The homebuilding industry has a favorite word: Scale.
Executives use it. Analysts use it. Investors use it. Consultants use it. Whenever a builder grows, someone inevitably explains why the company has become more “scaled.” I have become skeptical of the word. Not because scale does not matter. It matters enormously. The problem is that scale, too often, is treated as an answer, a solution.
Rather, it should be a question. Scale of what? Purchasing? Capital? Corporate overhead? Land? Communities? Labor? Market density? Production?
Recent results from K. Hovnanian Enterprises put the issue in sharp relief. Revenue fell to $705.7 million, down from $800.6 million a year earlier, while adjusted EBITDA declined from $77.1 million to $31.9 million. SG&A as a percentage of revenue increased from 11.3% to 12.3%.
That illustrates the conventional case for scale. Homebuilders carry substantial fixed and semi-fixed costs. Corporate personnel, technology, finance, land operations, sales infrastructure, and divisional management do not disappear when a house does not close. When volume falls, those costs weigh more heavily against revenue. But a builder cannot simply buy more land to manufacture scale. Bad land is not scale. It is a future write-down. A weak community does not solve an overhead problem; it can make it worse.
The aim, therefore, is not simply more communities or more closings. It is: enough profitable throughput to make the organization more productive. That shifts the question from: How big can we become? to How much productive density do we need?
Most importantly: What actually becomes cheaper, faster, or more profitable because the company got bigger?
That is the question that matters. A homebuilder does not necessarily become more efficient every time it adds another dot to the map. Sometimes it just has another dot to manage.
Century Communities and the 45-market question
Century Communities’ scale is no illusion. Century delivered 10,387 new homes in fiscal 2025 and generated approximately $4.1 billion in total revenue. It ended the year with 305 selling communities and more than 60,000 lots owned or controlled. Century operates in more than 45 markets across 16 states.
Those are substantial numbers and substantial competitive advantages. A company that size can spread corporate infrastructure across thousands of closings, negotiate nationally, deploy technology across a large organization and access capital and land opportunities unavailable to smaller competitors.
That is real scale. But another number is worth considering.
Divide Century’s home deliveries across more than 45 markets, and you get roughly 230 homes per market. That is only an average. Century obviously doesn’t build exactly 230 homes in every market, and “market” itself isn’t a standardized unit across the industry.
But the calculation raises a question we seldom ask: Is being present in 45 markets the same as achieving scale in those markets?
I don’t think we should assume that it is.
Drive south
Consider Highland Homes. Highland is not a national builder with dozens of dots scattered across the country. It primarily builds in four Texas markets: Dallas-Fort Worth, Houston, Austin, and San Antonio. Highland produced approximately 3,482 homes in 2025, or 871 homes per market.
Again, this is simple arithmetic, not an efficiency study. Production does not divide evenly across those four metros. But 871 versus 230-or-so is a large enough difference to merit attention. Highland proves that a homebuilder can achieve substantial organizational scale without becoming geographically enormous.
And then there is Bloomfield Homes. Bloomfield takes the idea a step further. It is essentially a Dallas-Fort Worth homebuilder, producing roughly 2,000 homes within a single metropolitan ecosystem.
Think about what that concentration could mean. The same municipalities. The same entitlement environment. The same engineering community. The same trade pools. The same inspectors. The same transportation network. The same consultants. The same warranty territory. The same land market. And over time, an enormous amount of institutional knowledge about what works and what doesn’t. That is a different kind of scale.
Operating density
A builder operating at sufficient density should be able to learn which trades perform, which localities move efficiently, where infrastructure causes problems, which plans sell, how long construction takes, and where the organization is losing time and money. That does not prove Bloomfield has lower costs than Century, Highland, or anyone else.
It proves something simpler: Local concentration can be a form of scale.
Now try a thought experiment
Take Bloomfield’s 2,000-plus-home DFW operation and imagine that level of density replicated across five more metropolitan markets. Six markets at approximately 2,000 homes each would get you to roughly 12,000 homes annually. Nobody who has ever opened a homebuilding operation in another city believes you can simply copy and paste Dallas-Fort Worth into Phoenix, Tampa, Atlanta, Nashville, or Charlotte.
Land is different. Labor is different. Municipalities are different. Buyers are different. Competition is different. That’s not the point. The arithmetic raises a more interesting question: Should 45 markets automatically be considered more “scaled” than six exceptionally productive ones? Maybe. But we ought to be able to explain why.
The national numbers make the question more interesting
Century is not alone.
Looking across the large public homebuilders, the differences in closings per market are striking. Some builders appear to push far more production through each geographic market than others. That is not an efficiency ranking. The companies serve different buyers, define markets differently, carry different product mixes and employ different land strategies. Some markets are substantially larger than others. An average can also conceal enormous variation within a single company.
But the spread is too large to ignore. It tells us something important: national size and local operating density are not the same thing.
That means the industry’s favorite word needs a clearer definition.
Bigger can help. Bigger can also hurt
Builders pursue scale for a perfectly legitimate reason. Homebuilding carries high fixed and semi-fixed costs. Corporate management, accounting, technology, finance, land departments, purchasing organizations, sales infrastructure, and divisional leadership do not disappear when closings decline. When volume falls, those costs weigh heavier. The traditional answer is more volume, and sometimes that’s exactly right. But a builder cannot simply buy more land to manufacture scale. Bad land isn’t scale. It’s a future write-down.
A weak community does not solve an overhead problem. It can make it worse. So the goal cannot simply be more communities, more markets or more closings. The strategic business imperative isprofitable throughput to make the organization more productive.
That changes the question from: How big can we become?
To: How much productive density do we need?
The house is still the problem
This is where the scale discussion gets compelling. The homebuilding industry has become remarkably good at scaling the corporation. But have we scaled the house? A house is still assembled on a single, specific lot. One lumber package has to arrive. A plumber has to show up. An electrician has to show up. Drywall has to be installed. An inspector has to inspect it. A superintendent has to coordinate it.
And eventually, a Texas rainstorm screws up someone’s schedule.
A corporation can close 80,000 homes even as its physical production system still, in many respects, operates one house at a time. That is certainly repetition at scale.
But is it manufacturing scale? A manufacturer doesn’t just buy more components. It redesigns production around repeatability, standardization, sequencing, quality control, logistics, and material flow. Homebuilding has adopted much of the manufacturing language. It has been much slower to adopt the physical production system behind it.
It is important to clarify this. Purchasing scale and production scale are not necessarily the same.
If I negotiate a window at a lower price but then deliver it individually to a job site 40 miles away, store it there, move it around, install it with a fragmented labor force, and absorb the delays that system creates, how much of my purchasing advantage did I actually preserve?
Buying cheaper is useful. Building cheaper is scalable.
Those are not always the same.
Maybe we need three definitions
Rather than talking generically about scale, it would serve the industry well to separate three things.
Corporate Scale is capital, technology, purchasing, data, overhead leverage, and national infrastructure. Century unquestionably has it.
Operating Density is the amount of profitable production pushed through a local or regional operating system. Highland demonstrates it regionally, while Bloomfield demonstrates it within a single metropolitan area.
And then there is Production Scale: the ability to make the physical act of building a house cheaper, faster, more predictable, and more productive as volume increases. That last one may ultimately be the most valuable because it changes the economics of the house itself.
Show me the economics
Everyone says scale creates an advantage. Fine. If it’s purchasing scale, show me purchasing savings per house. If it’s overhead leverage, show me SG&A per closing. If it’s capital scale, show me the return on that capital. If it’s market scale, show me profit per market. If it’s land scale, show me a better land basis or better inventory turns. And if it’s production scale, show me that the 10,000th house is actually cheaper, faster, or more predictable to build than the 1,000th.
Otherwise, we may be measuring size and calling it scale. That distinction matters. The number of markets alone doesn’t tell us enough, nor does the number of closings. The better question is what economic advantage those additional markets and closings actually created.
A better metric: scale density
Maybe the metric we are missing is what I would call Scale Density:
How much profitable production can a builder generate in each market it enters? Not merely closings per market. Profitable closings per market. That is an important distinction. Two thousand homes in one market aren’t particularly impressive if they require excessive incentives, bloated inventory, or poor returns on land and capital.
Likewise, 200 homes in a market could be highly attractive if the builder earns exceptional returns while deploying very little capital. The point isn’t that density automatically wins. The point is that density should be evaluated alongside the economics it produces.
This deserves more digging
Closings per market do not prove operating efficiency. They do, however, raise a question I don’t think the industry has adequately answered. If geographic scale creates an economic advantage, where does that advantage actually show up? Does it appear in SG&A per closing, gross margin, inventory turns, cycle time, land investment per closing, community absorption, purchasing costs, or return on invested capital?
And perhaps more importantly, what happens when we compare those measures not simply by company, but by Market Density?
I don’t have enough data to answer that yet, and I think pretending otherwise would weaken the argument. But the differences are large enough that the question deserves considerably more digging. If builders producing 800 or 1,000 homes in a market consistently generate better economics than those producing 200 or 300, we may be measuring homebuilding scale incorrectly.
If they don’t, that’s equally interesting. Then we should be able to pinpoint the specific economic advantages the geographically broader builder receives that offset the added organizational complexity. Either answer would teach us something. What shouldn’t satisfy us anymore is pointing to more markets and simply calling it scale.
The ultimate national builder may look different
This suggests a possibility I find more interesting than Century versus Highland versus Bloomfield. Maybe the ultimate national homebuilder doesn’t look like a single giant operating system spanning the United States. Maybe it looks like a collection of high-density regional operating systems connected by shared capital, technology, purchasing and data.
In other words: Leverage Century’s corporate scale. Combine it with Highland’s regional concentration. Add Bloomfield-like metropolitan density. Then figure out how to achieve genuine production scale at the house level. That could be a very different homebuilding company.
The goal wouldn’t be to occupy the most markets. It would be to develop an operating system that delivers exceptional economics in a market before replicating it elsewhere. Replication should follow a strong operating system, not replace it.
The builder with the smallest map
We have learned to measure homebuilders by the size of their maps. But a map isn’t a factory. A market isn’t a unit of production. A closing isn’t necessarily an efficiency gain. And buying more material doesn’t automatically mean building the house more cheaply. The next generation of homebuilding may not be won by the company that enters the most markets. It may be won by the company that achieves the greatest profitable production density in a handful of markets and then figures out how to replicate that operating system without destroying what made it work in the first place.
Century’s national footprint gives us one model. Highland’s Texas footprint gives us another. Bloomfield’s concentration in Dallas-Fort Worth gives us a third. None is automatically right or wrong.
But they force us to ask a question I think is much more useful than simply asking who is biggest: How much scale is actually enough?
Maybe the goal isn’t to own the biggest map. Maybe it is to build the most profitable houses with the fewest unnecessary miles, layers, handoffs, trucks, warehouses, delays and dollars.
There is a pretty simple Texas way to put that: Don’t get bigger just to get bigger. Get better. Then get bigger because getting better made it possible. That’s scale
