Financial ratios disagree because there is no universal formula for most of them, and providers make different but defensible choices. They use different definitions of the same term, mix trailing-twelve-month and full-year figures, adjust for one-off items differently, and draw from different source filings. The company reports one set of accounts, but the ratios built on top of them are constructed choices, not fixed facts.
Key takeaways
- Most ratios have several valid formulas, so two providers can compute the same named ratio differently and both be defensible.
- Trailing-twelve-month figures and full-year figures cover different periods, so ratios built on each will differ, especially mid-year.
- Providers adjust for exceptional items, leases, and other accounting choices in different ways, which changes the inputs to a ratio.
- Different source-of-truth filings, consolidated versus standalone and provisional versus audited, feed different numbers into the same formula.
Why is there no single formula for a ratio?
Many ratios sound precise but have no universally agreed definition. Take Return on Equity: it divides profit by shareholders' equity, but which profit (before or after exceptional items) and which equity (year-end or average over the year) are choices the provider makes.
The same is true of debt ratios, margins, and coverage ratios. Each has a numerator and a denominator that can be defined in several reasonable ways. When two providers pick different definitions, they report different values for a ratio that carries the same name, and neither is making an error. This is why comparing a ratio across two websites is only meaningful when both use the same definition.
How do TTM and annual figures create differences?
Ratios can be built on the trailing twelve months (TTM), meaning the most recent four quarters, or on the last full financial year. These cover different periods, so they naturally produce different results.
Midway through a financial year, a TTM figure already includes recent quarters that the last annual report does not, so a TTM-based ratio reflects newer performance than an annual one. A provider that uses TTM and a provider that uses the last audited year will disagree simply because they are measuring different stretches of time.
This mirrors the trailing-versus-forward split behind Why PE ratios differ: the ratio changes with the window you measure, not only with the company's performance.
Use the Artha Terminal screener to compare companies on ratios calculated the same way, so differences reflect the companies rather than the method.
How do accounting adjustments change the inputs?
Reported accounts contain items that providers treat differently. Exceptional or one-off items, such as an asset sale or a write-off, may be left in by one provider and stripped out by another to show "core" performance.
Accounting standards themselves allow choices, for instance in how leases, depreciation, or provisions are handled, and providers sometimes normalise these to make companies comparable. Every such adjustment changes the numerator or denominator of a ratio. A margin computed on reported profit and the same margin computed on adjusted profit can look quite different, and both can be labelled the same way.
Why does the source of the numbers matter?
Two providers can also start from different underlying documents. One may use consolidated group accounts while another uses standalone parent-only accounts, which changes the EPS, the Book Value, and every ratio built from them.
There is also a difference between provisional results a company releases first and the final audited figures that follow, which can be revised. A provider that captured the provisional numbers and one that captured the audited numbers will carry different inputs. Because financial data providers each maintain their own pipeline from filing to database, small differences in what they read and when they read it propagate into every ratio downstream. Judging which data to trust is the theme of Reliable vs unreliable data.
How can you compare ratios without being misled?
The practical defence is to compare like with like. Check whether both figures use TTM or annual periods, consolidated or standalone accounts, and reported or adjusted inputs before concluding that a company looks cheap or expensive on one site versus another. A ratio in isolation means little without knowing how it was built.
It also helps to look at the raw components rather than the finished ratio. If you can see the profit, equity, and share count that went into a return figure, a disagreement becomes explainable instead of mysterious.
On Artha Terminal, ratios are shown with the underlying figures and the basis used to calculate them, and you can ask the built-in assistant, Ask Warren, to explain why a particular ratio differs from what you saw elsewhere. That turns a confusing mismatch into something you can reason about.