Strategy guides

Housing and Amenities: Keeping Cities Growing

How housing throttles growth as a city nears its cap, where every point of housing and every amenity actually comes from, when an Entertainment Complex beats a Water Park, and the settling and expansion habits that quietly starve an empire.

What this guide assumes

Gathering Storm rules. The housing and amenity thresholds themselves are old, but loyalty interactions with unhappiness only exist from Rise and Fall onward, and the Preserve and several wonders mentioned here come from later content.
Standard game speed on Prince through Emperor. Higher difficulties do not change how a city grows, but the AI's luxury monopolies make trading for amenities harder.
A mid-sized empire of roughly six to twelve cities in which several are expected to grow large. A four-city empire rarely runs short of amenities, and a thirty-city empire has problems this guide only touches on.
Any victory condition. Growth is the prerequisite for district slots, which every plan on this site eventually depends on.

How Housing Throttles Growth

Every city grows by piling up surplus food, and housing is the lid on that jar. The number beside the house icon in the city panel is not a hard wall; it is the point where growth starts being taxed. A city sitting two or more housing above its population grows at full speed. Once it closes to within one point of the cap, growth is cut roughly in half, and when population reaches or passes the cap the remaining surplus is trimmed so hard that a city with a fat food income can still need twenty turns for its next citizen. Nothing about the food itself changes; the penalty applies to growth progress, which is why the city screen can show a healthy surplus and a miserable turns-to-growth figure at the same time.

The practical reading is that housing only matters in cities that have food to spare. A hill city with a handful of mines and no surplus does not care about spare housing, and queuing a Granary there changes nothing. Conversely, a river city with three farms and a fishing boat will bump into its cap by the early Classical era and then idle. Since specialty districts unlock with population, that idle city is also locked out of its third and fourth district, and that is the real cost. Housing is, in effect, the ceiling on how many districts a city will eventually hold.

Check the number at two moments in particular: the turn a city is founded, and the turn it grows to within one of its cap. In the first case you are deciding what its ceiling will be for the next fifty turns. In the second you are deciding whether the next production item is a growth building or something else, and a Granary queued the turn before the penalty bites is worth far more than one built after the city has stalled for ten turns.

Where Housing Comes From, Roughly in Game Order

The single largest housing decision is the settle location, because water at the city center sets the base figure that everything else stacks on. A center adjacent to a river, lake, or oasis gets the highest tier; a center on the coast but away from fresh water gets the middle tier; a dry inland center gets the lowest and starts life already cramped. The gap between top and bottom tiers is several points, more than a Granary and Sewer combined. Mountains do not count as fresh water for the center, even though they will satisfy an Aqueduct later. Mohenjo-Daro's suzerain bonus is the one way to erase this decision entirely, since it treats every city as riverside for housing purposes.

The next source arrives with builders. Farms, pastures, plantations, camps, fishing boats, and most other tile improvements each add half a point of housing, so a city working four such tiles has gained two housing without building anything. This is why a mine-heavy hill city stalls earlier than a farm-heavy river city on the same food: the mines add nothing. The Granary is the first building that adds housing, and it is cheap enough to belong in every city expected to grow, usually right after the first district.

The Aqueduct is the midgame fix for a bad settle. It must touch both the city center and a river, lake, oasis, or mountain, and it grants a large amount of housing to a city that lacked fresh water but only a little to one that already had it, so its value runs inversely to how well you settled. The Sewer adds a flat amount to any city later and is rarely urgent. Harbor buildings add housing in coastal cities, the Lighthouse first and the Seaport more, which is one reason coastal cities tolerate mediocre water. From the Urbanization civic on, the Neighborhood becomes the scalable answer: a city can hold more than one, and each one's housing depends on the appeal of its tile, which gets its own section below.

Settle: fresh water at the center is the largest single housing gain in the game, and it is fixed forever.
Improve: pairs of farms, pastures, plantations, camps, or fishing boats each add a point; mines and quarries add none.
Build: Granary early, Aqueduct in dry cities, Sewer late; Lighthouse and Seaport in coastal cities.
Place: Neighborhoods on high-appeal tiles, several per city if the city is meant to be large.
Extras: Hanging Gardens and Temple of Artemis, the Audience Chamber for governed cities, Insulae and Medina for district-heavy cities, the Religious Community belief, and Mohenjo-Daro's suzerain bonus.

Amenities Start with Luxuries

Amenities are the parallel constraint. A city needs none for its first couple of citizens, then wants roughly one more for every two citizens beyond that, so a size-ten city is asking for about four and a size-twenty city for about nine. The requirement is not empire-wide; each city runs its own balance, and the same luxury can be doing nothing in one city while preventing a revolt in another.

Luxury resources are the main supply. Each distinct improved luxury supplies one amenity to up to four of your cities, assigned automatically to whichever need it most. A second copy of the same luxury supplies nothing; it exists to be traded. Two consequences follow. Variety beats quantity, and the luxury supply scales with kinds owned, not tiles. And the arithmetic turns against wide empires: eight luxuries across eight cities averages four amenities per city, but the same eight luxuries across sixteen cities averages two. Every city past the fourth that a luxury cannot reach is a city that has to find its amenities elsewhere.

Trading for luxuries is the cheapest fix, and the AI is usually willing to swap one duplicate for another. Suzerainty over a city-state grants access to its luxuries as well. A few leaders and city-states bend the rule outright: Montezuma's Aztecs push each luxury to more cities than the usual four, the suzerain of Zanzibar receives two exclusive luxuries with unusually wide reach, and Buenos Aires lets bonus resources count as luxuries, which can double the supply of a wide empire sitting on wheat and cattle.

Entertainment Complex or Water Park

Once luxuries run out, districts take over. A city may hold an Entertainment Complex or a Water Park, never both, and the two share a shape: the district itself provides a point of amenity, the first building helps only its own city, and the second and third buildings reach every city center within six tiles. That regional radius is the entire design problem. A Zoo or Aquarium in the geographic middle of a five-city cluster is worth five buildings scattered one per city.

The choice between them is mostly terrain and timing. The Water Park must sit on coast or a lake beside land and unlocks at Natural History, while the Entertainment Complex is available from Games and Recreation and goes anywhere. In a landlocked cluster there is no choice to make. On a coast, the Water Park's buildings are somewhat stronger in the regional stage but arrive a full era later, so a coastal empire that is already displeased in the Medieval era should not wait for it. The Entertainment Complex also unlocks the Colosseum, which broadcasts amenities across the same six-tile radius and is the best early answer to a tight cluster, and much later Estádio do Maracanã, which also demands an Entertainment Complex beside it; no wonder attaches to the Water Park.

Placement follows the radius, not the city. Pick the city whose center sits within six tiles of the most other centers, even if that city itself is small, and put the district there; treat it as infrastructure like a Dam rather than as a personal upgrade. In a spread-out empire where no such center exists, luxuries and policies have to carry more weight, and these districts revert to being one-city fixes for your largest cities only. Huey Teocalli is a different tool: built on a lake, it feeds its own city amenities from adjacent lake tiles, which makes a lake-heavy capital unusually easy to keep happy.

Amenities from Religion, Policy, and Governance

Religion contributes through beliefs rather than through the religion itself. Zen Meditation adds an amenity in cities with enough specialty districts, Religious Community adds housing from Shrines and Temples, and Feed the World turns those same buildings into food and housing. A civ that is building Holy Sites everywhere anyway can therefore lean on them where a secular empire needs Entertainment Complexes. The catch is the district-count condition on these beliefs: a fresh two-district city gets nothing, which is fine, since two-district cities are rarely the unhappy ones.

Policy cards are the flexible source. Retainers grants an amenity for a garrisoned unit, which is free in cities that keep a defender anyway; Liberalism rewards district-heavy cities; New Deal is the late-game hammer, buying housing and amenities in developed cities for a gold cost per city. Among Government Plaza buildings, the Audience Chamber hands housing and an amenity to every city with an established governor and bleeds loyalty from cities without one, so it suits a compact empire with a full governor roster and punishes a sprawling one. Wonders sit on top: Temple of Artemis converts nearby camps, pastures, and plantations into amenities and adds housing, the Great Bath and Alhambra add a little each, and Estádio do Maracanã reaches the whole empire.

War weariness is the drain hiding in this ledger. Prolonged wars, especially ones fought inside your own borders, subtract amenities from every city, and the effect fades slowly after peace. A player who counted on a content empire and then fights a thirty-turn war will find several cities displeased with no change to their luxuries. Before declaring, keep a policy slot in mind for Retainers, or a Zoo half built somewhere central, so the fix is one turn away instead of fifteen.

What Unhappiness Costs, and the Ecstatic Upside

The city's amenity balance maps to a mood, and the mood applies percentages. Content is neutral. One or two short and the city is displeased: growth and non-food yields both take a modest cut. Three or four short is unhappy, and the cuts deepen sharply, with growth suffering more than yields. Five or more short brings unrest, where growth effectively stops and rebel units spawn beside the city and begin pillaging its tiles. Since Rise and Fall the mood also feeds loyalty, so a border city that is unhappy for long is a border city you may lose without a single enemy soldier arriving.

The upside is real but smaller than the downside. A surplus of one or two makes the city happy and adds a small bonus to growth and yields; a surplus of three or more makes it ecstatic and the bonus grows. Because it is a percentage, it is worth chasing only where there are large yields to multiply, which in practice means the capital and one or two core cities with big Campuses or Industrial Zones. Spreading luxuries thin so every city stays content is almost always better than making three cities ecstatic while five sit displeased, and the game's automatic luxury assignment already leans that way.

Displeased: small penalties, tolerable for a few turns while a Granary or Arena finishes.
Unhappy: growth roughly stalls in practice; fix it before the next war, not after.
Unrest and revolt: rebels and loyalty bleed; the city has stopped being an asset.
Happy and ecstatic: worth engineering in the capital and top yield cities, not everywhere.

Plan Appeal Early So Neighborhoods Land Well

Appeal is a per-tile score most players ignore until Urbanization, at which point they discover their best Neighborhood sites are wedged between an Industrial Zone and a strip of mines. It is computed from neighbors: mountains, coast, woods, natural wonders, wonders, Holy Sites, Theater Squares, and entertainment districts push it up; rainforest, marsh, mines, quarries, Industrial Zones, Encampments, Aerodromes, and pillaged tiles push it down. A Neighborhood on a breathtaking tile gives several more housing than one on an uninviting tile, and because a large city can hold several, the difference multiplies.

The planning move is to split each big city into a dirty side and a clean side while you pin districts in the early game. Put the Industrial Zone, Encampment, and mines on one flank, ideally the hilly one, and leave a woods-heavy flank facing coast or mountains untouched. Holy Sites and Theater Squares belong on the clean side since they raise appeal. Do not chop the woods there: grown woods on a charming tile are worth more as future housing than as a one-time production burst, and the Conservation civic makes replanting possible but slow. The same clean flank is where a Seaside Resort or National Park will later want to go, so the reservation pays twice.

Two late helpers change the math. The Eiffel Tower raises appeal across the entire empire, which can lift borderline Neighborhood tiles a full tier, and the Preserve district raises appeal on the tiles around it while its own buildings reward high appeal. Neither excuses early carelessness, because a Neighborhood built on a bad tile stays bad until the tile itself improves, but both mean a city that kept one clean flank has room left to grow in the last third of the game.

Mistakes That Stall an Empire

Growth failures rarely trace back to a single wrong building; they come from habits set in the first fifty turns. The commonest is settling one tile off the river for a prettier first ring, which trades a permanent housing tier for a few early yields and usually costs an Aqueduct later. The second is expanding to twelve cities on the strength of three luxuries, which produces an empire where every city is displeased and no single fix helps. The third is treating the Entertainment Complex as a default district and building one everywhere, paying maintenance on six Arenas when one central Zoo would have covered the cluster.

Settling a dry site when a river or lake tile is one move away.
Founding cities faster than luxuries and trade deals can support them.
Building an Entertainment Complex in every city instead of one per cluster.
Clearing every woods for production, then wondering why Neighborhoods give so little housing.
Waiting until the city has visibly stalled before starting the Granary or Aqueduct.
Chasing ecstatic in small cities whose yields are too small for the percentage to matter.

Last reviewed: 2026-09-03 · Based on game-data snapshot gs-1.0.12.68-build15296837