My thought here is that once you reduce volumes of testing, the equipment included in the test kits is not suitable for accuracy. For instance the vials or beaker used may not perform linearly at low levels, and the judgement of the meniscus becomes much more important as a source of error. The number of drops is important in titrations because as the number of drops required to determine endpoint is reduced so the resolution decreases. Similarly, the dropper bottles will not dispense drops of equal volume, they are engineered for testing average over several drops. Add to this the fact that the drop size depends on the particular solution being dispensed (see later).
The only way to test consistently and with good resolution with reduced volume sampling would be to move to scientific glassware and syringes (for both pool water and reagent dispensing) which would cost ten times more than buying the reagents, and testing would be a lot more cumbersome. You would also need to know the concentrations of the reagents and specific reactivity for calculations to move from counting drops to counting reagent volume. A decent syringe or burette for measuring 5 ml (glass) might be $50, one for measuring fractions of a ml (glass) perhaps $80. An easier dispenser with disposable tips can be about $300.
Having said that a 5ml medical syringe will dispense close enough, however the resolution of the test will be compromised because it is a drop basis test using a specific concentration of reagent.
I believe (I hope) that the Taylor folk did all the work to make kits that could be used easily and with reagent concentrations, dispensing, and test vessels that enabled them to be sold for somewhere between $50 and $100, be easily used, and be of appropriate quality for the task at hand. In this they most likely use spectrophotometers, other instruments, volumetric glassware, employ technicians, and engage manufacturers in the development of their vials, dropper bottles and so on to have appropriate manufacturing consistency. They performed some development relating to drop volume and concentrations of the reagents so that the particular tests could be performed on a drop count basis.
