I don't think it would be a reliable method at all. I don't know how big your system is but it takes a lot of RO water to move the needle. If you had a system with 100g in it at a PPT of 35 and lost 5 gallons, the PPT would only move to 36.8.
These conductivity probes only have an accuracy of +/- 1 PPT. So your 35 PPT system could read 34 or 36 and be within the accuracy of the probe. Now you've lost 5 gallons of water and your PPT only moved .8 PPT outside the accuracy of the probe.
And that's a lot of water to replace at one time. Most people like to put in ounces at a time to avoid fluctuations.
I wouldn't do it. I'd use floats instead. Way more precise IMO.