Why Cutting Boston Attic Insulation From R70 to R49 Moved the Model Under 1%

We cut a third of the attic insulation on a Boston townhome project, and the energy model barely moved. The project had a strict energy code to hit, so we figured it needed good insulation. Then someone on our team asked a simple question: "Why do we have so much insulation in the attic?" We cut it from R70 down to R49, and the model dropped less than 1%. That's almost nothing, and it saved the client a lot of money.

The Real Project

The project was a townhome development in Boston with a strict energy code target. When a code is strict, the instinct is to load up the envelope. More insulation feels like the safe answer, and nobody gets blamed for an attic that is too warm.

So the attic was specified at R70. That is a lot of insulation. It looked like the kind of spec that protects a project from failing the model.

Then the question came up in a design review. Why do we have so much insulation in the attic? Nobody had a number behind it. It was in the design because more seemed better.

This is a common pattern on strict code projects. Specs often get added to feel safe, then carried forward from job to job without anyone checking what they buy.

So we started cutting it back. R70 down to R49, and the model dropped less than 1%. That's almost nothing. The building still had a well-insulated attic, the energy model barely noticed, and the client kept a lot of money that would have gone into the ceiling.

The Diminishing Returns Problem

Insulation does not work in a straight line. The first few inches stop most of the heat. Each layer after that stops a smaller share of what is left. Heat loss through an assembly drops roughly with the inverse of its R-value, so every added R does less than the one before it.

By the time an attic is at R49, it is already blocking most of the heat that would move through the ceiling. Going to R70 does cut heat loss further, but it is cutting a small slice of a number that is already small. The cost of the extra insulation does not shrink the same way. You pay for every inch at roughly the same price, and each inch buys less.

The energy model sees the whole building, not just the attic. Heat leaves through the walls, the windows, the air leaks and the ventilation. It also gets used by the water heater, the dryer and everything else plugged in. Once the attic is well insulated, its share of the total is small. Making a small share smaller barely moves the total.

This happens more than you'd think. A lot of energy code money goes into upgrades that barely improve the energy efficiency of the home. They feel productive because they sound like efficiency. In the model, they are a rounding error.

Energy code money is limited. Every dollar that goes into an R70 attic is a dollar that cannot go into a measure the model rewards. On a townhome project, that spec repeats in every unit, so a small overspec becomes a large line item across the job. Cutting it does not lower the quality of the building. It moves the money to where it does something.

Energy Code Upgrades That Barely Move the Energy Model

The attic is not the only place this shows up. When we run each measure on its own and look at what it actually does in the model, the same pattern keeps coming back.

Thicker wall cavity insulation. Going from R21 to R30 in the wall cavity when you already have continuous insulation is about 1%. The continuous insulation is already doing most of the work, so the cavity upgrade adds very little.

Triple pane windows. Going from a double pane to a triple pane window is about 2%. Windows are a real heat path, but a good double pane window already handles most of it.

Extra attic insulation. An attic at R60 or R70 instead of R49 is about 1%. That is the Boston project in one line.

A slightly better heat pump water heater. A heat pump water heater at UEF 4 instead of 3.5 is about 1%. Both units are already very efficient, so the step between them is small.

An efficient dishwasher. An "efficient" dishwasher is basically zero, since they're all about the same.

Compare that to swapping a standard dryer for a heat pump dryer, which is 5 to 10% in the model. One appliance swap can outperform a whole stack of envelope upgrades. A standard dryer is a big energy user, and a heat pump dryer does the same job with far less energy. The model counts that difference, and it counts it big.

None of these upgrades are bad products. The issue is what they cost against what they return. When a design carries several of them at once, the money adds up fast while the model barely moves.

When Extra Attic Insulation Is Still Worth It

None of this means insulation is a waste. It means each upgrade should earn its place on the numbers. There are cases where the extra R-value is worth paying for.

The model is right at the line. If a project is passing by a hair, even 1% can be the difference between a pass and a fail. Then a small gain has real value, and the question becomes which small gain is cheapest.

The cheaper moves are already used up. If the design already has a heat pump dryer and the other high-value measures, the remaining gains may only be in the envelope. At that point an upgrade that buys about 1% might be the best option left.

The assembly is doing more than saving energy. Insulation also affects comfort and how an assembly handles moisture. Cutting insulation without checking how the assembly behaves can create problems the energy model does not show.

The cost difference is small. If going from R49 to a higher attic value costs very little on a given project, a small gain can still be worth it. The test is always cost against what the model says it buys.

The Decision Framework

The trick is to keep asking "how much are we actually getting out of this?" on every assembly and mechanical system. Run each upgrade on its own and look at what it moves. By the end of it, you'll usually see a stack of savings.

Ask the question out loud in design reviews. On the Boston project, the savings started because someone on our team asked why the attic had so much insulation. Nobody had to run a complex study to find it. They had to ask, then check the model. The habit matters more than any one answer.

Start with the expensive items that sound efficient. Thick attics, thick wall cavities and triple pane glass are the usual suspects. Then look at the measures that move the model the most for the money, like a heat pump dryer at 5 to 10%. Put the money where the model actually responds.

It takes a lot of work to find, which is why most people don't do it. Every assembly gets checked, every mechanical system gets checked, and every answer depends on the rest of the building. What was almost nothing on a Boston townhome could matter more on a building with a different shape, a different target or a different mix of systems. No two projects stack up the same way, and the only way to know where your money is being wasted is to model your building one measure at a time.