A latte is espresso with steamed milk poured into it and a thin cap of foam on top. “Coffee”, in this comparison, means the drink with no milk in it — ground coffee and water, brewed by drip, press or pour-over. So the difference is not the coffee: it is that one drink is mostly milk and the other is not. If you want the shortest useful version: a latte is a single or double espresso inside a drink of about 236–340 ml that is otherwise steamed milk, and everything else on a café’s milk-drink list is the same espresso with the total volume and the foam depth changed.
The espresso underneath is the one part of this that anybody has written down. The World Barista Championship’s rules define it as “a beverage (around 1 fl. oz. / 30 ml) made from ground coffee, poured from 1 side of a double portafilter in 1 continuous extraction”, brewed “between 90.5-96 degrees Celsius (195-205 degrees Fahrenheit)” at “between 8.5 and 9.5 bars”, with “Crema… present when espresso is served, with no break in coverage” (2025 WBC Official Rules and Regulations, §3.1). That temperature range is the same one our page on brewing temperature gives for coffee generally, from a different standard.
Table of Contents
Nobody defines a latte, and that is not an oversight
Having pinned down the espresso in that much detail, the same WBC rulebook then defines the milk half like this, in full: “A milk beverage is a combination of 1 single shot of espresso (per the ‘Espresso’ section) and milk steamed with the WBC-provided espresso machine, which should produce a harmonious balance of rich, sweet milk and espresso.” No volume. No ratio. No foam depth.
We searched the full text of both the 2024 and the 2025 WBC Official Rules and Regulations: the word “cappuccino” appears zero times in either, and so does “flat white”. The competition that decides who the best barista in the world is does not name these drinks. So when a page gives you an exact espresso-to-milk ratio for a latte, it did not get it from a standards body, because there is not one to get it from.
What exists instead is trade consensus, and it is openly contested. Perfect Daily Grind ran a whole piece in July 2026 on whether the industry should standardise these sizes at all, and its conclusion was that “The definitions of different milk drinks are shaped by history, geography, and shifting customer tastes rather than by a universal standard” and that “Full standardisation seems unlikely given regional differences in taste and preference” (PDG, “No more ‘large’ flat whites”). The same article reports the drift plainly: a cappuccino in early-1900s Italy was “25ml of espresso topped with 100ml of steamed milk”, and today it is a larger drink.
The espresso-and-milk drinks, side by side
| Drink | What distinguishes it |
|---|---|
| Latte | The most milk of the group and the least foam. Perfect Daily Grind puts today’s version at “between 236ml and 340ml (8oz-12oz), topped with around 0.5cm of microfoam.” |
| Cappuccino | Smaller than a latte and distinctly foamier — PDG describes at least 1 cm of microfoam against roughly 0.5 cm for a latte, and gives the drink as “usually around 180ml (6oz)”. |
| Flat white | Smaller again, with a latte’s shallow foam rather than a cappuccino’s. PDG reports it as “generally around two-thirds milk”. Its origin is disputed between Australia and New Zealand and has never been settled. |
| Cortado | Espresso “cut” with a small amount of warm milk. Small, milk as a correction rather than as the body of the drink, and not sweetened as part of the recipe. |
| Cortadito | The Cuban version, and the sugar is the difference: it goes into the coffee before the milk arrives. Covered in full on what a cortadito is and the four types of Cuban coffee, which this page does not restate. |
One caveat on the cappuccino line. Perfect Daily Grind attributes the 1 cm microfoam figure to “the Specialty Coffee Association and the World Barista Championships”. We downloaded and searched the WBC rules ourselves and, as above, they do not mention the cappuccino at all — so treat that attribution as PDG’s, not as something we have confirmed in an SCA or WBC document.
The “1:2” you will see quoted is not an espresso-to-milk ratio
This trips people up constantly. When a barista says they build a latte on a 1:2 ratio, they mean the espresso: one part dry coffee in the basket to two parts liquid out of it. One of PDG’s interviewees spells out the same recipe in grams — “a 20g dose and 40g yield, extracted in 28 seconds” — which is what 1:2 means here. It says nothing about how much milk goes in afterwards.
It is also a different number from the brew ratio for filter coffee, which is far weaker by design. The Specialty Coffee Association’s home brewer standard specifies “The standard brew ratio of 55 g of coffee per 1.000 kg of 17±3°C water” (SCA Standard 310-2021, Home Coffee Brewers, §7.1.5) — about 1:18 by weight. Espresso at 1:2 and filter at 1:18 are not in competition; they are different drinks. And in both cases, stronger means more coffee, never less. Our page on how many beans per cup of coffee works the filter figure through in grams and beans, and notes that a coffee “cup” is 6 fl oz rather than 8.
Does a latte have less caffeine than a coffee?
An earlier version of this page said yes, and gave a mechanism that is wrong. Its words: “A standard serving of coffee usually has more caffeine than a latte because lattes contain a lot of milk, which dilutes the caffeine concentration.” Its comparison table put coffee at “Generally higher” and latte at “Generally lower”.
Milk does not remove caffeine. It lowers the concentration per millilitre while leaving the dose in the cup exactly where it was, and the dose is what you drink. The caffeine in a latte is set by how many shots went in; the caffeine in a filter coffee is set by the dose of grounds and the serving size. Either can be higher. A two-shot latte and a small filter coffee are not a fixed ordering, and the honest answer is that you have to count shots and ounces rather than drink names. The FDA’s own figures for brewed coffee span more than a two-fold range for a single serving size, which is quoted on our page on what food group coffee is.
Making a latte without an espresso machine
The usual substitute is a stovetop moka pot, which produces a small, concentrated brew. Be clear that it is a substitute: a moka pot does not meet the espresso definition above, because it does not run at 8.5–9.5 bars and it does not produce the same crema. It gets you a coffee strong enough to survive milk, which for a home latte is the part that matters. Heating and frothing the milk separately, then pouring it in, gets you the rest of the way.
Affiliate disclosure, next to the link rather than in a footer. The product link below is a paid link. As an Amazon Associate I earn from qualifying purchases. If you buy through it the site may be paid a commission and your price is unchanged. We have not bought, used or tested this product — it is named because it is the standard example of the type, not as a tested recommendation. The disclosure sits here because the FTC’s Endorsement Guides require that “In any communication using an interactive electronic medium, such as social media or the internet, the disclosure should be unavoidable” (16 CFR 255.0(f)).
Bialetti Moka Express Iconic Italian Stovetop Espresso Maker, 6 Cups, on Amazon
An earlier version of this page carried the same product behind a shortened amzn.to link labelled only “Moka pot”. Shortened links hide where they go and hide which affiliate tag they carry, and that one had no disclosure anywhere near it. It has been replaced with the full product URL above, under the product’s actual name, with the disclosure attached.
Other corrections to an earlier version of this page
- A link whose label did not match its destination. The text “different coffee brewing methods” was linked to the National Coffee Association’s Coffee Roasts Guide — a page about roast levels, not brewing methods. That URL also now redirects, because the NCA has moved its consumer site to aboutcoffee.org. Roast levels are covered on our own page on coffee roast levels.
- Two origin dates with no source. The page stated that coffee’s “inception… traces back to Ethiopia, around the 9th century” and that in “the 20th century in Italy… the delightful concoction known as a latte was born”. Its comparison table then printed “Ethiopia” and “Italy” as bare facts. The latte’s history is not that tidy: Perfect Daily Grind describes it as having “started in 17th-century Europe as plain coffee with milk, then evolved once it reached the US.” The origin row has been dropped rather than replaced with a different guess. For Ethiopia’s actual documented role in coffee, see Ethiopian coffee beans.
- Invented tasting notes. The page told readers that arabica has “notes of sugar, fruit, and berries” and that robusta has “a grain-like overtone and peanutty aftertaste”, and elsewhere assigned each brewing method its own flavour verdict — none of it measured or cited, and nobody at this site tasted any of it. What is actually known about the two species is on arabica and robusta.
- Nutrition claims stated as fact. “Lattes, due to the added milk, contain more calories and a significant amount of protein and calcium” is directionally obvious but was given without any figure or source, and “significant” is doing work it cannot support. The size of the effect depends entirely on which milk and how much.
- A “Conclusion” heading, an FAQ block, and an opening paragraph about the “fascinating journey of beverages”. All removed. The FAQ’s one genuinely useful answer — that you can build a latte on strong brewed coffee if you have no espresso — is in the section above, with the caveat that a moka pot is the closer substitute.
Nobody at this site has pulled a shot, steamed milk or compared these drinks side by side, so there are no tasting notes here that are ours. If you want the same espresso taken in the opposite direction, with hot water instead of milk, see what a long black is. If milk is the part giving you trouble, why almond milk curdles in coffee covers that.
