Ozon (OZON): a breakdown by the numbers
· 4 min · beginner
Automated material · TradeAlmanac editorial deskDraft prepared by a language model from our stored data; not reviewed by an editor.
Analysing OZON with the same multiples that are applied to Sberbank or Gazprom produces a figure that is formally calculated but explains nothing. A different logic is at work here: the company lives on the turnover of goods it does not own and on a percentage of other people's transactions, so it has to be read from the revenue structure downwards, not from profit upwards. Always take the current values for the stock from the OZON card and from the financial reports section — what is examined here is the mechanism that makes those values meaningful.
Turnover and revenue are different things
A marketplace brings a seller and a buyer together. The platform's gross merchandise turnover includes everything sold through it, including the goods of third-party sellers that the company never bought and never put on its own balance sheet. Revenue, by contrast, contains only what the company earned itself: the commission charged to the seller, fees for logistics and storage, advertising inside the platform, and revenue from its own direct sales.
This is the first fork in the analysis. If the company is growing turnover through third-party sellers, revenue grows more slowly than turnover — and that is normal, not a slowdown. If it is growing through its own purchases, revenue jumps, but inventories swell along with it and margins fall. One and the same turnover trend means opposite things depending on which model is producing it. That is why comparing the growth rate of turnover with the growth rate of revenue is more informative than looking at either of them in isolation.
Why P/E says little here
The value of no data for a company of this type is a metric with a catch. The denominator of a fast-growing platform is small, unstable and easily turns negative because of decisions that have nothing to do with operating efficiency: revaluations, non-recurring write-offs, spending on stock option programmes, exchange-rate effects. The multiple then either shoots up to absurdly high values or loses its meaning altogether.
What can actually be read:
- The share of commission and advertising revenue. Advertising inside a marketplace is the most profitable line, and it scales with almost no increase in costs. Its weight in revenue says more about the maturity of the platform than the profit for a quarter does.
- Logistics costs relative to turnover. Warehouses and delivery are the main capital-intensive block. If turnover is growing faster than logistics costs, the infrastructure is starting to earn back the investment.
- Cash flow versus profit. A marketplace has a gap between receiving money from the buyer and paying it out to the seller, and this gap provides free financing of working capital. Positive cash flow alongside a reported loss is a routine situation here, not an accounting trick.
Fintech inside the platform
A marketplace develops a lending business of its own: instalment plans for buyers, loans to sellers against future receipts, a payment service. This is a separate set of economics, and it must not be mixed with the trading business. The loan portfolio brings in interest income but carries the risk of non-repayment, and the provisions set aside for it eat into profit unevenly. In the analysis, look at the segment breakdown: what part of the result came from trading and what part from financial services, and how the cost of risk is behaving in the lending segment.
A banking unit inside a non-financial company is the reason why balance-sheet multiples become blended and stop being comparable with retail peers.
The order of analysis
Step 1. Open the financial reports and take the trends in turnover and revenue, and the ratio between them, for comparable periods.
Step 2. Break revenue down by source: own sales, commissions, advertising, logistics services, financial services.
Step 3. Check operating cash flow and the change in working capital — that is exactly where you can see whether the platform is financing its growth itself.
Step 4. Compare the result with stocks where profit is stable and the multiples work by the textbook: the Sberbank breakdown and the Rosneft breakdown are useful as a contrast, not as a benchmark.
Dividends and tax
A growing platform usually channels all of its free cash flow into infrastructure, not into payouts. You can check whether any payouts have been declared in the dividend calendar:
The absence of a payout is not a flaw; for a company at this stage it is a consequence of the fact that investing internally yields more than distributing cash externally.
The tax side — the calculation of the tax base for income from securities — is determined by 13%; when calculating your result, take it into account before you compare the return with alternatives such as OFZ bonds.
Unfamiliar terms from the financial statements are explained in the glossary, and the full list of traded securities is in the stocks section.
Related instruments
Draft prepared by a language model from our stored data; not reviewed by an editor.
Model: claude-opus-5
How we use language modelsSimilar articles
- Abrau-Durso (ABRD): a metrics-based breakdownIn short: the stock is hard to read through a head-on comparison of its multiples with the market.
- Sberbank (SBER): a breakdown by the numbersSberbank cannot be analysed with the same multiples used for a factory or a retailer: a bank has no revenue or EBITDA in the usual sense; it has an interest margin, a cost of risk and capital, which by law constrains everything else, the dividend included.
- Rosneft (ROSN): a metric-by-metric breakdownRosneft is read not through its earnings multiple but through the link between debt, capital expenditure and the base from which the dividend is calculated: for a state-controlled company with a heavy investment programme and revenue that depends on the export price and the tax regime, net profit is the most volatile
- Gazprom (GAZP): a metric-by-metric breakdownGazprom's multiples cannot be read the way a bank's or a retailer's are: the holding's profit is built from segments with different economics, and non-cash items — revaluations, impairments, foreign exchange differences — regularly land on the bottom line.