A company can book £10m of revenue without a single pound of cash arriving. Understanding when and how revenue is recognised changes how you read every set of accounts.
In 2019, WeWork filed to go public. Its S-1 prospectus (the IPO document) claimed revenues were growing rapidly. On the surface it looked like a high-growth tech company — which is exactly what founder Adam Neumann wanted investors to believe.
But analysts started reading the small print on how WeWork recognised revenue. WeWork signed long-term leases (10–15 year commitments, billions in liabilities) and then rented those spaces month-to-month to members. Revenue was short-term and cancellable. Costs were locked in for decades.
WeWork had also invented a metric called "Community Adjusted EBITDA" which excluded most of its actual costs, making it appear profitable when it was losing over $1.9bn per year. This is a textbook example of revenue and earnings manipulation.
The IPO was pulled. The valuation crashed from $47bn to under $3bn within months. Neumann was removed as CEO. WeWork eventually went bankrupt in 2023.
This is one of the most tested concepts in finance interviews because it explains the gap between P&L profit and cash flow — a gap that trips up analysts and creates opportunities for manipulation.
A company can recognise £10m of revenue today without a single pound arriving in the bank. This happens constantly — subscription companies collect cash upfront and recognise it over time; services firms do the work first and get paid later; construction companies recognise revenue over years of a project.
Understanding this is what separates analysts who can read accounts from those who just look at the headline revenue number.
The accounting standard that governs revenue recognition (IFRS 15 in the UK/EU, ASC 606 in the US) uses five steps:
1. Identify the contract with the customer
2. Identify the performance obligations — what has the company promised to deliver?
3. Determine the transaction price — how much will be paid?
4. Allocate the price to each performance obligation
5. Recognise revenue when (or as) each obligation is satisfied
You don't need to memorise all five for most interviews — but knowing the principle (revenue when obligation is met, not when cash arrives) is essential.
Revenue is recognised evenly over the subscription period, regardless of when cash is received. A customer pays £120 upfront for an annual subscription on January 1st. The company receives £120 of cash immediately but only recognises £10 of revenue each month for 12 months.
The remaining unrecognised revenue sits on the balance sheet as deferred revenue (a liability — the company still owes the customer the service).
Deferred revenue is actually a signal of health in SaaS — it means customers have paid in advance, showing confidence. High deferred revenue = strong future revenue visibility.
Watch out for: annual contract value (ACV) vs revenue. A SaaS company might sign £50m of new contracts but only recognise £12.5m in the first quarter. The revenue headline understates the commercial momentum.
Revenue is recognised at the point of transfer of control — when the product is delivered to the customer and the risks and rewards of ownership pass over. For a retailer, this is typically when the customer takes the product. For a manufacturer selling to a distributor, it's when the goods are shipped (FOB shipping point) or received (FOB destination).
This seems simple — but complications arise with warranties, returns policies, and multi-element bundles (like an iPhone + AppleCare).
Apple sells an iPhone that includes iOS updates and iCloud. Under IFRS 15, Apple must allocate the transaction price across each performance obligation — the hardware (recognised at point of sale) and the software/services (recognised over the update period).
This is why Apple's recognised revenue is actually lower than its billing — a portion is deferred for future services. Analysts watching only the headline miss this entirely.
For contracts spanning multiple years (building a bridge, delivering a defence system, a 3-year IT implementation), revenue is recognised using the percentage of completion method. If a £100m contract is 40% complete, £40m of revenue is recognised — even if only £20m of cash has been invoiced so far.
This method matches revenue to the work actually done, preventing a company from recognising all £100m on day one (when it signs the contract) or day last (when the project completes).
This is where construction companies get destroyed. If a company recognises revenue based on estimated completion, but then costs overrun significantly, the revenue is already on the P&L but the profit evaporates.
This is exactly what happened to Carillion (UK construction giant, collapsed 2018). They were recognising revenue aggressively on long-term contracts while concealing that the underlying projects were massively over budget.
For time-and-materials or hourly billing, revenue is recognised as the service is delivered — as hours are worked. For a fixed-fee engagement, revenue is recognised over the period of the work. For success-fee arrangements (common in investment banking — the bank only gets paid if the deal closes), revenue is recognised only when the deal completes and the fee is virtually certain to be received.
Services firms often do the work before invoicing — a law firm works all of October and sends the invoice in November. That October work is accrued revenue (or "work in progress") — recognised on the P&L but cash hasn't arrived yet. This is receivables risk: if the client disputes or delays, the revenue is on the books but the cash might not come.
