That is the real issue, as you approach pure H2O pH becomes less and less meaningful as a parameter and very small additions of acids <"
and bases"> cause <"
huge variations in pH">. I use <"
shell attrition"> in my Ramshorn snails as a rough proxy for dKH (and therefore indirectly pH) now.
I agree with @brhau you can get around the problems in measurement to some degree by adding a <"
neutral salt (usually NaCl)"> to the water sample you are testing. It doesn't change the pH (the Na+ and Cl- are charge balanced) but it does mean there are more ions in solution and the meter stabilises more quickly.
You still only have a pH "snap-shot", which is the advantage of the snail shells, they are a real time data logger.