The problem with the undissolved crystals is that it doesn't work that way as an indicator because the water does NOT get saturated with DPD (or with other powder contents). If you add very fine powdered DPD to the water, you can add more than 2 scoops and not get any undissolved crystals after a swirl. I've done this by intentionally scooping out the most powdery part from the container. So what do you do, keep adding scoop after scoop after scoop where it all keeps dissolving?
At the other extreme, you have a large big crystal clump which you add and then stop because you see something undissolved. However, this doesn't put enough DPD into the water.
I must be missing something because I don't see how seeing undissolved clumps means anything useful except that it tells you you've added something that hasn't fully dissolved. If anything, you need to add more powder beyond the amount if it were to have fully dissolved (or you need to put something in to crush the clumps to help them dissolve faster). That is, the amount you add should be an amount that gets into the water, not that sits there at the bottom undissolved.
This MSDS shows the components of the DPD powder where every single item is very soluble in water. The dye itself is as a soluble salt, N,N-Diethyl-p-phenylene-diamine sulfate, which as shown in
this link has solubility of 0.1 g/ml and
this link says 1000 mg/ml (1 g/ml) so let's use 0.1 g/ml. According to the MSDS, the dye is less than 2% of the weight. A 10 ml sample should be able to hold at least (10 ml) * (0.1 g/ml) = 1 gram of dye which would be (1 gram)/(2%/100%) = 50 grams of powder which is of course more than the entire 10 g container. Likewise, the solubility of the main components of the powder, potassium and sodium phosphate (see
this table), are very high at 92.3 g/100g so 0.923 g/ml and 12.1 g/100g so 0.121 g/ml so in 10 ml this would be 9 g and 1 g respectively. At <65% and <35% of the powder, that's 14 g and 3.5 g, respectively so
the limit for saturation of the least soluble component appears to be for sodium phosphate at 3.5 grams or about one-third of a 10 g container in 10 ml.
If one doesn't want to waste powder, then crushing any clumps back into powder so that they quickly dissolve and so one need not overdose would be the way to go. In fact, do this experiment: take a mortar and pestle and thoroughly grind to a fine powder the entire contents of a 10 g container. Then take a 10 ml water sample and add powder to it swirling each time. See how much powder it takes before the water is saturated and adding any more powder simply sits on the bottom undissolved. You may find that you can dissolve about a third of the 10 g container in 10 ml. Note that some DPD may get oxidized by air and that part might not be soluble, but it will also be useless, but it should be a very small amount if you crush to a fine powder.