Sorry for the late reply!
sadly with all inductive clamps, they utilise a design which inherently suffers from drift.
First battery voltage will drop over time, and for our task when we learnt all of this we were doing 24 hour and longer parasitic drain tests, so 50-500ma range for long times. Tried using multi battery and larger battery configs but these all suffer from voltage drop over time, and this affects the output. Even tried a mains based stable output supply, and this reduced the drift but not entirely.
Second the "press button" zero style clamps seem to suffer from fast and more drift than the rotate knob user adjusted style, but regardless both still drift! It's in the design! (unless anyone can tell us all otherwise!)
Third we tried (and got inconclusive results, and bored!) using no zero, and using a maths channel to zero in software (although you might be able to do this in the probe config now?), but I'm sure in our tests we still saw drift.
We tried a 10 loop of thin wall cable to allow amplified current reading without amplifying the drift, so the drift would be 1/10 of normal and this worked well for some 4 hour based time frames.
For long time frame measurements of low current we moved to "volt drop across fuses" method (see another post here https://www.picoauto.com/support/topic12061.html
), which allows for long term measurement with high accuracy.
see also by Steve Smith:https://www.picoauto.com/library/training/parasitic-drain
and VW technical bulletin 270711 with a downloadable PDF in the page here:https://wiki.bentleypublishers.com/x/AoD3B
So basically your clamp is probably fine and behaving as they all do!
If you find any way to stabilise the clamp or another method please let me know