Ratios

Dividing one number by another is the simplest analytical move. It’s also one of the most powerful.

Why divide

Raw numbers resist comparison. Is £5m EBITDA good? Depends on the size of the business. Is 50 defects acceptable? Depends on how many units you produced. Is £2m in receivables a problem? Depends on your revenue.

Ratios normalise. EBITDA ÷ Revenue gives you margin — now you can compare a £10m business to a £100m one. Defects ÷ Units gives you defect rate — now you can compare factories. Receivables ÷ Revenue gives you receivables days — now you can compare collection efficiency.

The division strips out scale and reveals structure.

Numerator and denominator

Every ratio embeds a choice: what goes on top, what goes on bottom.

The numerator is what you’re measuring. Profit, revenue, defects, customers lost. This is the outcome you care about.

The denominator is “per what”. Per pound of revenue, per employee, per customer, per pound of capital. This is the normalising base.

The same numerator with different denominators answers different questions:

Same profit. Three different lenses. Each focuses attention somewhere different.

Choice shapes behaviour

The denominator you choose shapes the behaviour you get.

EBITA ÷ Working Capital focuses managers on receivables, inventory, and payables — the capital they control day-to-day. Improve collections, reduce stock, manage supplier terms.

ROIC (Profit ÷ All Capital) focuses on total capital efficiency including fixed assets and acquisition premiums. But a general manager can’t do much about the building they occupy or the price the parent company paid to acquire them.

Same business, different ratio, different focus. Serial acquirers like Addtech use EBITA/WC precisely because it directs attention to controllable levers. The ratio design drives the behaviour.

What ratios hide

Ratios collapse information. That’s their power and their danger.

Ratios hide absolutes. A 30% margin looks identical whether the business is £1m or £100m. Sometimes scale matters — a 30% margin on £100m is a very different business from 30% on £1m.

Ratios can be gamed. Improve a ratio by growing the numerator or shrinking the denominator. Revenue per employee rises if you outsource — the productivity looks better, but you’ve just moved the people off your books. The ratio improved; the economics might not have.

Ratios hide composition. A blended 25% margin might be one product at 40% and another at 10%. The average looks healthy; the portfolio has a problem. Aggregate ratios can mask segment-level issues.

Ratios are snapshots. A ratio at period-end might not represent the period. Inventory turnover calculated on December stock looks very different from June stock for a seasonal retailer.

Designing ratios

When you need a metric that doesn’t exist, design it as a ratio. Three questions:

What outcome do you care about? This becomes the numerator. Be specific — “profit” might mean gross profit, operating profit, or cash profit depending on what you’re trying to understand.

What’s the natural base? This becomes the denominator. Usually it’s the resource being consumed (capital, people, time) or the scale of activity (revenue, customers, transactions).

What behaviour do you want? The ratio will focus attention on the relationship between numerator and denominator. Make sure that’s where you want people looking.

A well-designed ratio makes the right trade-offs visible. A poorly designed one hides them.

The practice

Compare like with like. Ratios enable comparison, but only if the definitions match. One company’s “EBITDA margin” might exclude stock-based compensation; another’s might include it. Same label, different calculation.

Watch for denominator manipulation. When a ratio improves, ask whether the numerator grew or the denominator shrank. Both can be legitimate; only one might be.

Pair ratios with absolutes. Margin tells you efficiency. Revenue tells you scale. You usually need both. A 40% margin on declining revenue is a different story from 40% on growing revenue.

Track trends, not just levels. A ratio moving from 20% to 25% tells you more than a snapshot at 25%. Direction reveals whether things are improving or eroding.

Ratios are how you turn raw numbers into insight. The division is simple. The choice of what to divide — and what to divide by — is where the thinking happens.

Connects to Library: Goodhart’s Law — The denominator you choose shapes the behaviour you get.