@DavidWilson has some very good points here and it’s what led to this massive thread:
Long story short, VOC readings from ‘low cost’ sensors are very difficult to interpret and shouldn’t really be relied on much other than looking for general trends. This, unfortunately, is common across all air quality monitors brands in this price range, whether Airthings, PurpleAir, AirGradient, or otherwise.
I can’t remember exactly what scale/index Airthings uses (I will need to double-check this), but most monitors use a 1-500 index, where anything below 100 indicates a decrease in VOCs and anything above indicates an increase. 100 itself is the baseline (a rolling average based on exposure over the past 12 or 24 hours).
Based on this, I believe Airthings may be showing an ‘absolute’ number in parts per billion, but this is just derived from the relative index I just mentioned so it’s not particularly useful. Of course, that’s before we get to VOCs themselves, some of which are very harmful but some of which are also harmless.
I don’t say this to add more confusion (I hope that’s not the case!), just to say that a VOC sensors are, unfortunately, quite limited in usefulness and I wouldn’t conclude that the purifier doesn’t work based purely on this. If this is a big concern for you, I would definitely recommend a VOC patch test where you basically place a collector in a room for a few days and then send it off to a lab for analysis. I’m not sure where you’re based, but if you’re in the U.S. I believe these are generally not too expensive and they’re definitely the best way to get conclusive answers.