Adding a datapoint from our coworking build: phone dB apps aren’t lab-grade, but they’re great for comparisons if you control the setup—same spot, same time, A-weighting, “slow” response. We tracked before/after for a tighter door and later for sound masking; the deltas lined up with what people felt. If your app shows frequency, note whether the pain is low rumble or speech band—that decides inserts/seals vs treatment. I also did a quick consult with https://newyorksoundproofing.com/ they translated our notes into “holes first, then heft” (seal doors/windows, then consider added mass) and it matched what the phone logs hinted at. TL;DR: use the app to rank problems, not to chase a magic number.