I have two mental models for understanding and improving building performance. The first model I’ll call “Everything is Sort of Bad”. This model says that a building is pretty bad – but not horrible – pretty much everywhere and in most of the building systems (thermal boundary, pressure boundary, ducts, water-resistive barrier, etc).
The second model I’ll call “Everything is Sort of Bad but at Least One Thing is Awful”.
The first model is helpful. There are lots of areas in buildings where slight improvements yield large benefits. Attic air-sealing is usually a pretty good example of this. Cracks at the drywall ceiling in attics tend to be fairly uniform. Yet, plugging all the holes in an attic will yield a significant improvement. Usually these cracks add up to a 20%-30% overall leakage reduction, or no less than 2/3 the potential improvement on one of our projects. In this is a model geeky diagnostic tools are pretty helpful. Blower doors and duct blasters are great for verifying the effect of these measures. However, it’s often hard to tell if the marginal improvement left hard-to-find problems. A leakage reduction of 35% is great, but doesn’t tell us whether the cantilevered 2nd floor joists accessible only from a small roof in the garage were effectively sealed or not.
The second model includes all the lessons of the first, but tells us to be on the lookout for egregious performance violations. When looking for isolated but big problems, it’s unfortunate, but you may have to actually listen to the people who live in buildings.