François Mansart (1598–1666) helped shape the distinct elegance and grandeur that define French Baroque architecture.

His crowning achievement, the Château de Maisons, still stands today as a testament to his genius and a blueprint that would inspire generations of French country-house design.

I know of François Mansart because of the Mansard Roof (spelled differently for some reason).

He didn't "invent" the roof style; the steep double-pitched roof type existed before him. However, he adopted the style and utilized it craftily with the classically designed structures his roof adorned.

It was only through my meandering that I learned Mansart was an uncompromising perfectionist. So much so, he reportedly had works demolished and rebuilt when he found them unsatisfactory. That practice earned him a spectacularly expensive reputation, as you can probably imagine.

It seems that Mansart had clients with unlimited budgets, and he managed to exceed them.

More on that, along with a tool I created inspired by perfectionism, but first, a quick look at the roof style that ingeniously adds a story within the roof structure.

A mansard roof is a four-sided roof with two slopes on each side—a steep lower slope and a flatter upper slope. In plain terms: it's a roof that looks like it has a "second story" because the lower slope is so steep it appears almost vertical.

Unlike the gambrel roof (named after a horse's hock), which is a two-sided roof with two slopes on each side and a gable at each end.

Yet, both are composed of a steep lower slope and a much flatter upper slope. The steep lower slope creates the illusion of an additional story, while the flatter upper portion provides usable attic space.

Another distinction I've found: the mansard roof nearly always has dormers.

Looking for the perfect "formula," I found that there isn't an exact pitch for each of the slopes, but rather a range that gives the designer freedom to fine-tune the roof with the proportions of the structure below.

The steep lower slope typically ranges from 60° to 80° (almost vertical), creating the dramatic visual effect. This steep angle makes the roof appear to be part of the building's wall.

The flatter upper slope is usually between 5° and 15°, providing a nearly horizontal surface that maximizes usable attic space.

The transition between the steep upper slope and the flat lower slope — commonly referred to as the "curb" or "break" — is a defining design element that can be either accentuated or softened depending on the desired aesthetic. In practice, however, it is frequently omitted in the interest of cost savings and the pursuit of affordable housing. Whether that trade-off is worth making is something I'm still weighing up.

Mansard roofs come in straight, concave, and convex forms. The greatest collection of photos I found that illustrates the different styles is "Buffalo as an Architectural Museum." Buffalo, NY, seems like an unlikely source, but there was a time when Buffalo had more millionaires per capita than any other city in America. There're grand estates along Delaware Avenue.

Anyway, as I mentioned earlier, Mansart was known for having works demolished and rebuilt when he found them unsatisfactory. While I can only find one well-documented example of this, the Château de Maisons, it speaks volumes about his uncompromising pursuit of perfection.

The fact that he was willing to tear down and start over until he got it right is a remarkable testament to his vision and standards, as well as his ability to convince the client to fund it.

What an impressive way to carve out a reputation and establish a niche.

Mansart was almost too large for life. He had colossal ideas that captured the imaginations of the rich. But I've learned that not every client approved his "change orders."

In 1645, Anne of Austria hired Mansart to build a magnificent church and monastery complex to fulfill a religious vow. When the costs of merely laying the foundation exceeded the queen's entire provided budget due to his persistent modifications, she refused to continue funding his unchecked scope creep.

Effectively rejecting his structural "change orders," the Queen dismissed Mansart from the project and replaced him with Jacques Lemercier, who agreed to finish the project using Mansart's core design but under strict financial management.

However, the pendulum swings both ways.

History offers examples of architects who found themselves guarding their creative vision and had to treat the client as their own worst enemy.

To protect their core vision, these architects used everything from subtle psychological manipulation and outright deception to aggressive contractual threats.

For example, when Edgar Kaufmann Sr. commissioned Frank Lloyd Wright to build a weekend home in rural Pennsylvania, Kaufmann wanted the house positioned facing the beautiful waterfall on the property so he could look at it.

Wright completely ignored this request. He famously designed Fallingwater to sit directly on top of the waterfall, arguing that the Kaufmanns shouldn't just look at the waterfall, but live with it as an active part of their lives.

Of course, Château de Maisons and Fallingwater are examples of extremes. But chances are, you’ve encountered similar tensions in your own work, even if the stakes were a little lower.

Once a project begins, you know that every change has a ripple effect. Yet, formal Change Orders are often skipped or overlooked.

And even when they are used, standard Change Order templates, including the one found in the AIBD Designer Docs, fall short. They fail to help the Owner, Designer, and Contractor anticipate that ripple.

Today, September 23rd—the 360th anniversary of Mansart’s death—I propose a better approach: the “Mansart Risk Style” template.

Before you change the plan, change the conversation.

A Change Order should not merely record what changed. It should reveal what the change will change.

When there is a conversation, it rarely begins with, “We have a major problem.”

They begin with something much friendlier:

None of those requests is unreasonable. In fact, some might be exactly the right decisions.

The trouble begins when the team approves a change by looking only at the immediate price or only at the immediate convenience.

A “similar” window can change sightlines, frame depth, impact performance, finish durability, and the entire character of an elevation. A cost-saving material substitution can increase maintenance. A field fix can create a structural, MEP, permit, or sequencing problem three trades later.

François Mansart offers a useful reminder. The lesson for us is not to avoid ambitious design. It is to make decisions that allow the project to remain coherent when budget, timing, availability, and construction reality intervene.

That is why I've created the Mansart Risk–Style Change Order.

It retains the essentials: a clear change description, contract-sum adjustment, time adjustment, and signatures. But it also asks the questions a standard form may leave unasked:

• Does this change alter the design intent or composition?
• Is it funded by an allowance, contingency, or additional Owner money?
• Does it affect procurement, sequencing, or substantial completion?
• Does it require a permit revision, consultant review, or AHJ approval?
• Is the “equivalent” material truly comparable in appearance, performance, weathering, and maintenance?
• What other alternatives were considered?

The form also gives the Owner a useful comparison: proceed as proposed, preserve the primary design intent through another solution, or defer and phase the work properly.

That protects the Owner from a rushed contractor substitution. It also protects the Owner from a designer’s well-intended but unnecessarily expensive recommendation. And it protects you by ensuring that everyone sees the cost, time, technical, and design consequences before the work begins.

​Download the Mansart Risk–Style Change Order​

AIBD members will soon find an editable file in the Designer Docs.

To be candid, I have not used this new version on one of my own active projects. So I will not tell you it has already prevented disputes or saved a client money.

What I can tell you is that my goal is to force the conversation many of us wish had happened earlier: before the “equivalent” product was ordered, before the framing was changed, before the contingency quietly disappeared, or before a deferred phase left an awkward and expensive problem behind.

If it helps you and your client ask those questions before approving the work, it has already done something valuable.

Reply to this email and let me know what you think.