By measurement, not by the hum. Every install here ends with a verification retest of at least 48 hours, run after the system stabilizes, placed where the original test sat. The written result is the proof, and it goes in your records.
More detail
A running fan proves airflow, not air quality, and the gap between the two is where bad installs hide. The verification test runs under the same closed-house protocol as any short-term measurement, with the system on, and the target is simple: below 4 pCi/L, and usually well below. Most properly designed systems land under 2. If verification comes back high, the finding is diagnostic, not a shrug: it usually means part of the slab is outside the suction field, an unsealed path is feeding the fan house air, or a second foundation level was never addressed, and the fix belongs to the install, not to you. Keep the result with the quote and the startup manometer reading. Years later, that trio is what tells you whether a drifting gauge means anything. See reading the manometer.