Guides

A chocolate wrapper carries three kinds of statement — legally controlled, conventional, and pure marketing — and they look identical.

Takeaway: The ingredients list and the percentage are regulated and informative. Almost everything else on the front of the pack is not, and the difference is invisible unless you know which is which.

Most of it is made rather than grown. Fermentation creates the precursors, roasting converts them, and conching removes what is left over.

Takeaway: Fermentation, not origin or variety, is the largest single variable in how chocolate tastes — and it happens on a farm, days after harvest, long before any chocolate maker is involved.

A repeatable method, and — more usefully — how to tell a genuine characteristic from a manufacturing fault.

Takeaway: Let it melt rather than chewing. Chocolate's aromatics are released by body heat, and chewing sends it past the palate before most of them have appeared.

The right chocolate depends on what it has to do — coat, set, melt into a batter or hold its shape — and those requirements pull in different directions.

Takeaway: Match the chocolate to the job, not to the price. Couverture for coating, ordinary dark for baking, chips only where you want distinct pieces, and never compound where the chocolate will be tasted on its own.

A food made from fermented, roasted and ground cacao seeds, sugar and usually milk — and, unusually, a category whose names are defined in law.

Takeaway: Chocolate is ground cacao seed plus sugar, held together by cocoa butter. Everything else — the type, the percentage, the origin — is a variation on those three components.

A tropical tree, its pods and its seeds. Cacao is the raw material; chocolate is what processing turns it into.

Takeaway: Cacao is a plant and chocolate is a manufactured food. Everything in between — fermenting, drying, roasting, grinding — is what separates them.

There is no technical or regulatory distinction. The difference is marketing convention, and the health implication attached to it is not supported.

Takeaway: Neither word tells you anything checkable about processing or nutrition. What does matter — alkalisation and fat content — is a separate declaration that either word can conceal.

The combined mass of everything that came from the bean — cocoa solids plus cocoa butter — as a share of the bar. It is a statement about proportion, not about quality.

Takeaway: The percentage tells you how much of the bar is not sugar. It does not tell you how that cacao mass splits between solids and butter, which is much of what determines whether a bar tastes rich or harsh.

Sixteen steps across two continents, and the four that matter most happen on a farm before any chocolate maker is involved.

Takeaway: Fermentation is the largest single variable in how chocolate tastes, and it happens at origin within days of harvest — long before the beans reach a factory.

Cocoa butter melts just below body temperature — an unusual property for a natural fat, and the single physical fact most of chocolate's behaviour follows from.

Takeaway: Cocoa butter melts sharply just under 37°C. That is why chocolate is solid in your hand, liquid in your mouth, and why substitute fats never feel right.

Cocoa solids, cocoa butter and sugar with no milk. The type where the beans are audible, and where percentage choice actually changes the experience.

Takeaway: Start at 65–70%. It is where most well-made dark chocolate sits, and where origin character is clear without bitterness dominating.

Dark chocolate's components plus milk solids — and the one category where American and European products carrying the same name are genuinely different.

Takeaway: Milk chocolate's legal minimum cacao content is far lower in the US than in the EU, and the milk-crumb process used by several long-established manufacturers gives a distinctly cooked, tangy flavour.

Yes, legally, in both the EU and the US. The fair criticism is not about legitimacy but about deodorised cocoa butter.

Takeaway: White chocolate is legally defined and contains a real cacao ingredient — cocoa butter. Most of it tastes of nothing much because the butter has been deodorised, which is a quality problem rather than a definitional one.

A pink, tart chocolate launched commercially in 2017. No colouring or fruit is added — and the process that achieves that is proprietary.

Takeaway: The colour and the berry tartness come from bean selection and processing rather than from added ingredients. Beyond that, the mechanism is not in the public literature and this guide does not reconstruct it.

Chocolate with a high cocoa butter content, made to be melted and used as a coating. If your home tempering never works, this is usually why.

Takeaway: Couverture means high cocoa butter, not high quality. You need it for coating, moulding and enrobing, and you do not need it for baking.

A claim about who did the manufacturing. It means the maker roasts, grinds and refines the beans themselves rather than remoulding chocolate someone else made.

Takeaway: Bean-to-bar tells you who made the flavour decisions. It does not tell you whether those decisions were good ones, and it is not the same claim as single-origin.

Read the ingredients before the front of the pack, and match the bar to what you are going to do with it.

Takeaway: A short ingredients list with cocoa mass first, a stated percentage, and a specific origin claim are the three signals that carry real information. Everything else on the front is marketing.

An ordinary dark bar at a stated percentage, chopped — with the recipe's assumptions checked first, because 'baking chocolate' means two different things.

Takeaway: Match the sugar. A recipe built on unsweetened chocolate and made with a 55% bar gains roughly 45g of sugar per 100g of chocolate, which is enough to change the result.

65–75% for eating dark, 60–70% for baking, 35–45% for milk. But the number is a preference guide rather than a quality ranking.

Takeaway: There is no best percentage, only a best percentage for a purpose and a palate. Two bars at the same number can differ by ten percentage points of cocoa butter, which matters more than the number does.

No. A higher percentage means less sugar and more cacao, which is a statement about proportion — and it exposes bad beans as readily as good ones.

Takeaway: Percentage measures proportion, not quality. The number cannot distinguish a carefully fermented bean from a badly fermented one, and at high percentages it amplifies both.

Melt fully, cool while stirring in tempered seed, then hold at working temperature. A thermometer is not optional.

Takeaway: Use the seeding method, use couverture, and use a thermometer. Two of the three most common failures are materials problems rather than technique problems.

It controls which of cocoa butter's six crystal forms the chocolate sets in. Skip it and the chocolate is soft, dull, and blooms within days.

Takeaway: Tempering has no effect on flavour and total effect on everything else — snap, gloss, mould release, melt and shelf stability.

Two failures, both irreversible and both counterintuitive: a small amount of water seizes it, and modest heat scorches it.

Takeaway: A drop of water ruins melted chocolate; a lot of water does not. That is why a seized batch can be rescued as a sauce but never as melted chocolate.

Five common failures, what each one is telling you, and which of them are materials problems rather than technique problems.

Takeaway: Streaky, soft and won't-release are all the same fault presenting differently — the wrong crystal form. Grainy and seized are different problems entirely.

It is safe. Bloom is cocoa butter or sugar that has migrated to the surface — two different phenomena with two different causes, neither of them mould.

Takeaway: Bloomed chocolate is edible. Fat bloom comes from warmth or poor tempering; sugar bloom comes from moisture, usually the fridge. Both are texture problems rather than safety ones.

Steady 15–18°C, wrapped, dry, away from strong smells, and not in the fridge.

Takeaway: Steady temperature matters more than cool temperature. Cycling between warm and cool causes more bloom than a constantly slightly-too-warm cupboard.

Solid chocolate carries a best-before date rather than a use-by date, and the distinction is a legal one about quality rather than safety.

Takeaway: Plain solid chocolate has very low water content, so microbial spoilage is essentially not a risk. What degrades is flavour and texture, which is what a best-before date describes.

You can, and it will usually bloom unless you manage the condensation. For most purposes there is a better option.

Takeaway: Freezing does not damage chocolate; thawing does. Condensation on the cold surface causes sugar bloom, which is why the wrapping and the thawing procedure matter more than the freezing.

No. It is bloom, and telling the two kinds apart tells you what went wrong and how to prevent it.

Takeaway: Smooth and slightly greasy is fat bloom, from warmth. Rough and gritty is sugar bloom, from moisture. Both are safe; neither is mould.

The most natural partner and the easiest to get wrong, because both are roasted, bitter and acidic — so one has to give way.

Takeaway: Only one partner should be bright. A fruity acidic coffee with a fruity acidic chocolate reads as thin and sour, however good both are individually.

A wider range than coffee and a more forgiving partner, as long as the tea's tannin is kept in check.

Takeaway: Match weight, and brew slightly lighter than usual. Tannin is the variable most likely to break the pairing and the one most easily controlled.

The most recommended pairing in food writing and one of the least reliable. The classic dry-red-with-dark-chocolate combination fails for three specific reasons.

Takeaway: The wine must be at least as sweet as the chocolate and preferably lower in tannin. The standard recommendation violates both conditions.

Better than its reputation, and structurally sound — both are fermented, both are fatty, and salt is the bridge.

Takeaway: Salt in the cheese suppresses chocolate's bitterness, exactly as a pinch of salt does to any sweet thing. Match intensity and the pairing works far more often than it fails.

Contrast the acidity rather than matching it — and use peel and zest rather than juice, because the aromatics are fat-soluble and the juice is not.

Takeaway: Sharp fruit wants round chocolate. The common mistake is pairing an acidic fruit with an acidic single-origin bar, which produces sourness rather than brightness.

Cocoa powder for the crumb, chocolate for richness — and the choice between natural and Dutch-process determines which leavening the recipe needs.

Takeaway: Natural cocoa pairs with bicarbonate of soda; Dutch-process pairs with baking powder. Swapping one for the other without changing the leavening gives a flat bake or a soapy taste.

Chopped bar rather than chips, at 60–70%, with the recipe's sugar assumptions checked — brownies are where substitution errors show most.

Takeaway: Brownies are mostly chocolate, sugar and fat, so a chocolate substitution changes a large fraction of the recipe. The sugar arithmetic matters more here than anywhere else.

Most substitutions work if you adjust the sugar and the fat. Two do not work at all, and it is worth knowing which.

Takeaway: Substituting chocolate changes sugar, fat and cocoa solids simultaneously. Adjust the sugar first — it is the largest and most predictable difference.

Because it is thickened with starch. Four distinct traditions, and knowing which one you are trying to make is most of the problem.

Takeaway: Thickness comes from starch, not from more chocolate. Richness comes from cocoa butter, which is why a chocolate-based drink beats a powder-based one for body.

Match the recipient's existing taste rather than trying to educate it, and remember that filled chocolates have a shelf life measured in weeks.

Takeaway: The most common gifting mistake is buying a very high percentage bar to signal quality. If the recipient enjoys milk chocolate, a good milk chocolate is a better present than an 85% single-origin bar.

Four bars, one variable, room temperature. The comparison is what teaches you, not the individual bars.

Takeaway: Hold everything constant except one thing. Four bars at the same percentage from different origins teaches more than four bars chosen at random ever will.

Three questions — sweetness, acidity and intensity — will narrow the field faster than any amount of reading about origins.

Takeaway: Acidity is the axis that divides people most sharply and the one nobody asks about. If you have disliked expensive single-origin chocolate, acidity is the most likely reason.

Place affects flavour, but far less directly than in wine — and most of what gets attributed to terroir is actually fermentation.

Takeaway: Place matters, but fermentation matters more. A region name is worth far more than a country name, and both are worth less than knowing how the beans were processed.

The three-variety model is a useful simplification that modern genetics has largely superseded — and it never predicted flavour on its own.

Takeaway: The three-variety scheme is a simplification you will meet everywhere, and variety sets potential rather than outcome. Processing decides what is actually in the bar.

An unregulated claim that can describe one farm or one country. The precision of the claim is what carries the information.

Takeaway: Single-origin is unregulated. Judge it by how specific the claim is — farm beats region beats country — and remember a deliberate blend can be the better bar.

A band roughly twenty degrees either side of the equator, and the constraints that define it are narrow enough to explain the whole geography of chocolate.

Takeaway: Cacao grows in a narrow equatorial belt because it tolerates no frost and needs constant humidity. Where it grows most is a product of colonial trade history rather than natural range.

West Africa grows most of the world's cacao and receives the least of its attention. The reasons are structural rather than agricultural.

Takeaway: West Africa grows most of the world's cacao and sells nearly all of it as an untraceable commodity. That is a supply-chain fact, not a quality one — the beans make excellent chocolate when someone separates and ferments them properly.

A small share of world volume and most of its fine-flavour reputation — because this is where the genetic diversity actually is.

Takeaway: Latin America holds cacao's genetic diversity, which is why its flavour range is wider. The prestige is real and rests on scarcity and genetics rather than on any inherent superiority of the region's farming.

The fastest-changing producing region, with one shared quality problem that has defined its reputation for decades.

Takeaway: The region's aromatic potential is real and its historic reputation problem is smoke taint from wood-fired drying — a solvable equipment problem, and where it has been solved the chocolate is excellent.

Four causes, in descending order of influence: fermentation, roast, genetics, growing conditions. Most writing gets the order backwards.

Takeaway: Fermentation is the dominant cause of origin character, roast is second, genetics third and growing conditions last. Almost all popular writing reverses that order.

A trade classification, administered country by country, that is about market category rather than a measurable property of any given bean.

Takeaway: Fine-flavour is an ICCO country-level trade classification tracking genetics and processing convention. It signals potential, and says nothing certain about any particular bar.

Most origins have a main crop and a mid crop, and they are not the same. Craft bars increasingly print the harvest year for good reason.

Takeaway: Main and mid crops differ, and years differ. Estate and single-harvest bars are effectively vintages, which is why a printed harvest year carries real information.

The suitable growing belt is projected to shift. That is the structural pressure behind recent cacao prices, alongside disease and ageing trees.

Takeaway: Cacao's suitable belt is projected to shift rather than vanish, and current price pressure comes from disease and ageing trees as much as from climate. The effects are already visible in what is on the shelf.

A country can span several climates, genetic populations and fermentation conventions. A region usually spans one of each.

Takeaway: A region implies one climate, one genetic base and often one fermentation protocol. A country implies none of those, which is why regional labels are worth far more.

Two of the most abused words in chocolate marketing, with a small number of cases where they are literally true.

Takeaway: Both terms are unregulated. Genuinely wild cacao exists — Bolivian Beni is the clearest case — and genuine heirloom claims come with genetic characterisation named on the packaging.

The structural facts behind ethical chocolate claims: smallholder farms, commodity pricing, and a farmgate share far below what most buyers assume.

Takeaway: Cacao is a smallholder crop sold into a commodity chain, and the farmgate share of a mass-market bar's retail price is in the low single digits. Certification and direct trade both address this partially and neither is a guarantee.

A method for hearing the bean rather than the recipe: same percentage, same maker, three or four origins, room temperature.

Takeaway: Hold the maker and the percentage constant and vary only the origin. Madagascar against Ecuador at the same percentage from one maker is the clearest first lesson available.

A three-phase microbial succession over five to seven days that creates essentially all of chocolate's flavour precursors. Nothing downstream can replace it.

Takeaway: Fermentation creates the precursors roasting turns into flavour. Get it wrong and no later step can compensate — it is the single highest-leverage stage in chocolate production.

Maillard chemistry converting fermentation's amino acids and reducing sugars into hundreds of aroma compounds — and destroying some in the process.

Takeaway: Roasting converts fermentation's precursors into aroma via Maillard chemistry. Light preserves origin character and exposes defects; dark builds cocoa depth and hides them. Neither is better in the abstract.

Two jobs in one machine — driving off harsh volatiles and coating particles in fat. Longer is not automatically better, and it can flatten a bar.

Takeaway: Conching drives off harsh volatiles and coats particles in fat. Longer gives smoother and blander — smoothness and distinctiveness trade against each other directly.

The tongue stops detecting individual particles somewhere around 25–30 microns. Everything about chocolate texture follows from that threshold.

Takeaway: Smoothness means getting particles below about 25 microns. Finer is not automatically better — it costs cocoa butter and can read as pasty — and apparent grittiness is often sugar bloom instead.

Cocoa butter is polymorphic — it can solidify six different ways, and only one of them makes good chocolate. Tempering is the management of that fact.

Takeaway: Cocoa butter crystallises six ways and only Form V gives snap, gloss and clean release. Tempering is the procedure for making Form V dominate, and bloom is the unstable forms converting later.

Sixteen stages across two continents and several months. Knowing the order makes every individual step easier to place.

Takeaway: Fermentation and roasting make flavour, refining and conching make texture, tempering makes appearance and snap. Most faults map cleanly onto one stage.

The most expensive ingredient in chocolate and the one doing the most work — melt, flow, snap, mouthfeel and shelf life all sit on it.

Takeaway: Cocoa butter controls melt, flow, snap and mouthfeel, and it is the expensive part. The cocoa percentage on a bar does not tell you how much of it there is.

A processing aid used at well under 1% that replaces several times its weight in cocoa butter. Its presence is not a quality signal.

Takeaway: Lecithin at 0.3–0.5% replaces several times its weight in cocoa butter. Its presence signals a cost-conscious formulation, not a bad one.

Treating cocoa with an alkali to raise its pH — which changes colour, flavour, solubility and its behaviour with baking soda.

Takeaway: Alkalisation raises cocoa's pH, deepening colour and smoothing flavour at the cost of flavanols — and it removes the acidity that baking soda needs to work.

The quiet step whose failure shows up as harshness three stages later — and the biggest practical obstacle to making chocolate at home.

Takeaway: Winnowing removes the harsh, fibrous 12% of the bean that refining cannot fix. It is also the step that stops most home chocolate-making, which is why buying nibs is the sensible shortcut.

Getting from 60% moisture to 7% without mould, case hardening or smoke. Three failure modes, all irreversible, all common.

Takeaway: Drying must reach 6–7% moisture slowly. Case hardening, mould and smoke taint are the three failures, all irreversible, and indirect heating or a solar dryer eliminates the worst of them.

Water is chocolate's enemy, so milk enters as powder or as crumb. Which route is taken changes the flavour substantially.

Takeaway: Milk enters as powder or as crumb, never as liquid. The crumb process cooks milk, sugar and cocoa together and is why British milk chocolate has its distinctive caramelised character.

Not shortcuts — different equipment solving different problems. Scale enables things craft cannot do and forecloses things it can.

Takeaway: Scale swaps lot expression for consistency, using different equipment rather than worse. Craft's advantage is specificity; industry's is that it never fails.

A diagnostic list: what each off-flavour tastes like, which stage produced it, and whether anything downstream could have saved it.

Takeaway: Astringency and sourness point at fermentation, smoke and must at drying, ash at roasting, grit at refining, and dullness at tempering. Only the last is fixable after the fact.

Yes, with one genuine bottleneck and one piece of equipment you cannot improvise. Starting from nibs skips both.

Takeaway: Winnowing is tedious and grinding needs a melangeur — everything else is achievable. Start from nibs, which removes both obstacles and most of the cost.

The ratio depends on the chocolate type, not on the recipe. Dark 2:1, milk 3:1, white 4:1 — and most ganache failures are this and nothing else.

Takeaway: Dark 2:1, milk 3:1, white 4:1 by weight for a firm ganache. A loose ganache is almost always a ratio error, and chilling will not fix it.

Counterintuitively, the fix for too little water is more water. Seized chocolate cannot go back to smooth melted chocolate, but it makes an excellent sauce.

Takeaway: Add a tablespoon of boiling liquid per 100g and it becomes a smooth sauce. It will never be melted chocolate again — change the plan rather than fighting it.

Four causes with four different fixes: no temper, wrong chocolate, too warm, or something added that softened it.

Takeaway: Soft chocolate is untempered, the wrong product, too warm, or has something added. Diagnose which before remelting, because three of the four are not tempering problems.

Possible by seeding and by lip test, with a real accuracy cost. Worth knowing, and worth buying a thermometer anyway.

Takeaway: Seeding plus the lip test gets close enough for coating. Always verify with a test strip — three minutes, glossy, no streaks — and buy a thermometer when you can.

Decide first whether you want molten pools or distinct pieces. That answer picks the product; everything else is detail.

Takeaway: Chips hold their shape, chopped bar melts into pools. Chopped 60–70% bar gives better flavour and a range of piece sizes; flaked salt on top is the cheapest upgrade available.

Natural cocoa is acidic and activates baking soda. Dutch-process is not. Swapping them without adjusting the leavening flattens the cake.

Takeaway: Natural cocoa activates baking soda; Dutch-process does not. Swap the cocoa and you must swap the leavening — roughly three parts baking powder for one of soda.

The microwave is faster and safer than its reputation. The bain-marie's advantage is control, and its risk is the steam.

Takeaway: Microwave in short bursts with stirring, or bain-marie with the bowl clear of the water and no lid. Stay under 50°C for dark and 45°C for milk and white.

Above about 24°C, chocolate work gets progressively harder and then impossible. Some of it is manageable and some of it is not.

Takeaway: Chocolate needs a room below about 20°C to set well. Work early, use a cold slab, set in the coolest room you have, and avoid the fridge — condensation causes sugar bloom.

Sugar and fat are what change when you swap percentages. Adjust those two and most substitutions work.

Takeaway: A percentage swap is a sugar and fat swap. Adjust both and most baking substitutions work; setting applications need the ratio recalculated instead.

Curls, shards, piped shapes and transfer sheets — four techniques with wildly different difficulty, and the easiest looks the most impressive.

Takeaway: Shards give the best appearance-to-effort ratio and transfer sheets give the most professional finish. All four techniques require properly tempered chocolate and none work without it.

Three failures — lumps, collapse and grease — with three different causes, and only one of them is about folding technique.

Takeaway: Lumps are a temperature mismatch, collapse is over-whipping or over-folding, and grease is over-chilling. Only the second is about technique in the way people assume.

Cool, dry, dark, sealed and away from anything aromatic. The fridge fails on two of those five.

Takeaway: Cool, dry, dark, sealed, away from smells. The fridge is humid and full of odours, so use it only if the room is above about 25°C — and wrap airtight.

Real chocolate rather than powder, milk heated but not boiled, and a pinch of salt. Thickness comes from starch, not from more chocolate.

Takeaway: Use chopped bar chocolate at 40–60g per 200ml, heat the milk to 70°C without boiling, whisk rather than stir, and add a pinch of salt. Thickness needs cornflour.

Skip tempering. Compound coating and a handful of forgiving techniques give good results without the one step that reliably ruins the afternoon.

Takeaway: Compound coating removes the tempering problem entirely. Bark, dipped fruit and silicone moulds are the four projects that reliably work.

Dark keeps for years, milk and white for months, and filled chocolates for days to weeks. The variable is water and milk fat, not the chocolate.

Takeaway: Dark keeps for years, milk and white for six to twelve months, cream-filled chocolates for weeks. Bloom is not spoilage; a rancid smell in milk or white chocolate is.

Order is by weight, which makes the first two ingredients the most informative thing on the packet — and more informative than the percentage.

Takeaway: Check what comes first and check for vegetable fat. A short list signals intent, not quality, and nothing on the label reveals the four things that actually determine flavour.

Minimum compositions differ by jurisdiction, which is why the same product can be chocolate in one country and not in another.

Takeaway: The EU sets 35% cocoa solids for chocolate and 31% cocoa butter for couverture. Nearly every term a buyer cares about — craft, single-origin, raw — is undefined anywhere.

It is a mass fraction of cacao-derived ingredients, and it says nothing about the split between solids and butter, the beans, or the roast.

Takeaway: The percentage predicts sweetness and nothing else. It hides the solids-to-butter split, and the best chocolate mostly sits around 70% rather than at the top of the scale.

Start with families rather than specific notes, and learn the four pairs of terms that are constantly confused.

Takeaway: Start with ten flavour families and learn the four confusable pairs. Write something down every time — the discrimination comes from repetition, not from reading a wheel.

One variable, four bars, mild to intense. Six flight designs that each teach one specific thing.

Takeaway: Vary exactly one thing across four bars, taste mild to intense at room temperature, and do not tell people what to expect before they taste.

Judge the disclosure, not the adjectives. A maker who names the farm, the harvest and the percentage is telling you something checkable.

Takeaway: Pay for disclosure rather than adjectives: farm, harvest, variety, short ingredient list. Start at 70% with three origins from one maker, and expect to dislike some of them.

Most chocolate health reporting describes studies on cocoa flavanol extracts at doses no bar delivers. Knowing that resolves nearly all of it.

Takeaway: Most cocoa research uses flavanol extracts at doses ordinary bars do not deliver, and processing degrades flavanols. Treat any eat chocolate for your health headline as a conversion of association into instruction.

Chocolate contains both, in much smaller amounts than coffee — and theobromine is the reason chocolate is dangerous to dogs.

Takeaway: Chocolate contains theobromine and a smaller amount of caffeine, both scaling with cocoa solids. The consequential fact is that theobromine is toxic to dogs at doses harmless to people.

Cacao itself is rarely the allergen. Milk, soy, nuts and cross-contamination account for nearly all chocolate allergy problems.

Takeaway: Cacao is rarely the allergen — milk, soy, nuts and cross-contamination are. Dark chocolate is not reliably dairy-free, and may-contain labelling on nuts should be taken seriously.

Four thousand years as a drink, and less than two hundred as a bar. Nearly everything familiar about chocolate is Victorian.

Takeaway: Chocolate was a bitter drink for most of its four-thousand-year history. The bar, milk chocolate and conching are all Victorian inventions from a single fifty-year window.

National chocolate cultures differ for specific, traceable reasons — process, regulation and taste conventions, not vague national character.

Takeaway: National differences trace to specific causes — the crumb process in Britain, long conching in Switzerland, butyric acid in the US, and a preserved drinking tradition in Mexico.

Fairtrade, Rainforest Alliance, organic and direct trade make different claims, verified differently. None is a guarantee and none is meaningless.

Takeaway: Fairtrade, Rainforest Alliance and organic certify different things and none covers the others. Published prices and named producers are a stronger signal than any logo, and direct trade is transparent but unverified.

Six persistent beliefs, what is actually true, and why each one survives — usually because it contains a fragment of something real.

Takeaway: Most chocolate myths contain a fragment of something true — white chocolate really has no cocoa solids, bloom really is a change. The error is in what gets concluded from the fragment.

Match the gift to what you know about the person. When you know nothing, a milk-and-dark range beats an impressive single bar.

Takeaway: A range beats an impressive single bar when you do not know their taste. A comparison flight from one maker is the best gift for someone who already likes chocolate.

An ordered path from first comparison to informed buying, with the mistakes that waste the most money marked.

Takeaway: Start by tasting two bars side by side, learn the four confusable concepts, then run flights that vary one thing. Do not start at 90%, and do not bake with the expensive bars.

Start from the symptom. Eleven common failures, what each one actually is, and whether it can be undone.

Takeaway: Water causes seizing, heat causes scorching, crystallisation causes dullness and bloom. Tempering failures are fully recoverable, seizing is recoverable into a sauce, and overheating is not recoverable at all.

A little water dissolves sugar surfaces and glues them together. A lot of water dissolves the sugar completely. The danger zone is in between.

Takeaway: Seizing is sugar particles wetted just enough to stick together. Adding a tablespoon of boiling liquid per 100g dissolves them fully and gives a sauce — which is a different product, not a repaired one.

Four different causes produce the same sensation, and only one of them is anything to do with the chocolate being badly made.

Takeaway: Grain through a whole bar is refining; grain on the surface only is sugar bloom; grain that appeared in your bowl is seizing or scorching. Break a piece to tell which.

Both look like white marks and neither is mould. One is fat migrating outward, the other is sugar recrystallising — and only one can be reversed.

Takeaway: Fat bloom is greasy and blotchy and comes from crystal migration; sugar bloom is sandy and even and comes from moisture. Neither is mould and both are perfectly usable for melting or baking.

Gloss comes from Form V crystals and, for moulded work, from the mould surface. Dullness means one of those two is wrong.

Takeaway: Dull plus soft plus a crumbly break means poor temper. Dull with a good snap means the mould surface, condensation or handling — not the crystallisation.

Usually the product rather than the technique. Cocoa butter content decides viscosity, and no amount of heat compensates for not having enough.

Takeaway: Thick melted chocolate is usually the wrong product rather than the wrong technique. Cocoa butter thins it correctly; oil and cream thin it and destroy the temper.

A less common problem than the opposite, and usually caused by something having been added — which cannot be taken out again.

Takeaway: Too warm is recoverable by cooling; added cocoa butter is correctable by adding more chocolate; added oil or butter cannot be undone and the batch will not set hard.

A broken emulsion, not a ruined one. Almost every split ganache can be rebuilt, and the method is counterintuitive.

Takeaway: A split ganache is a broken emulsion and is nearly always recoverable: add a spoonful of warm liquid and stir from the centre outward, or use a stick blender.

Six causes, and the two most common are invisible: the melt was not hot enough to start with, and the room is too warm to finish in.

Takeaway: Check the melt reached 45–50°C and the room is below 20°C before blaming the technique. Those two account for most failures that follow apparently correct temperatures.

Release is caused by contraction, and contraction is caused by temper. A piece that will not come out was never in temper.

Takeaway: Release comes from contraction and contraction comes from temper. Chill briefly to force it, never prise with a knife, and never wash a polycarbonate mould in detergent.

The one chocolate failure with no recovery. Sugars caramelise and milk solids scorch, and both are chemical changes.

Takeaway: 50°C for dark, 45°C for milk and white. Past those the change is chemical and permanent — unlike seizing, there is no rescue, only a decision about where the damaged chocolate can hide.

Visible fat pooling on the surface or at the edge of a filling. Three causes, two of them recoverable.

Takeaway: Separation in a ganache is a broken emulsion and is recoverable. Separation in melted chocolate is added fat or scorching. On a stored bar it is bloom, and needs nothing.

There are four kinds of answer to 'what type is this', and consumer writing lists them as if they were alternatives.

Takeaway: Composition, grade, sourcing and format are four separate questions. A bar answers all four at once, and treating them as one list is the source of most confusion about chocolate types.

The same plant as cacao, under a word that has come to mean the processed and defatted form — and, in marketing, the unglamorous one.

Takeaway: Cocoa and cacao are the same plant. Modern usage puts cacao before processing and cocoa after, but the words carry no nutritional difference of their own.

70% is where most makers do their best work. Above 85% the trade-offs change character, and not in the direction most people expect.

Takeaway: Buy at 70% from a maker who names a region and lists few ingredients. Higher percentage raises the bar the beans must clear rather than guaranteeing a better one.

The cocoa percentage is the whole decision, and 25% and 55% are barely the same product.

Takeaway: Look at the percentage, and try dark milk at 50–55%. The gap between a 25% bar and a 55% bar is larger than the gap between many dark chocolates.

Cocoa butter content is the only variable that matters, and the cheap products replace it with something else entirely.

Takeaway: Cocoa butter must be the first ingredient and ideally 30%+. Anything built on vegetable fat is not white chocolate and is where the category's bad reputation comes from.

Origin is the best available proxy for flavour, and it is a weak one. Here is how weak, and how to use it anyway.

Takeaway: Origin is a real but weak predictor. Use it to choose a direction, hold the maker constant when comparing, and trust your own notes over any map.

Couverture if you have it, chopped bar if not, never chips — and the ratio changes with the type, which is where most ganache goes wrong.

Takeaway: Couverture or a 60–70% bar, never chips. Then get the ratio right for the type — dark 2:1, milk 3:1, white 4:1 — because that is the failure everyone has.

A chopped bar rather than powder, because body comes from cocoa butter — and powder has had most of it pressed out.

Takeaway: Chopped bar at 40–60g per 200ml, around 70%, whisked into milk heated to 70°C. Thickness needs cornflour, not more chocolate.

Two different chocolates for two different jobs: the centre needs flavour, the shell needs cocoa butter.

Takeaway: A good 60–70% bar for the centre and couverture for the shell. Dust in cocoa instead if you have no couverture — it is the better first project anyway.

Cocoa solids set the foam, so the percentage decides whether the mousse sets at all — and milk and white need substantially more chocolate.

Takeaway: 65–75% dark is the default because cocoa solids are what set the foam. Milk and white need substantially more chocolate, and the usual failure is a temperature mismatch rather than the chocolate itself.

The one application where the product genuinely decides the outcome, because coating thinly is a cocoa butter problem.

Takeaway: Use couverture. If you cannot, compound coating is an honest trade for decorative work — but a normal eating bar will not coat thinly however carefully you work.

Two decisions: natural or Dutch-process, and how much fat. The first changes your leavening; the second changes the crumb.

Takeaway: Match the cocoa to the leavening — natural with soda, Dutch-process with powder — and pay for 20–24% fat where richness matters.

One is defined by law and the other is a shelf position. Which you need depends entirely on whether the chocolate has to flow.

Takeaway: Couverture for anything that has to flow and set. Ordinary chocolate for anything baked. Check what 'baking chocolate' means in your country before substituting, because it means two different things.

Most of what makes chocolate good is free to check, and most of what makes it expensive is invisible in a recipe.

Takeaway: Do not bake with expensive chocolate. Check the first ingredient and the fat. Dominican Republic bars are the best value in fine chocolate.

Vary one thing. Four bars maximum. The selection matters more than anything you do on the night.

Takeaway: Four bars, one variable, same maker. Madagascar against Ecuador at 70% is the clearest first lesson available, and nothing above 85%.

An unregulated term describing a sourcing model and a scale, not a quality grade — and the model is the substantive part.

Takeaway: Craft describes a sourcing model and a scale, not a quality grade. Judge the disclosure — farm, harvest, variety, ingredients — rather than the word.

Ground cacao nibs, containing no alcohol whatsoever. The name is a translation accident that has misled readers for a century.

Takeaway: Chocolate liquor is ground nibs and contains no alcohol. Cocoa mass and cacao paste are the same thing, and it is the fork point between chocolate and cocoa powder.

Cracked roasted bean with the shell removed — chocolate before sugar, and the most useful thing in the catalogue for a home maker.

Takeaway: Nibs are roasted, cracked, shell-free bean. Use them for crunch or, better, infused into cream — and never as a substitute for chocolate in a recipe.

Cocoa treated with an alkali to raise its pH. It changes colour, flavour and — the part that matters in a recipe — how it behaves with baking soda.

Takeaway: Alkalisation deepens colour, smooths flavour and removes the acidity baking soda needs. Swapping cocoas means swapping the leavening too.

Untreated cocoa powder — acidic, fruitier, reddish, and the only kind that will activate baking soda.

Takeaway: Natural cocoa is untreated, acidic, fruitier and redder. It is the only cocoa that activates baking soda, and it retains more flavanols than alkalised cocoa does.

Not the most expensive bar, and not 85%. A dark milk at around 55% converts more people than anything else on a shelf.

Takeaway: Start with a dark milk around 55%, then buy two bars from one maker at one percentage from different origins. Avoid 80%+ until you know what you like.

The number predicts sweetness and nothing else. Here is what each band actually feels like.

Takeaway: 65–75% is where fine chocolate is most legible. The number predicts sweetness; it hides the solids-to-butter split that decides texture and intensity.

Frequently the same product under two words. Where a real difference exists, it is processing — and processing is checkable.

Takeaway: The two words are not a nutritional distinction. Buy on the processing claims — natural or alkalised, raw or roasted, and the fat percentage — which are checkable.

In a bake, the difference is mostly sugar and mostly ignorable. In anything that sets, it is neither.

Takeaway: Eating chocolate bakes fine — adjust the sugar for the percentage. It will not coat or temper properly, and 'baking chocolate' means two different things depending on the country.

Cocoa butter is one component of chocolate — the fat, with essentially no flavour. Swapping one for the other changes everything.

Takeaway: Cocoa butter is the flavourless fat component. It thins chocolate correctly and carries colour, and it cannot substitute for chocolate anywhere flavour is needed.

55, 70, 85 and 100 side by side: what changes, what it costs, and where each is actually useful.

Takeaway: 70% is the reference point and where fine character reads best. Higher percentages are drier, less forgiving of poor beans, and need recipe adjustment when baked.

Temper properly and store at a stable temperature. Cycling matters more than warmth.

Takeaway: Temper properly, store sealed at a stable 15–18°C, and keep chocolate out of the fridge. Temperature cycling causes more bloom than steady warmth does.

The gloss is a cast of the mould surface, not a property of the chocolate. Everything else follows from that.

Takeaway: Polish polycarbonate moulds with dry cotton wool and never wash them. Gloss comes from the mould; release comes from temper.

Five techniques, in order of how much they return for the effort. Shards win.

Takeaway: Shards give the best result for the least effort and transfer sheets give the most professional finish. All of it needs temper, and none of it should be set in the fridge.

Room temperature, break it, smell it, let it melt. Four steps, no equipment, and most people have never done all four.

Takeaway: Room temperature, break and smell immediately, melt rather than chew, and wait thirty seconds afterwards. No equipment and it changes the experience completely.

The brew method changes the pairing more than the bean does. Espresso and filter want different chocolate.

Takeaway: Espresso needs 70%+ to stand up to it; filter suits milk and dark milk. Both share roast chemistry with cacao, so the risk is redundancy rather than clash.

Tea's tannin is the constraint. It compounds with chocolate's, which rules out more combinations than people expect.

Takeaway: Tannin compounds, so brew weaker than usual and avoid strong black tea with high-percentage dark. Matcha with white chocolate is the standout.

Alcohol anaesthetises the palate, which is why dilution matters more than the pairing does.

Takeaway: Add a few drops of water to the spirit — it matters more than the match. Bourbon suits milk chocolate, peat needs earthy dark, and aged rum suits almost anything.

Two acidities compound. That one principle explains most fruit pairings that disappoint.

Takeaway: Two acidities compound rather than balance. Use dried fruit or a low-acid chocolate, and remember that citrus oil pairs far better than citrus juice.

A pairing that works on fat and salt rather than on flavour matching, and one that fails badly when the cheese is too mild.

Takeaway: The bridge is fat and salt, not flavour. Aged hard cheeses and blue cheese work; mild fresh cheeses contribute nothing.

Minimally processed cacao paste, sold at a large premium on the strength of claims the evidence does not support — wrapped around a genuine and much older practice.

Takeaway: Ceremonial cacao is cacao paste with no standard behind the word. The compounds cited do not support the effects claimed, the Mesoamerican history is real but not continuous with the modern market, and 100% chocolate prepared the same way is the same product.

There is no single answer, and the reason why is the most interesting fact about chocolate: cocoa butter can solidify six different ways, each with its own melting point.

Takeaway: Tempered chocolate melts between about 30°C and 34°C — a range rather than a point, because cocoa butter can crystallise six ways and each melts differently.

Carob is the standard substitute and it is not a substitute — it is a different food that happens to be brown. Knowing that makes it much easier to use well.

Takeaway: Carob is a different food rather than a substitute, and it fails where texture matters because it has no cocoa butter. For caffeine or theobromine avoidance, white chocolate is a better answer than carob.

Almost always plain sucrose, and the alternatives that appear on craft bars each change the manufacturing as well as the taste.

Takeaway: Refined sucrose dominates because it is dry, grindable and cheap. Unrefined and coconut sugars bring flavour and moisture problems; polyols bring a different tempering curve and a digestive limit.

Four pressures at once — disease, ageing trees, weather and a market that cannot respond quickly — and three visible consequences on the shelf.

Takeaway: Disease, ageing trees and weather have hit a crop that cannot respond quickly. The visible results are smaller bars, more declared vegetable fat, and formulations drifting toward the legal minimum.

Nothing but ground cacao. It is not a stronger version of 85% — it is a different eating experience, and most people should meet it as an ingredient first.

Takeaway: 100% is a different food rather than a stronger bar. It is the most revealing test of bean quality and the least forgiving, and it is easiest to meet as a baking ingredient.

Four inputs and one structural constraint: cacao is a perennial tree crop, so supply cannot answer a price signal for years.

Takeaway: Four inputs move the price, but the constraint that matters is biological: a cacao tree takes three to five years to bear, so supply cannot answer a shortage for years.

Disease, ageing trees and weather, concentrated in two countries that grow more than half the world's crop.

Takeaway: Two countries grow over half the world's cacao, so West African disease and ageing trees are a global supply problem — and replanting takes years to bear.

Bean cost, batch size and yield — and the premium is smaller than the retail gap suggests once you account for the last two.

Takeaway: Bean cost, small-batch overhead and yield loss, in that order. The premium buys specificity rather than consistency — and it buys nothing at all in a brownie.

Bulk sweeteners can replace sugar's volume; intense sweeteners cannot. That distinction explains every formulation you will meet.

Takeaway: Bulk sweeteners replace sugar's volume; intense ones cannot and always arrive with a bulking agent. None of them meaningfully reduces the energy, because cocoa butter supplies most of it.

Percentage tells you directly: the remainder above the cocoa figure is very close to the sugar content.

Takeaway: For dark chocolate, sugar is roughly 100 minus the cocoa percentage. A higher percentage is usually a better route to less sugar than a sugar-free product is.

A compositional question with a clear answer, and a dietary question this site will not answer.

Takeaway: Carbohydrate falls with percentage and is on the wrapper; energy does not fall meaningfully at any percentage. Whether that suits a given diet is a clinical question, not one this site answers.

A numbered procedure rather than an explanation. Three temperatures, one test, and the two mistakes that account for most failures.

Takeaway: Melt fully to 45–50°C, seed down to 31–32°C, test on paper for three minutes. Not melting hot enough and a warm room account for most failures.

Because it contains no cocoa solids. It is still legally chocolate, and both halves of that are worth knowing.

Takeaway: White chocolate contains no cocoa solids, which is the true and useful statement. It is nonetheless legally chocolate in the EU and US, and the bad reputation comes from sub-threshold products built on vegetable fat.

Possible, and the risk is entirely in the thawing. Wrap airtight, thaw sealed, and never rush it.

Takeaway: Freeze wrapped airtight, and thaw in two stages — freezer to fridge to room — leaving the wrapping on until it reaches room temperature. The thawing causes the damage, not the freezing.

The overlooked variable. It does not affect crystallisation directly, and it ruins the results anyway.

Takeaway: Humidity does not affect crystallisation and does ruin the results, by condensing on any surface cooler than the air. Keep the room below 50% and never work with a cold mould.

The same product under two spellings, frequently at two prices.

Takeaway: The two spellings name the same fat and carry no compositional difference. Check deodorised against natural, and food grade against cosmetic — those are the differences that matter.

A format that constrains the product almost completely, and a genuine formulation story rather than a complaint.

Takeaway: The format demands thin, cheap, warehouse-stable pieces, which pushes toward compound coating or minimum-cocoa milk chocolate. Reading the fat on the ingredients list tells you which you have.

A place of manufacture and a texture standard — not a bean, not a variety, and not a guarantee about sourcing.

Takeaway: Swiss means made in Switzerland. It describes processing and a smooth, mild style — not the origin of the beans and not the terms they were bought on.

A filled-chocolate shop tradition, and a word that describes where a chocolate was finished rather than where it was made from beans.

Takeaway: Belgian describes work done in Belgium on chocolate made from beans elsewhere. Ask which couverture the shop uses — it is the single most informative question available to you.

Chocolate treated as pastry — built around ganache, infusion and seasonal collections rather than around a fixed range.

Takeaway: The French tradition treats chocolate as pastry: ganache-centred, flavoured by infusion, and organised into seasonal collections rather than a permanent range.

A hazelnut tradition born of a cocoa shortage, and a drinking chocolate thick enough to stand a spoon in.

Takeaway: Hazelnut chocolate is Italy's contribution and it began as a way of stretching scarce cocoa. Treat gianduja as a filling, not as a coating — it will not set.

Europe's oldest chocolate culture, and the one that kept the drink thick when everybody else thinned it out.

Takeaway: Spanish drinking chocolate is thick by design, thickened with starch, and descends more directly from the Mesoamerican drink than any other European preparation.

Coarse, stone-ground, cinnamon-spiced and made for drinking or cooking — and its grain is a characteristic rather than a fault.

Takeaway: Mexican chocolate is a drinking and cooking product, deliberately coarse, and it is not a substitute for couverture in any technique that requires tempering.

Seasonal, regional, softer and less sweet than European chocolate — and organised around gifting to a degree no other market matches.

Takeaway: Expect softer, less sweet chocolate, delicate flavours carried on white chocolate, and a market built around limited-period and regional releases. Try nama chocolate while you are there — it does not travel.

A lower legal minimum, a distinctive milk tang, and a confectionery tradition built around the filled bar rather than the plain tablet.

Takeaway: A lower cocoa minimum, a characteristic milk tang and a filled-bar tradition. It is a different standard rather than a worse execution of the European one.

Chocolate joined a confectionery tradition that was already highly developed, and the interesting products are the hybrids.

Takeaway: Chocolate here is a component of an older sweets tradition rather than a category of its own — and pistachio, dates and pastry textures are the through-lines.

A filled bar of pistachio cream, tahini and toasted shredded pastry — created by one company in 2021 and made globally famous by a video in 2023.

Takeaway: A moulded shell around pistachio cream, tahini and toasted kadayif. It is a 2021 commercial product, not a traditional confection, and the crunch is the whole thing.

A very old pairing that a very new product made globally visible — and a nut that behaves differently from hazelnut and almond.

Takeaway: Read the ingredient order. Pistachio is expensive, and the difference between a pistachio product and a pistachio-flavoured one is visible on the label and in the colour.

Recent chocolate phenomena travel through video rather than through retail, which changes what gets popular and how fast it is copied.

Takeaway: Modern chocolate trends are demand-first and video-driven, so virality selects for visual distinctiveness and scarcity — neither of which tells you anything about how it tastes.

Two different products are sold as premium — expensive presentation and expensive sourcing — and the packaging is designed to blur which is which.

Takeaway: Decide whether you are buying presentation or sourcing. Only the second is visible in the chocolate, and only specific checkable claims — named farm, harvest year, price paid — evidence it.

Factory visitor centres, museums, shops and tastings are four different experiences with four different reliabilities.

Takeaway: A visitor centre is marketing with good machinery in it. A small maker's workshop teaches more, and the two questions that cut through any shop are which couverture and how fresh.

The occasion decides the format before it decides the chocolate — and getting the format wrong is what makes a good gift land badly.

Takeaway: Match the format to the occasion first. The best chocolate in the wrong format is a worse gift than good chocolate in the right one.

A Victorian commercial invention with sharply different rules in different countries — and a season where the format matters more than the contents.

Takeaway: The custom differs sharply by country, and the season's products are made ahead for moulding rather than for eating. Check the fat in the ingredient list and buy perishable centres early.

Hollow moulding is the hardest routine job in chocolate manufacture, and every egg on a shelf is evidence of a tempering process that worked.

Takeaway: A hollow egg is tempering made visible. Expect a different formulation from the brand's bars, expect occasional bloom from long storage, and check the fat in the ingredient list.

The biggest chocolate season and the one with the widest spread of formats — each of which is a different product doing a different job.

Takeaway: Decide whether you are buying a ritual, a shared household item or a personal gift. They are three different products and the packaging deliberately makes them look like one.

A season defined by a packaging requirement rather than by a product — and miniatures are a reformulation rather than a smaller bar.

Takeaway: The miniature exists because the custom requires it, and it is a reformulation rather than a smaller bar. Keep the outer packaging — the individual wrappers often carry no allergen information.

Volume, heat and time are the three constraints, and they rule out most of what people first think of.

Takeaway: Heat and time decide it. Anything with fresh cream in it and anything relying on a temper holding through a warm afternoon should be planned around, not hoped over.

Three things on the outside of a box tell you most of what is worth knowing, and none of them is the brand.

Takeaway: Date, fat, and whether the centres are genuinely different. Those three are on the outside of the box and tell you more than the brand does.

The usual gift instincts invert. Packaging matters less, unusual sourcing matters more, and the risk is buying something they already know.

Takeaway: Buy unfamiliar rather than expensive, look for specific checkable sourcing claims, and prefer a set built around one variable over a broad selection.

Most plain dark chocolate already qualifies. The vegan claim is doing work about the factory, not about the recipe.

Takeaway: Check the ingredient list for milk, honey and glazes — that settles the recipe. A vegan mark additionally speaks to the production line, which is a separate question.

A narrower claim than vegan and a different one from allergen-safe — and white chocolate can never qualify.

Takeaway: Scan for milk fat as well as milk solids, remember white chocolate is never dairy-free, and treat cross-contact as a separate question the ingredient list does not answer.

Chocolate contains no gluten by nature. Everything that puts gluten into a chocolate product is something added to it or something on the line.

Takeaway: Plain chocolate has no gluten in it. Malt, biscuit, wafer and cereal inclusions do — and for coeliac disease the certification mark, not the ingredient list, is the meaningful signal.

The hardest dietary requirement in chocolate, because nuts are everywhere in the category and the risk is overwhelmingly on the equipment rather than in the recipe.

Takeaway: The recipe is easy to check and is not the problem. For a serious nut allergy the relevant assurance is a dedicated facility, not an ingredient list.

Two different statements appear on every chocolate package, they answer different questions, and only one of them is legally required.

Takeaway: The ingredient list is a legal declaration about the recipe. 'May contain' is a voluntary, unstandardised statement about the equipment — and its absence proves nothing.

Three different claims — sugar-free, no added sugar and reduced sugar — mean three different things, and none of them means calorie-free.

Takeaway: Sugar-free, no added sugar and reduced sugar are three different claims. None reduces the fat, and high-percentage dark chocolate gets to low sugar without any substitution.

A marketing category rather than a regulated one, built on a carbohydrate accounting convention that is not universally accepted.

Takeaway: Keto is an unregulated marketing term and net carbs is a calculation convention, not a measurement. Read the declared carbohydrate figure, and note that high-percentage dark chocolate reaches the same place without substitution.

Milk chocolate carries lactose because it carries milk powder — and the amount varies more than the category suggests.

Takeaway: Milk and white chocolate contain lactose; plain dark generally does not unless milk fat has been added. Ingredient order is the best available indicator of how much.

Sugar in chocolate does structural work as well as sweetening, and the particle size matters more than the source does.

Takeaway: Refined cane and beet sugar are the same molecule. Where sugar genuinely changes a chocolate is through unrefined varieties carrying molasses and moisture, and through how finely it was refined.

Four different dairy ingredients appear on chocolate labels and they are not interchangeable — the choice changes flavour, texture and how the bar tempers.

Takeaway: Dairy goes in dry because water seizes chocolate. Milk fat softens the set and lowers the tempering window; milk crumb produces caramelised flavours that mixing the same components separately cannot.

Vanilla in chocolate is usually there to round off roughness rather than to taste of vanilla — which is why removing it is a stylistic choice and not automatically an upgrade.

Takeaway: Vanilla in chocolate is a rounding agent, not a flavour. Leaving it out exposes origin character and also exposes roughness — which is why it is a style rather than an upgrade.

Cocoa butter's melting behaviour is why chocolate works at all — and it is also why every substitute involves a genuine trade-off rather than a simple downgrade.

Takeaway: Cocoa butter's narrow melt just below body temperature is what makes chocolate snap and melt cleanly. Substitutes trade that curve for cost or for not needing tempering, and the mouth notices.

Anything added to chocolate has to be dry, and that single constraint explains why fruit chocolate uses powder and why so many flavoured bars are white.

Takeaway: Everything added to chocolate must be dry. That is why fruit is freeze-dried and citrus goes in as oil — and why delicate flavours are carried on white chocolate, which has nothing competing with them.

A perennial understorey tree, harvested by hand twice a year, on farms averaging a few hectares — and every structural problem in the sector follows from those facts.

Takeaway: Cocoa is a perennial tree crop harvested by hand on smallholdings. Supply cannot answer a price rise for three to five years, and the flavour-determining steps happen at the farm.

Several million smallholder households, concentrated in West Africa, most of whom have never tasted chocolate made from their own beans.

Takeaway: Millions of smallholder households, concentrated in two West African countries, receiving a farmgate price that does not track the exchange price upward. That gap is the sector's central problem.

Seven or eight parties stand between a cocoa tree and a chocolate bar, and the two that set the price are neither the farmer nor the brand.

Takeaway: Aggregation at three separate points is why cocoa is hard to trace, and grinding is where bean identity is lost. Keeping a lot separate costs money at every step, which is what a traceability claim is paying for.

Four connected problems — income, deforestation, child labour and climate — and income is upstream of the other three.

Takeaway: Income is upstream of deforestation, child labour and climate vulnerability. An intervention that addresses one of the three without addressing income tends to push the pressure sideways.

A large, measured and persistent problem that is mostly household labour on smallholdings — which is why two decades of voluntary commitments have not resolved it.

Takeaway: Most measured child labour in cocoa is household labour on farms that cannot afford to hire. That is why monitoring alone has not resolved it and why income is the mechanism that matters.

Clearing forest is cheaper than replanting a tired farm, which is the economic logic every forest-protection scheme has to overcome.

Takeaway: Clearing forest is the cheaper alternative to replanting an ageing farm. Any scheme that does not make replanting affordable is arguing against an economic logic rather than changing it.

The newest mechanism is not a consumer choice at all — it places legal obligations on importers, which is a different kind of lever from a certification mark.

Takeaway: Regulation moves the obligation from the consumer to the importer, and the geolocation requirement runs directly against a supply chain built on aggregation. It is a market-access condition, not an income mechanism.

Very little of what a package asserts is checkable. Three specific claims are, and they are the ones worth looking for.

Takeaway: Look for a named producer, a harvest year and a published price paid. Those three are checkable; everything else on a sustainability-forward package is either an audit against a standard or an unverifiable adjective.

Frosting has to stay soft and spreadable at room temperature, which rules out the chocolate most people reach for.

Takeaway: Frosting needs to stay soft, so cocoa powder often beats melted chocolate — and chocolate chips are formulated specifically not to do what a frosting needs.

A sauce must pour hot and not set solid cold, which makes it the one chocolate preparation where less cocoa butter is an advantage.

Takeaway: Cocoa powder makes a sauce that stays fluid; melted chocolate makes one that sets. Choose by what the sauce has to do cold, not by which chocolate is better.

Three words for three different products, used interchangeably on packaging — and the difference is whether there is cocoa butter in it.

Takeaway: Cocoa powder is defatted; drinking chocolate keeps the cocoa butter; instant mix is mostly sugar. Read the first ingredient — the words on the front are not used consistently.

Quantity planning for truffles, coating and moulding — including the coating losses nobody accounts for and everybody discovers.

Takeaway: Weigh one finished piece and multiply — then add a working volume for dipping, which is several times what ends up on the product. That last part is what catches people out.

The next question after deciding to make chocolate at home — and the one where a bad purchase makes everything downstream impossible.

Takeaway: Fermentation quality is the thing to establish, because nothing downstream fixes it. Ask for a cut test and a harvest year, and buy small lots of several origins before committing to a sack.

Five stages need equipment and only two of them need equipment you cannot improvise — which makes the melanger the real decision.

Takeaway: Roasting, cracking and winnowing can all be improvised. Refining and conching cannot — a stone melanger is the one purchase that decides whether the project is possible.

An award sticker means a panel liked a sample on a day, in a category, against the other entries. That is worth something and it is not what most people read it as.

Takeaway: An award says a submitted sample was judged well in a category in a year. Use it to find makers worth trying, not to rank the bar in your hand.

A discovery mechanism rather than a value one — and the useful question is who curates it and whether you can steer it.

Takeaway: It is a discovery product, not a value one. Judge it on whether the curation is explained and whether you can steer it away from what you will not eat.

The same fat appears in cosmetics and pharmaceuticals, and the deodorised grade sold for those uses is a genuinely different product from the one a chocolate maker buys.

Takeaway: Same fat, two grades. Deodorised cocoa butter is the cosmetic and pharmaceutical one and is also used in chocolate where added butter must bring no flavour of its own.

There is no standard bar size, and the material of the mould matters more than its dimensions.

Takeaway: There is no standard bar size — work from cavity capacity. Polycarbonate gives gloss and needs correct tempering; silicone tolerates poor temper and never shines.

Plain chocolate involves few questionable ingredients. Fillings, glazes and flavourings involve several — and the answer is a certification question rather than an ingredient-list one.

Takeaway: Plain chocolate raises few questions; fillings, glazes and flavourings raise most of them, and the label cannot resolve the source of an emulsifier or a gelatine. That is what certification is for — and which certifier is the question, not whether a mark exists.

The dairy question turns on the same milk fat that the dairy-free guide already covers — and the equipment question matters here in a way it does not for most dietary claims.

Takeaway: The dairy-or-parve classification is the chocolate-specific part, and added milk fat in dark chocolate is the thing most often missed. Equipment status is part of the determination here, not a footnote to it.

Eight different words on an ingredient list all mean milk, and the one that catches people out is on dark chocolate.

Takeaway: Learn the eight words that mean milk, check dark chocolate for milk fat specifically, and treat the recipe and the equipment as two separate questions — only the first is on the label.