Customer Relationships as Collateral: Can You Lend Against Customer Lifetime Value in the AI Era?

During my years at NM Rothschild, we structured asset-backed securities across a spectrum of collateral classes — from the conventional (aircraft leases, real estate) to the innovative (intellectual property, toll road revenues). The common requirement was the same: predictable, measurable cash flows with defensible recovery mechanisms.

The question I am asked increasingly by PE firms and growth company finance teams is whether customer lifetime value — the present value of all future revenue from a customer relationship — meets these criteria and can therefore serve as collateral for structured lending.

The answer is more nuanced than simple "yes" or "no." CLV can be collateral. The structures exist. But the measurement and risk management requirements are more stringent than most companies recognise. And AI, which offers genuine advantages in CLV prediction, introduces new complexities.

The Precedent: Revenue-Based Financing Already Values CLV

Revenue-based financing (RBF) already exists as a proven asset class. An RBF lender advances capital against the projected future revenue of a company or product line, repaying the principal plus a percentage of revenue until a cap is hit. This is, in essence, lending against customer lifetime value — the mathematical foundation of the underlying collateral is the stream of future customer payments.

RBF has moved from novelty to mainstream. Suppliers of capital (including large financial institutions) now offer RBF facilities to profitable SaaS companies with transparent revenue data and high net revenue retention. The structure works because:

Revenue is measurable and contractual. SaaS revenue is not based on sales projections or customer relationships that might evaporate. It is contractual recurring revenue — customers are committed for defined terms, and the terms are documented.

The cash flow pattern is predictable. With sufficient historical data and cohort-based analysis, the revenue stream can be modelled with reasonable precision. Churn rates, expansion rates, and downsell rates are observable across customer cohorts.

The recovery mechanism is direct. In the event of borrower distress, the lender has a claim on future revenue, which is analogous to a claim on cash flows from physical assets.

These three characteristics — measurability, predictability, recovery mechanism — are precisely what structured finance requires.

From Revenue-Based Financing to CLV-Backed Lending

The conceptual leap from RBF to CLV-backed structures is small but material. Instead of advancing against the total revenue of the company, a lender could advance against the CLV of a defined customer segment or portfolio.

This creates more granular, flexible financing. A SaaS company with £50 million ARR could potentially secure:

• A £3 million facility against the CLV of enterprise customers (higher CLV per customer, lower concentration risk)
• A £1.5 million facility against the CLV of SMB customers (lower CLV, higher churn risk)
• A £500k facility against the CLV of newly acquired cohorts (unproven, higher risk, lower advance rate)

The economics are more efficient than a single blended RBF facility because the lender can price each segment according to its specific risk profile.

What CLV Must Demonstrate to Be Collateral

For CLV to function as collateral in a structured finance context, three conditions must be met:

First, defensible measurement. CLV must be calculated consistently using defined, auditable methodologies. The calculation should include customer acquisition cost, gross margin contribution, retention probabilities, and expansion revenue. It should be reviewed by independent accountants and be stable over time.

Second, recovery mechanisms in place. The lender must have a clear path to recover against the collateral. This is less straightforward than with physical assets. One structure involves assigning customer contracts or revenue streams to a special purpose vehicle that sits outside the operating company's insolvency waterfall. Another approach is a charge over future customer payments. The precise mechanism depends on the jurisdiction and the underlying revenue contracts.

Third, concentration and risk management. This is where CLV lending differs most from traditional RBF. A single large customer relationship might represent 15-20% of CLV. That concentration risk must be quantified, priced, and managed actively. The lender should require the borrower to maintain minimum diversification thresholds and monitor customer health metrics continuously.

How AI Enhances CLV Prediction — and Introduces New Risks

Artificial intelligence is already transforming CLV prediction. Machine learning models can identify patterns in customer behaviour that improve churn forecasting, predict expansion propensity, and estimate CLV with greater accuracy than traditional methods.

The opportunity is real. A SaaS company using ML-enhanced CLV prediction can:

• Improve forecast accuracy by 12-18% through better churn identification and expansion scoring
• Enable granular lending by accurately predicting CLV for customer cohorts below the blended company level
• Manage concentration risk actively by identifying which customers are at high churn risk and require relationship investment

But the risks are equally real. An AI-enhanced CLV model depends on data quality, feature engineering discipline, and continuous monitoring. If the model drifts — if the patterns it learned from historical data no longer predict future behaviour — the collateral value can decline sharply.

For lenders, this introduces a new due diligence requirement: assessing the quality and robustness of the AI systems that underpin CLV calculations.

The AI Due Diligence Framework for CLV Lending

A responsible lender will require:

Model transparency and validation. The borrower must explain the model architecture, the features used, and the validation approach. The lender should have access to performance metrics: how accurately did the model predict churn in the last 12 months? How has model accuracy trended?

Data governance. The data feeding the model must be clean, consistent, and auditable. Missing values, outliers, and data quality issues should be documented and their impact on model performance assessed.

Monitoring and update procedures. The model must be monitored continuously for drift. If actual churn exceeds predicted churn by more than 5%, the model should be flagged for retraining. The borrower should have documented procedures for model updates.

Human oversight. AI-driven CLV predictions should not be black boxes. Someone on the borrower's team should understand why a given customer's CLV was calculated at a specific level. Interpretability is not optional.

The Structured Product Opportunity: CLV-Backed Securities

The further innovation horizon is CLV-backed securities — pooled portfolios of customer relationships with defined CLV characteristics, financed through tranched debt instruments.

The parallel is the asset-backed securities market, where loans (mortgages, auto loans, credit card receivables) are pooled, securitised into tranches (senior, mezzanine, subordinated), and distributed to investors. A CLV securitisation would follow the same structure:

This structure has not yet been implemented at scale, but the legal and financial engineering required is well understood. The barrier is not technical — it is institutional. Rating agencies have not developed CLV securitisation rating methodologies. Investors have not built CLV credit models. Trustees do not yet offer CLV portfolio servicing.

But the conditions are ripening. As CLV measurement becomes more rigorous and AI-enhanced CLV prediction gains institutional acceptance, the infrastructure for CLV securitisation will develop. The first mover — the lender or investment bank that successfully structures a rated CLV securitisation — will unlock a significant capital market opportunity.

The Practical Challenges: What Still Needs Solving

CLV-backed lending works in theory. In practice, four challenges remain material:

1. Customer Concentration Risk

A £50 million SaaS company might have 2,000 SMB customers (average CLV £25k) and 10 enterprise customers (average CLV £2.5 million). The 10 enterprise customers represent 55% of total CLV. If one customer churns, the lender's collateral value drops 5%. If two customers churn, it is down 10%.

This concentration risk is qualitatively different from concentration risk in traditional lending. A mortgage portfolio with 1,000 homeowners is uncorrelated; the probability of one homeowner defaulting is independent of others. But customer churn is often correlated: a recession affects multiple customers simultaneously, or a product issue affects a customer cohort.

Responsible CLV lending requires active concentration risk management. A lender might require:

• No single customer to exceed 20% of collateral
• Top 10 customers to represent no more than 50% of collateral
• Continuous monitoring of customer health indicators
• Mandatory relationship investment if a key customer shows churn warning signs

2. Churn Volatility and Market Conditions

Customer churn is not constant. During economic downturns, churn rates spike. A SaaS company might normally see 5% annual churn but 12% churn in a recession. This creates mark-to-market losses for CLV collateral.

A £20 million CLV pool with 5% assumed churn is valued at £19 million. If actual churn spikes to 12%, the pool is worth £17.6 million — an 8% loss. For a lender using this as collateral, this translates directly to increased risk.

CLV-backed lending structures must incorporate volatility buffers: advance rates must be conservative enough to absorb foreseeable volatility; haircuts must be applied to account for macro stress scenarios; and pricing must reflect the volatility risk.

3. Customer Stickiness and Switching Costs

CLV depends on the assumption that existing customers remain customers. But customer switching is increasingly frictionless. SaaS products are web-based, with no switching costs. If a competitor launches a better product, customers can migrate in weeks.

The CLV of a SaaS company with weak switching costs and undifferentiated features is structurally weaker than the CLV of a company with embedded switching costs or genuine product differentiation.

A rigorous CLV collateral assessment must measure switching costs explicitly:

• How expensive is it for a customer to migrate to a competitor (data migration, training, integration)?
• How differentiated is the product vs. alternatives?
• What is the Net Revenue Retention — is the company expanding within existing customers or losing share?

4. Regulatory and Tax Clarity

CLV-backed lending exists in a regulatory grey area in many jurisdictions. Is a claim on customer relationships a security? A commercial loan? A derivative? The treatment affects pricing, capital requirements for lenders, and tax treatment for borrowers.

The SNA 2025 update represents progress — it officially recognises customer relationships as capital assets. But tax treatment, accounting standards, and regulatory clarity remain inconsistent across jurisdictions.

A CLV Collateral Assessment Framework

For companies seeking CLV-backed lending, and lenders evaluating CLV as collateral, here is a structured assessment framework:

A company scoring green across most dimensions can support CLV collateral at near-blended valuations. Yellow requires discounting and heightened monitoring. Red makes CLV collateral impractical.

The Path Forward

CLV is not yet a mainstream collateral class. But it is moving rapidly in that direction. The convergence of three factors is accelerating adoption:

Better AI-driven CLV prediction is making customer lifetime value measurable with greater accuracy and finer granularity than historical methods allowed.

Institutional acceptance is growing. Rating agencies are beginning to study CLV securitisation methodologies. Investment banks are exploring pilot transactions. Large financial institutions are hiring specialists in customer-based lending.

Market demand is evident. SaaS companies need capital-efficient financing alternatives to dilutive equity rounds. PE firms need flexible financing for portfolio company leverage. Growth company CFOs are asking about CLV-backed lending as a strategic financing tool.

The company that structured the first institutional CLV securitisation (aiming for a rating agency rating) would establish itself as the market leader in this nascent asset class. The first CLV-backed SPAC or investment vehicle would unlock significant capital.

For companies with strong, defensible, AI-enhanced CLV measurement, CLV-backed lending is no longer theoretical. It is a practical capital source that improves on traditional equity dilution and unsecured debt alternatives. The structures exist. The lenders are emerging. The question is whether your company's customer relationships meet the collateral standards required.

Tony Hillier is Co-Founder of Opagio. He holds an MA from Balliol College, Oxford and an MBA with distinction. His career includes executive board roles at NM Rothschild & Sons and GEC Finance, and a non-executive directorship at Financial Security Assurance in New York, where he specialised in structured finance and asset-backed securities across global capital markets.