU; emerging data asset recognition Post-Brexit divergence creating potential for data finance innovation
United States CCPA/CPRA + sector-specific laws No federal data property right; varies by state Fragmented landscape; sector-specific opportunities (healthcare, financial)
Singapore PDPA + data economy initiatives Progressive approach to data commercialisation Government-supported data financing schemes being piloted
China PIPL + data trading exchanges Data recognised as production factor; data exchanges established Beijing, Shanghai, and Shenzhen data exchanges enable structured trading

China's approach is particularly notable: the establishment of government-sanctioned data exchanges in major cities represents the most explicit recognition of data as a tradeable financial asset. While the regulatory context is fundamentally different from Western markets, the structural innovation provides a useful reference point for how data finance infrastructure might develop globally.

Practical Considerations

For CFOs and investors evaluating data financing opportunities, several practical considerations determine whether a data-backed transaction is feasible.

Consideration Key Question Implication
Exclusivity Is the data exclusively held, or is it replicable from public sources? Only exclusive, proprietary data has meaningful collateral value
Separability Can the data be separated from the business and transferred to a buyer or licensee? Data embedded in operational systems is harder to collateralise than data products
Regulatory compliance Is the data collected, stored, and processed in compliance with applicable regulations? Non-compliant data is a liability, not an asset
Technical custody Can the data be held in escrow or by a third-party custodian for lender protection? Physical (or cloud-based) custody is essential for enforcement
Decay rate How quickly does the data lose value if not updated? High-decay data (social media, market data) requires shorter tenors and more frequent revaluation
⚠ Warning

Data obsolescence is a materially different risk from patent or trademark obsolescence. A patent remains valid for its full term regardless of market changes. Data, by contrast, can become valueless through obsolescence (outdated market data), regulatory change (a new privacy law that prohibits previously lawful uses), or competitive replication (a competitor building an equivalent dataset). Lenders must underwrite data decay risk explicitly — and borrowers must demonstrate ongoing data refresh capabilities.

What Comes Next

In Lesson 5: Revenue-Based Financing and Intangible Assets, we examine how recurring revenue streams — particularly SaaS subscriptions and licensing income — enable non-dilutive financing backed by intangible revenue. This structure has become the dominant growth capital instrument for software and technology companies.

Tony Hillier is an Advisor to Opagio, bringing over 30 years of experience in structured finance, M&A advisory, and business valuation. His work spans due diligence, purchase price allocations, and intangible asset monetisation for institutional clients across the UK and Europe. Meet the team.