Ratio Analysis for Accountants: A Complete Guide to Liquidity and Leverage Ratios
7/5/2026
The most useful thing to understand about ratios is that a ratio is a question, not an answer.
Its entire value is comparative — against the same company over time, against a peer, or against a covenant threshold. A ratio computed once, for one period, with nothing to compare it to, tells you essentially nothing. Which means the analytical work is in the comparison and the decomposition, not in the arithmetic.
The Four Families and What Each Actually Asks
Liquidity: can it pay what is due soon?
Leverage and solvency: can it survive, and who has a claim on the value?
Activity and efficiency: how hard are the assets working?
Profitability and return: is the capital earning anything?
Most analysis over-weights the first family and under-weights the fourth, which is backwards — liquidity problems are usually symptoms, and return on capital is the question that determines whether the business is worth owning.
Liquidity
The current ratio — current assets over current liabilities. Universally cited and, as our post on working capital analysis argues, easily inflated by exactly the conditions that should worry you: dead inventory and uncollectible receivables both raise it.
The quick ratio, removing inventory. Better, and still a level measure.
The cash ratio, the most conservative.
The defensive interval — liquid assets divided by average daily operating expenditure, expressed in days. Underused and genuinely informative, because it answers a question an owner understands: how long could we operate if collections stopped?
And the measure that beats all four: the cash conversion cycle, which converts these balances into a duration and is trendable. Liquidity ratios describe a position; the cycle describes a process.
Leverage and Solvency
Debt to equity, and immediately the definitional question: which debt? Interest-bearing obligations only, or all liabilities? Analysts differ, covenants define it explicitly, and a ratio quoted without its definition is not comparable to anything.
Debt to tangible net worth, which is the version most covenants use — equity less intangibles and goodwill — and which our post on preparing for a loan application covers from the lender's side. A company whose equity is substantially goodwill has far less tangible net worth than its balance sheet shows.
Debt to total assets, and debt to capital.
And the coverage ratios, which are income statement measures and which are the ones that actually decide a credit:
Interest coverage — earnings before interest and taxes over interest expense.
Fixed charge coverage, which adds lease and other fixed obligations, and is more informative for a business with substantial leases.
Debt service coverage — cash available over principal and interest due. This is the ratio a lender decides on, and a balance sheet with excellent leverage ratios and inadequate coverage does not get the loan.
Activity and Efficiency
Receivable turnover and days sales outstanding. Computed properly — the countback method rather than a single period's revenue, per the working capital post.
Inventory turnover and days inventory, by category rather than in aggregate, since aggregate turns conceal dead stock.
Payable turnover and days payable, read together with whether terms were negotiated or stretched.
Total asset turnover, which asks how much revenue each dollar of assets produces, and fixed asset turnover for capital-intensive businesses.
These are the ratios that most often explain a change in the profitability ratios, which is why they belong in any set.
Profitability and Return
Gross margin, which isolates pricing and direct cost.
Operating margin, which adds the cost of running the business.
Net margin, which includes financing and tax and is therefore the least useful for comparing operations.
Return on assets — earnings over assets, asking what the asset base produces.
Return on equity, which is the most quoted and the most misleading, for reasons the decomposition below makes clear.
And return on invested capital, which is the most important ratio in this list and the one least often computed. It asks whether the business earns more on the capital employed than that capital costs. A company earning modest returns on a small capital base can be a far better business than one earning large absolute profits on an enormous one — and no other ratio here asks that question.
If a practitioner adds one measure to their standard analysis, this is it.
The Decomposition That Turns a Ratio Into a Diagnosis
The single most valuable technique in ratio analysis, and it is not a ratio.
Return on equity decomposes into three drivers:
net margin × asset turnover × leverage
Which converts "return on equity improved" into an actual explanation. Did the improvement come from better margins (pricing or cost control), better asset utilization (more revenue per dollar of assets), or simply more debt?
Those have entirely different implications. Margin and turnover improvements are operational achievements. A leverage-driven improvement means the business became riskier and the return rose because equity shrank relative to debt — which looks identical in the headline ratio and is the opposite in substance.
This is the analysis that distinguishes a useful ratio review from a table of numbers, and it takes two minutes once the underlying figures exist.
The Computation Traps
Where ratio analysis actually goes wrong.
Average versus ending balances. A ratio mixing an income statement flow with an ending balance sheet stock is distorted by growth — a fast-growing company's turnover ratios look worse than they are if computed on ending balances. Use averages, and be consistent.
Definitional inconsistency. Debt, equity, capital, and earnings each have several defensible definitions. Consistency across periods matters more than which definition you chose, and any ratio compared to a covenant must use the covenant's definition, not yours.
Seasonality and period-end management. A period-end balance sheet may have been managed — collections pushed, payments deferred, shipments timed. Averages and same-period-prior-year comparisons address this; sequential period comparisons do not.
EBITDA-based ratios inherit EBITDA's weaknesses. Net debt to EBITDA is the market's standard leverage measure and, per our post on EBITDA versus net income, the denominator excludes interest, taxes, and the cost of the assets generating the earnings.
Lease obligations. The requirement to recognize operating lease assets and liabilities changed leverage ratios structurally. A company's current debt-to-equity is not comparable to its own pre-adoption figures, and comparing a lease-heavy business to a property-owning one requires care in either direction. This is a discontinuity in most multi-year trend analyses and analysts frequently miss it.
Non-recurring items distorting a period's earnings, with the adjustment discipline from our EBITDA post: adjust only what is genuinely non-recurring and documented.
Framework differences. Ratios computed from a tax-basis or other special purpose framework statement are not comparable to ratios from a statement prepared under generally accepted accounting principles — a point our post on year-end statement preparation covers.
Related-party balances inflating apparent assets, and — in a closely held business — owner compensation decisions making margins incomparable between two otherwise similar companies.
Comparability: Choosing What to Compare To
The company's own history is usually the best comparator, and it is free. Twelve or more periods, with the same computation method throughout.
Peer comparison requires genuine comparability: industry, size, capital intensity, and lifecycle stage. A mature company and a growing one in the same industry have systematically different ratios for reasons that are not performance.
Published industry averages deserve real skepticism. The sample composition is usually unknown, the definitions may differ from yours, the data may be dated, and the average of a wide distribution describes no actual company. Use them as a sanity check and never as a target.
Covenant thresholds are the one comparison with a definite answer, and they must be computed the lender's way.
Where Ratios Mislead
Worth internalizing, because each of these looks like good news:
A rising current ratio from inventory that is not selling and receivables that are not collecting.
A high return on equity produced by leverage rather than performance — visible only through the decomposition.
Improving turnover from a shrinking denominator, where a business that wrote down assets or reduced inventory looks more efficient while contracting.
Margin improvement from deferred maintenance or reduced discretionary spending, which is a timing shift rather than an operational gain.
Improving days payable that reflects an inability to pay rather than negotiated terms.
A flattering leverage ratio at a company with substantial off-balance-sheet or contingent obligations.
How to Actually Use Them
Pick six to eight and use them consistently. A short set, trended, outperforms a comprehensive table computed once.
A defensible standard set: the cash conversion cycle, the quick ratio, debt to tangible net worth, debt service coverage, gross and operating margin, total asset turnover, and return on invested capital.
Trend over twelve or more periods, and compare to the same period in the prior year rather than to the sequential period.
Decompose anything that moved, per the DuPont technique.
Then ask the question that prevents wrong conclusions:what would make this number move? Every ratio has several possible causes, and identifying which one applies is the analysis. A ratio that improved for a bad reason is the most common error in this discipline.
Structured coverage is available through Analyzing Financial Statements, the Certificate in Financial Reporting and Analysis, the Business Credit Analysis Bootcamp, the financial statements training catalog, Understanding EBITDA, Fundamentals of Accounting, Essential Excel Skills, and High Impact Excel: Dashboard Edition.
Presenting Ratios So a Client Acts
The failure mode in client reporting is a page of ratios that produces no decision.
Three numbers, not twelve. Choose the three that matter for this client's situation.
Each with its driver. Not "days sales outstanding rose to fifty-two" but "days sales outstanding rose to fifty-two, driven by two customers now beyond ninety days."
And one action. "Call those two customers this week" is a report a client can use. A table is a report a client files.
Plus the trend, visually, because an owner sees a direction faster than they read a number.
Where Ratio Analysis Goes Wrong
- A ratio computed with nothing to compare it to
- Over-weighting liquidity and never computing return on capital
- Never decomposing return on equity, so a leverage-driven improvement reads as performance
- Mixing flows with ending balances rather than averages
- Inconsistent definitions across periods, or a covenant ratio computed the analyst's way
- Sequential period comparison in a seasonal business
- Trending across the lease standard adoption without noting the discontinuity
- Published industry averages used as targets
- Peers selected by industry alone, ignoring size, capital intensity, and lifecycle
- Comparing ratios across reporting frameworks
- Aggregate inventory turns, concealing dead stock
- Treating a rising current ratio as good news
- Not asking what would make the number move before concluding
- Reporting twelve ratios to a client instead of three with a driver and an action
The summary: use a short consistent set, trend it against the company's own history, decompose anything that moved before explaining it — and add return on invested capital, because it asks the only question that determines whether the business is worth the capital in it, and almost nobody computes it.
Frequently Asked Questions
What makes a ratio useful?
The comparison. A ratio computed once for one period, with no prior periods, no comparable peer, and no covenant threshold, tells you almost nothing — the analytical value is in trending it against the company's own history, comparing it to genuinely similar businesses, or measuring it against a defined threshold.
Which ratio is most important and least computed?
Return on invested capital, which asks whether the business earns more on the capital employed than that capital costs. A company earning modest returns on a small capital base can be a much better business than one earning large absolute profits on an enormous one, and no other common ratio asks that question.
What does decomposing return on equity reveal?
Whether an improvement came from margin, asset turnover, or simply more leverage — three drivers with entirely different implications. Margin and turnover improvements are operational achievements; a leverage-driven improvement means the business became riskier while the headline ratio looked better. The decomposition takes two minutes and it is the technique that turns a ratio into a diagnosis.
Why is the current ratio a weak liquidity measure?
Because it rises for bad reasons — inventory that is not selling and receivables that are not collecting both increase it, and it is unaffected by a company that has simply stopped paying suppliers. The cash conversion cycle is the better tool, because it converts balances into a duration and is trendable.
What discontinuity breaks multi-year leverage trends?
The requirement to recognize operating lease assets and liabilities, which changed leverage ratios structurally. A company's current debt-to-equity is not comparable to its own pre-adoption figures, and analysts trending across the adoption without noting it draw wrong conclusions about a change in leverage that was an accounting change.
How should ratios be presented to a business owner?
Three numbers rather than twelve, each with its driver stated — "days sales outstanding rose to fifty-two, driven by two customers now beyond ninety days" rather than the figure alone — plus one action and a visual trend. A table of ratios gets filed; three numbers with a cause and an action get used.
