Your Biggest Contract May Be the Reason You Cannot Borrow

The usual explanation for a failed raise is that a business has not met the right investor. The harder truth is that demand, revenue and growth can still leave a company with no cash flow a capital provider can safely underwrite.

Standard Bank arranged $202 million of sustainability-linked, multi-currency facilities for M-KOPA, the Kenya-founded company that lets customers in Kenya and Uganda acquire smartphones, solar systems and loans through daily repayments. The facilities included $165 million in Kenya and $37 million in Uganda.

That was the breakthrough. Not the headline growth figure.

The U.S. International Development Finance Corporation, or DFC, the American development-finance institution, then approved $51 million specifically for M-KOPA’s smartphone, solar-system and cash-loan receivables. In other words, the money was tied to identifiable customer payments, rather than to the general optimism surrounding a fast-growing company.

Only after that financing picture comes the scale: by April 2023, M-KOPA had reached three million customers and delivered more than $1 billion of products and services.

M-KOPA’s daily digital repayments, not customer scale alone, created cash flows lenders could finance. Photo: Nicholas Githiri / Pexels, Pexels licence (free commercial use).

That is a good operating story. But the lending story was more precise. M-KOPA had digitally recorded, granular repayments that could be examined and tied to particular assets. It could match local operating-company debt with local-currency collections.

Imagine lending to two businesses. One says it has a huge order book. The other can show you dated repayments, who owes them, whether they have paid, and what claim you have if they do not. Which cash flow would you rather rely on?

Most founders think fundraising becomes a matching exercise once customers arrive: find the right bank, private-credit fund or equity investor. Yet that is often where the misunderstanding begins. A company may be growing rapidly and still not have created something a lender can safely fund.

Revenue is not the asset people think it is

The conventional view has an intuitive appeal. A company proves demand. Revenue rises. Expansion requires money. Surely the remaining problem is access.

It is not that simple.

A capital provider does not fund commercial activity in the abstract. It asks whether expected cash can be checked independently, whether it will arrive when promised, whether somebody else has a prior claim on it, whether it can be controlled if things go wrong, and what can be recovered in a downside case.

Those questions can sound fussy to an operator. They are the whole job for a lender.

The Federal Deposit Insurance Corporation’s commercial-lending guidance tests the age of receivables, customer concentration, weak bookkeeping, foreign enforcement risk, possession of collateral and lockbox control, meaning an account into which customer payments are directed. It does not treat revenue growth as the main test.

This is where most people stop looking. They see a growing business and assume a financing failure must be about relationships. Sometimes it is. More often, reported activity has not yet become cash that a funder can observe, protect and recover.

M-KOPA is the positive version. Carillion, NMC Health, WeWork and Hyflux show how the conversion can fail in very different industries.

The order book that could not be collected

Carillion, the British construction and services group, appeared to have the ingredients lenders like. It worked on major contracts, including public-sector work. It reported profits. Revenue was growing.

Yet borrowing rose from £242 million in 2009 to £689 million in 2016, while its current ratio, a basic comparison of short-term assets with short-term bills, remained around 1.0. Then came the turn. A July 2017 review produced an £845 million contract write-down that later exceeded £1 billion. Of the eventual provision, £729 million related to trade receivables.

The UK Parliament’s investigators found a company that appeared profitable but had not established whether contract income would turn into collectible cash. FTI, which reviewed Carillion’s information for lenders in January 2018, called its historical cash-flow and contract-profitability information “extremely weak”. Carillion entered liquidation on 15 January 2018.

Carillion’s large contracts could not protect lenders once receivables and project economics proved uncertain. Photo: Roger Kidd / Wikimedia Commons, CC BY-SA 2.0.

A public-sector customer does not magically make a receivable safe. Construction revenue may depend on certification, claims, disputed costs and work that overruns. Until those issues are settled, a number recognised in accounts can be a poor thing to lend against.

That should unsettle a common assumption. A large order book can be less fundable than M-KOPA’s thousands of small customer loans. Carillion had bigger counterparties. M-KOPA had payments that were easier to verify and attach financing to.

For a closer look at how lenders turn this distinction into practical controls, read GI Network’s guide to inventory-backed lending and borrowing-base controls. A borrowing base is simply the portion of specific eligible assets a lender is prepared to advance against. The word “eligible” does a lot of work.

The hospitals with a missing perimeter

NMC Health had physical hospitals, real patients and a London listing. Revenue grew from roughly $490 million in 2012 to $2.1 billion in 2018. At its August 2018 peak, the UAE-based healthcare group was worth £8.6 billion and sat in the FTSE 100.

Then the reliability of the whole picture broke.

The UK Financial Conduct Authority found that NMC used dual accounting records and understated debt by as much as $4 billion. It entered administration in April 2020.

The hospitals did not disappear. That was not the issue. Lenders, directors and public investors could not establish the complete liability perimeter: the full boundary of debts and obligations around the business.

A building, a customer base and audited accounts can all be real while the financial claims on them remain impossible to calculate. If a lender cannot know who else is owed money, it cannot sensibly assess its own chance of repayment.

Here is what the numbers do not tell you. Revenue does not reveal whether the cash is already promised to someone else. NMC’s defect was not weak demand. It was an unverifiable balance sheet. No investor introduction fixes that.

The 15-year promise funded by 15-month customers

WeWork, the American flexible-office company, offered another version of the same error. By the second quarter of 2019, it operated 528 locations across 111 cities. Its expansion made it look, to many people, like a technology growth story.

Its planned initial public offering was withdrawn in September 2019 after scrutiny of losses, related-party dealings, voting control and exposure to long property leases. SoftBank Group, the Japanese investor that had already backed WeWork, later accelerated $1.5 billion of funding and contemplated $5 billion of new debt, alongside governance changes and a reconstituted board.

WeWork’s customer demand could not bridge the gap between 15-month memberships and 15-year leases. Photo: Hnapel / Wikimedia Commons, CC BY-SA 4.0.

But a later SEC filing exposed the structural problem. At December 2020, average initial property leases were approximately 15 years, while average membership agreements were 15 months. WeWork had $40.6 billion of undiscounted lease commitments against roughly $3 billion of committed future sales.

Customers were real. Their payments were observable. Yet WeWork had promised landlords far more, for far longer, than its customers had promised it.

This is a duration mismatch: short-lived customer income supporting long-lived fixed obligations. Capital does not merely price growth. It prices how much pain remains when growth stops.

SoftBank is the important exception to the simple story. The right investor can matter. Equity can fund strategic optionality without conventional collateral, and SoftBank did provide funding after public investors refused. But it came with severe repricing, accelerated capital, new debt and governance changes. The money was not a vote that the old structure was safe. It was the price of changing it.

One risky stream can contaminate the rest

In Singapore, Hyflux built Tuaspring, a project combining a desalination concession with a 411MW power plant. A water purchase agreement can sound like the kind of contracted income infrastructure investors seek.

Yet the project also relied on wholesale electricity earnings. In 2017, Tuaspring lost S$81.9 million because electricity prices fell below fuel costs. Its disposal group carried S$514.8 million of borrowings. Hyflux later said that depressed power prices, difficulty repatriating overseas cash and demands for performance-bond deposits drove it into court-supervised restructuring in May 2018.

Tuaspring’s contracted water income was weakened by merchant-power exposure elsewhere in the project. Photo: Arian Fernandez / Pexels, Pexels licence (free commercial use).

The water contract was not fictional. It simply did not isolate the project from merchant-power risk, meaning power income exposed to changing market prices rather than a fixed contract.

Put M-KOPA, Carillion, NMC, WeWork and Tuaspring side by side and a pattern appears that none of the reports says outright: fundability is a chain, not a score.

A company can have demand, assets, named customers and signed contracts. One weak link, from disputed collection to hidden debt to a mismatch in promise lengths to an unhedged revenue line, can decide the outcome.

Legal conditions matter independently of business quality. World Bank research across up to 73 countries found that movable-collateral registries increased the share of firms with credit lines by about 8%, while reducing interest rates and extending maturities. Registries do not make a business better. They make creditor priority and recovery more predictable.

That is an uncomfortable conclusion. Sometimes a business is not less fundable because it is poorly run. It is less fundable because nobody can reliably establish who gets paid first if it fails.

Find the broken link before you raise

For founders and operators, investor readiness should not mean polishing a deck or filling a data room with documents that answer no underwriting question. A nearly complete data room can still fail if it cannot prove cash conversion, ownership, obligations and enforceable rights. GI Network has examined that trap in the data room that was 95% done but lenders still would not fund.

Run five operating tests before outreach. Each needs an owner, a threshold and a remedy.

See it. The finance lead should be able to produce payment-level evidence for every material cash-flow pool: who owes the money, when it fell due, whether it was paid and how concentrated the customer base is. The threshold is simple: no material pool should rely only on management assertion. If it does, fix the reporting trail, reconcile records and separate receivables that cannot be independently verified.

Collect it. The commercial and finance teams should identify revenue that remains subject to certification, dispute, delay or customer set-off. The threshold is that cash offered as support for financing must be demonstrably collectible, not merely booked in accounts. Where the evidence fails, exclude that income from the funding case, renegotiate payment milestones or build a separate plan for disputed claims.

Claim it. Legal counsel should map who owns the receivables and assets, whether they can be pledged, and who has priority if the business fails. The threshold is a documented, enforceable claim on every asset presented to a lender. If overseas, affiliate or restricted assets cannot meet it, do not present them as collateral. Ring-fence eligible assets or seek a capital structure that does not depend on them.

Match it. The chief financial officer should compare the currency and length of customer commitments with the currency and length of debt or fixed obligations. The threshold is no major long-term or foreign-currency obligation that depends on short-term or mismatched cash without a credible protection plan. The remedy may be local-currency borrowing, shorter commitments, a different capital instrument or separating the risky business line.

Survive it. The board and finance lead should maintain one complete view of debt, guarantees, facilities and performance-bond demands. The threshold is that a lender can identify the full liability perimeter and understand what happens if performance weakens. If that view does not exist, reconcile every obligation, resolve conflicting records and stop presenting a forecast as financeable before the perimeter is clear.

Only then should management choose the instrument. Receivables-backed debt may fit granular, visible collections. Equity may fit strategic expansion whose value lies beyond current collateral. A project mixing contracted water income with merchant power exposure may need the risks separated before it can be financed on infrastructure-like terms.

GI Network’s view: The first fundraising deliverable should be a fundability map: every material cash flow, the evidence behind it, the claim on it, the risk around it and the capital type that can genuinely bear that risk.

GI Network would test that map before outreach. In a case like M-KOPA, that means examining repayment data, collection routes, currency alignment and security over receivables. In a case resembling Carillion or NMC, it means stress-testing cash conversion and the full liability perimeter before presenting forecasts as financeable. Then we would align the documents and capital structure with the objections an investment committee is likely to raise, rather than asking a broad market to discover the weaknesses for itself.

What experienced capital sees first

A first-time founder sees revenue as proof that a company deserves funding. An experienced lender sees revenue as the beginning of diligence.

That is not cynicism. It is an attempt to answer one practical question: if the plan disappoints, what remains?

Investors may accept uncertainty because they can own future upside. Lenders are paid a fixed return and therefore care intensely about downside protection. Their concern is not whether a business is admirable, or even fast-growing. It is whether the cash can survive inspection.

That is why capital providers ask awkward questions about a single customer, an overseas contract, a shareholder arrangement, a performance bond or an old facility. Each question is really about one of five things: can the cash be seen, collected, claimed, matched and survived?

The deal often dies there, not in the pitch. As GI Network has argued in the investment committee memo where deals really fail, the audience after the meeting must be able to defend the risk in writing.

The Five-Lock Test

Before saying a business is ready to raise, run the Five-Lock Test. It is the lesson shared by a Kenyan consumer-finance business, a British contractor, UAE hospitals, American offices and a Singapore water project.

See it: Can an independent party verify where cash comes from, who owes it and whether it is arriving?

Collect it: Does reported revenue become cash without disputed claims, long delays or hidden leakage?

Claim it: Are contracts, receivables and assets legally capable of being pledged, with a clear order of priority?

Match it: Does the duration and currency of financing fit the duration and currency of the underlying cash?

Survive it: If performance weakens, are liabilities, governance arrangements and recovery rights clear enough for a capital provider to act?

Pass all five locks and the question becomes which investor or lender fits. Fail one, and the problem is not access to capital.

It is that the business has not yet produced a financeable claim on its own success.