What margin of safety really means

The phrase gets thrown around so often it stops meaning anything specific. Margin of safety. Everyone nods. Almost nobody could tell you the number they're using, or why they picked it.

Start with the part that gets skipped: an intrinsic value estimate is not a fact. It's a model's best guess, built from assumptions about growth, margins, and a discount rate, all of which could be wrong. Aperto doesn't even publish a single point estimate most of the time; it publishes a range, because the honest answer to "what is this worth" is rarely one number. A margin of safety is the buffer you build in for the model being wrong, not a bonus you collect for being clever.

The discount isn't the reward. It's the insurance.

Say a company's estimated intrinsic value is $100 and it trades at $90. That's a 10% discount. Is that enough? It depends entirely on how confident you are in the $100. If the valuation rests on a stable, well-understood business with predictable cash flows, 10% might be plenty of room. If it rests on a handful of aggressive growth assumptions holding up for a decade, 10% probably isn't a margin at all — it's rounding error against how wrong the estimate could plausibly be.

This is why Aperto shows the discount at several thresholds rather than one: what you'd pay for a 0% margin (the raw estimate), 10%, 15%, 25%. The gap between those numbers on a given stock tells you something concrete. A business whose price barely has to move to clear a 25% margin is not obviously cheap, it's obviously misunderstood, in one direction or the other.

Why a bigger number isn't automatically the right answer

It's tempting to think the safest rule is always demand the biggest discount you can get. That's wrong for a specific reason: the businesses that trade at deep discounts to a defensible valuation are often deeply discounted for a reason the model hasn't captured. A 40% margin of safety on a company whose moat is quietly eroding isn't a bargain. It's the market pricing in something the valuation lenses haven't caught up to yet.

That's the whole reason margin of safety comes after the other layers, not before them. Aperto asks whether a business deserves a moat rating first. Only once that case holds does the size of the discount mean what you think it means. A margin of safety on a business you haven't actually vetted is just a number that happens to be smaller than another number.

The honest version

Margin of safety works because it forces you to write down, in advance, exactly how wrong you're allowed to be and still come out fine. Skip that step and you're not investing with a margin of anything. You're hoping the model was right on the first try.