When we talk about board game components, it is easy to treat them as the physical layer that comes after the design. First we decide how the game works, then we decide whether a particular piece of information should be represented by a card, a token, a die, a miniature or a track. From that perspective, components are containers: the rules create meaning, while the physical objects simply carry it.
But components are not neutral containers. The choice of component determines what information can be stored, how easily it can change, what players can see at a glance, what they can manipulate, and even what kinds of decisions feel natural. A token is excellent at representing a quantity or a state. A card can combine text, images, categories, costs, actions and ownership in a single object. A die does not merely display a result; its faces define a probability distribution. Once we look at components this way, choosing between a card, a token and a die is no longer a production decision. It is a design decision.
This does not mean that one kind of component is inherently more sophisticated than another. A simple component can be exactly what a system needs, while a component carrying too much information can make a game difficult to read. The useful question is not which component can do the most, but how much meaning the game needs that component to carry, and whether players can understand that meaning while they are actually playing.
A component can represent state, information or possibility
Consider a coin token. In many games, its job is intentionally narrow: one token represents one unit of money. If I have five coins, I can count them and immediately understand part of the game state. This is a strength rather than a limitation. Because the component carries very little additional information, players do not need to interpret it every time they look at the table. The token becomes almost transparent; we stop seeing the piece itself and simply see “five money.”
Cards behave differently because their physical structure allows them to carry several kinds of information simultaneously. A card can have a title, an illustration, a color, a cost, an effect, a category, a value and an orientation, and the game can make any combination of those properties relevant. Even its location can change its meaning: the same card may behave differently in a player’s hand, on the table, in a discard pile or tucked underneath another card. The component has not changed, but the system can read it differently depending on context.
This is why a card often represents more than a current state. It can represent a possibility. A card in my hand tells me something I may be able to do, something I may want to preserve, something another player may need, or something that could become valuable under different circumstances. The physical object gives the game a place to store several potential futures until the player chooses between them.
The same piece of cardboard can perform several jobs
7 Wonders is a particularly clear example of how much meaning can be compressed into a card. At first glance, each card is simply one option in a draft. Once I take it, however, I may build it as a structure, discard it for money, or use it to construct a stage of my wonder. If I build it, its color and symbols can connect it to entirely different systems: resource production, military strength, science, commerce, civic scoring or future discounts. At the same time, the card I choose is also a card I deny to the player sitting next to me.
The important point is not merely that the card has several uses. It is that the component allows several systems to meet in one physical object. Drafting, economy, engine building, scoring and player interaction can all refer to the same card without requiring a separate component for each function. The result is a game in which a relatively compact set of objects can generate a much larger decision space.
That compression can create elegance because the player gradually learns to read more into the same object. A beginner may look at a card and focus on its immediate effect. An experienced player may simultaneously notice its cost, its future production, the set it contributes to, the card it prevents an opponent from receiving, and the opportunities it opens several turns later. Nothing new has been added to the table. What has changed is the player’s ability to interpret the information already there.
This is one reason component design and resource design overlap without being the same subject. A card may function as a resource, but it can also be an action, a rule, a piece of hidden information or a scoring condition. The component is the physical language through which those functions are expressed.
Dice are not only random number generators
Dice provide another useful example because their familiar form can hide how flexible they actually are. In the simplest case, a six-sided die produces a number from one to six with equal probability. If that is all the game needs, the conventional die is an extremely efficient component: players already understand how to roll it and how to read the result.
But a die does not have to represent numbers at all. Its faces can contain colors, hits, shields, resources, movement, special effects or combinations of symbols, and the distribution of those faces becomes part of the design. Descent: Journeys in the Dark uses different dice to encode attack power, range, defense and special results. Choosing which die belongs in a roll therefore changes more than the possible maximum result; it changes the shape of the probability space.
In Sagrada, by contrast, the familiar numerical values and colors are valuable precisely because the game needs the dice to remain readable as objects that will be placed into a pattern. Their meaning comes not from a dense collection of custom symbols but from the interaction between color, value and position. The component is doing less on each individual face, yet that restraint serves the system.
These examples illustrate an important principle: component density is not measured by how much ink appears on an object. It is measured by how many meaningful relationships the object can participate in. A visually simple die may matter because of where it can be placed, what other dice may sit beside it, how difficult a particular face is to obtain, and what constraints it creates later. A component can be mechanically dense while remaining visually simple.
Sometimes the players supply the missing meaning
Not all meaning has to be printed on the component. Codenames uses cards that are, in physical terms, remarkably simple. A word card contains a word. Yet that word can become connected to dozens of possible concepts depending on the clue, the other words on the table and the associations shared by the players. Much of the component’s meaning is therefore generated socially rather than encoded by the designer.
This is a different kind of efficiency. Instead of placing more rules or information onto the card, the game uses a minimal object as a trigger for interpretation. The card is deliberately incomplete, because the interesting part of the experience comes from what players project onto it and from the uncertainty over whether other people will make the same connection.
This matters because there is a temptation to equate a multifunctional component with a component that contains many printed functions. In practice, components can gain meaning from rules, position, ownership, timing, neighboring pieces and player interpretation. A wooden cube may represent a worker when it is in one area, a vote when it moves to another, and a score when it reaches a track. The physical simplicity of the object tells us very little about the complexity of its role.
More meaning is not automatically better
If one component can carry several systems at once, it is tempting to conclude that designers should always try to make components more multifunctional. That would be a mistake. Every additional meaning placed onto an object creates another thing the player may need to remember, distinguish or check. Compression can reduce the number of physical pieces while increasing the amount of interpretation required from the player.
A token that always means one coin is cognitively cheap. A card whose value depends on its color, orientation, location, neighboring cards and the phase of the round may be extremely efficient from a systems perspective while being exhausting to use. The question is therefore not how many functions can be fitted into a component, but how many can be fitted into it before the component stops being readable.
Good component design often relies on hierarchy. Some information should be visible immediately, while other information only matters when the player examines the component more closely. Color can communicate category before text communicates a specific effect. Shape can distinguish one family of pieces from another. Position can communicate ownership without requiring an additional marker. Familiar conventions can reduce the amount of explanation needed. When these layers support each other, a component can carry considerable mechanical meaning without forcing the player to consciously decode all of it every time.
The reverse is also true. If two different states look almost identical, or if one symbol means several unrelated things without a clear contextual cue, the component may technically be efficient but function poorly at the table. Physical economy is not the same as cognitive economy.
The table is an interface
Board games do not have a screen that can hide information until it becomes relevant, automatically recalculate values or change the appearance of an object when its function changes. Almost everything the system wants players to know has to exist somewhere on the table, in their hands or in their memory. Components are therefore part of the game’s interface.
That makes physical design inseparable from information design. If a player’s current money is represented by a pile of identical tokens, the quantity can be estimated from across the table. If it is recorded on a small personal track, the same information becomes harder for opponents to monitor. If a card is turned sideways after use, orientation becomes a state marker. If a piece is physically moved from an available area to an exhausted area, location replaces the need for another token or reminder.
Each of these decisions changes not only how the game looks but how players process it. The component system determines which information is public, which is private, which is easy to compare and which requires effort to reconstruct. It can reduce bookkeeping, create uncertainty, expose intentions or conceal them. The physical object is participating in the rules even when no rule explicitly says that it is.
Design components by asking what they need to communicate
When choosing components, I find it useful to begin with the function rather than the object. Instead of deciding that a game needs cards because cards are common in board games, ask what information or behavior the system needs to represent. Does the object need to hold several attributes at once? Does it need to be shuffled? Hidden? Traded? Stacked? Rotated? Counted quickly? Placed spatially? Does its probability distribution matter? Will players need to recognize it from across the table, or inspect it closely in their own hand?
Those questions often point naturally toward a component. Cards are excellent when individual objects need identity and information. Tokens are excellent when quantity and state need to be read quickly. Dice are powerful when probability itself needs a physical representation. Tracks can communicate progression without filling the table with repeated pieces. Boards and tiles can make spatial relationships part of the system. None of these categories is absolute, and some of the most interesting designs deliberately use a familiar component in an unfamiliar way, but the choice should still follow the experience the system needs to create.
A component earns its place not merely by representing something in the game, but by making that thing easier, richer or more interesting to interact with. Sometimes that means doing exactly one job with absolute clarity. Sometimes it means allowing several systems to converge on the same object. The difficult part is knowing which of those the game actually needs.
The most elegant component is therefore not necessarily the one with the most uses, just as the most elegant game is not necessarily the one with the fewest pieces. Elegance appears when the physical object, the information it carries and the decisions it creates reinforce one another. At that point, the component is no longer packaging for the mechanics. It has become part of the mechanics themselves.