Efficiency and working capital

In depth

Asset turnover

Asset turnover is revenue divided by average total assets; fixed-asset turnover uses average property and equipment, net, instead. Because revenue is earned over the whole year and assets change during it, the average of the opening and closing balances is the better divisor. Lantern 2027: average total assets $(93{,}676 + 327{,}883) \div 2 = 210{,}779.5$, so asset turnover is $577{,}860 \div 210{,}779.5 = 2.74$. For 2026 there is no opening balance sheet (Lantern started on 1 March with nothing), so the closing balance stands in, and the ten months of revenue must be annualised before comparing: $193{,}320 \times 12 \div 10 \div 93{,}676 = 2.48$.

Asset turnover depends on the business model as much as on management. A café chain with leased premises has a different asset base from a roaster that owns its plant. And since ASC 842 the right-of-use asset for leased premises is in total assets: Lantern’s $173,379 of right-of-use assets is more than half its year-end total. Leaving it out, average assets would be $124,090 and asset turnover 4.66. Neither figure is wrong; compare like with like.

Receivables: turnover and days

Receivables turnover divides credit sales by average net receivables; days sales outstanding (DSO) turns it into days: the average number of days customers take to pay. Only sales on credit create receivables, so only they belong in the numerator. Lantern’s café, shop and workshop sales are paid at the till or online at once; only catering is invoiced on 30-day terms.

Even catering revenue needs care. Lantern’s 2027 catering revenue of $118,360 includes $30,840 for the law firm, which is in the contract asset and has never been billed, and $1,640 for January events paid for by deposits in 2026. Neither passed through accounts receivable. The sales that did were the $85,880 invoiced in the year. Matching flow and balance:

Using all catering revenue gives 27.0 days, and using total revenue 5.5 days: both divide receivables by sales that never became receivables, so both understate how long clients take to pay. The aging schedule is consistent with 37.2: of the $11,460 owed at the year end, $3,060 was more than 30 days old, past Lantern’s terms. For 2026, with closing receivables of $6,796 and $59,920 invoiced in ten months, DSO is $6{,}796 \div 59{,}920 \times 304.17 = 34.5$ days. The days in a ten-month period are $365 \times 10 \div 12 = 304.17$; using 365 with ten months of sales would overstate the days by a fifth.

The contract asset: revenue waiting to be billed

The contract asset is not a receivable, so it stays out of DSO. It is still working capital: cash Lantern has earned and is waiting for. Measure it in days of revenue. At 31 December 2027 it held $30{,}840 \div 577{,}860 \times 365 = 19.5$ days of Lantern’s total revenue, and 274 days, nine months, of the law firm’s fee: every event since April. Counting it as if it were a receivable, DSO would be 75.6 days. That figure mixes two kinds of asset and is not a collection measure, but it shows a lender how much of Lantern’s revenue is still outside its bank account. Whether 2027’s profit is as good as it looks, given that $30,840 of it rests on a CPI estimate and one client, is the cash-flow-quality lesson’s question.

Inventory: turnover and days

Inventory turnover divides cost of goods sold by average inventory; days inventory outstanding (DIO) is the average number of days stock waits before it is sold. Cost of goods sold, not revenue, because inventory is carried at cost. Lantern’s cost of sales, $224,205, also includes $61,080 of cost of services (event baristas and catering supplies), which never passed through inventory, so use cost of goods sold, $163,125:

With the whole cost of sales the answer is 12.4 days, which flatters the stock. For 2026, $5{,}800 \div 52{,}800 \times 304.17 = 33.4$ days. Lantern’s stock moves faster now because café stock (milk, pastries, cups) is used within days, while bagged beans for the shop can wait for weeks. A blended ratio hides that, and a business with very different product lines is better analysed line by line where the data allow.

Payables: turnover and days

Days payables outstanding (DPO) is the average number of days the business takes to pay its suppliers: average payables divided by purchases on credit, times the days. Purchases, not cost of goods sold, because payables arise when goods are bought, and Lantern’s payables include supplies and unpaid utility bills as well as beans. In 2027 Lantern bought $187,780 on account ($166,800 of stock, $19,800 of supplies and $1,180 of December utility bills): $9{,}245 \div 187{,}780 \times 365 = 18.0$ days. In 2026, $9{,}210 \div 70{,}430 \times 304.17 = 39.8$ days. When purchases are not disclosed, as for most public companies, analysts use cost of goods sold (or cost of sales) and say so.

The 2026 figure is inflated by the year end. Lantern bought heavily for the December gift peak, so its payables at 31 December were high compared with an average month, and with no opening balance there is nothing to average against. Kilnworth’s terms are 30 days in both years. A year-end balance in a seasonal business can mislead any days ratio; quarterly balances, where available, give a better average.

The operating cycle and the cash conversion cycle

The operating cycle is inventory days plus receivables days: how long from buying stock to collecting cash for it. Suppliers finance part of it; the cash conversion cycle is the rest:

Lantern 2027: $17.1 + 37.2 - 18.0 = 36.3$ days, against 28.1 in 2026. Read it with care for a mixed business: it is a composite of two different flows, and no real cash flow runs through both. The 37.2 days of DSO apply only to catering, about 15% of revenue, a service business whose cycle is its collection period less whatever supplier credit its supplies get. The goods sold in the cafés and the shop are paid for at once, so their cycle is inventory days less payables days, $17.1 - 18.0 = -0.9$ days: they are paid for their stock a little before they pay for it. A firm-wide figure, with receivables measured in days of total revenue, is $5.5 + 17.1 - 18.0 = 4.6$ days. Each view answers a different question; the composite 36.3 is best used only to compare like with like over time. The cycle has lengthened on every basis, and Lantern now pays its suppliers within 18 days though their terms allow 30: as cash grew scarce, it was not taking the credit it was offered.

A negative cycle: Amazon

A business that sells its stock quickly, is paid at once and pays its suppliers late can have a negative cash conversion cycle: its suppliers finance its operations and growth adds cash rather than using it. Amazon’s 10-K figures for 2013 and 2014 show it.

Computed from Amazon’s consolidated balance sheets (31 December 2012, 2013 and 2014) and statements of operations. Receivables are Amazon’s “accounts receivable, net and other”; credit sales and purchases are not disclosed, so total net sales and cost of sales are used. At 31 December 2014 accounts payable ($16,459 million) exceeded inventories and receivables together ($13,911 million).

For Amazon, each extra dollar of sales brought cash in before the stock had to be paid for. For Lantern’s catering the opposite holds: each extra invoiced event ties up cash for as long as the client takes to pay, about 37 days, and the law firm’s contract for up to a year.

How growth consumes working capital

When the cycle is positive, working capital grows with sales: more receivables, more stock, partly offset by more payables. A rough rule:

If the days stay the same, operating working capital grows in proportion to sales, and that increase is cash the business must find before it can spend its profit. The indirect method shows Lantern’s 2027 bill: working capital absorbed $33,890, of which $30,840 was the contract asset, $3,902 the increase in receivables and $3,675 the increase in stock, partly offset by $2,960 more unearned revenue and $2,200 more wages payable. That is why $22,431 of net income and $19,800 of depreciation became only $8,341 of cash from operations.