I also found this in the comments:
"Question:
about how accurate are the weeks the product gives you?
Answer:
The ones that claim they count how far along you are in weeks is completely a sham. Medical professionals measure pregnancy based on the last day of your last cycle, meaning by you next missed period you would be a month pregnant no matter when you conceived. There is litterally no way to tell someone is "2 weeks pregnant" when implantation can happen around that time. You really need a medical PROFESSIONAL to tell you how far along you are not a pee stick made in bulk by machines in some factory. Good luck!"
EDIT: Quotes around the copy paste from comments. I do not know if it is fair reasoning or not, but it makes sense to a simple man such as myself in my ignorance, perhaps.
I know next to nothing about menstruation, past that it happens sometimes to be honest, and is nicknamed "a monthly" at times of convenience. If the test is only usable 4 or 6 days before the next expected "period", with 99% accuracy only after the expected day, wouldn't that make determining pregnancy and scheduling an appointment within a 30 day window quite difficult? I must assume that appointments would take time to set up, unless it is much more streamlined than usual medical appointments.
EDIT2: I may be misunderstanding the comment. Maybe there is a counter in these digital tests to guess how far along the pregnancy is. However I do think it raises a valid argument about the window of time, possibly, and further necessitates a cycle of usage of the product should a woman be sexually active without desiring pregnancy, likely with the additional use of contraceptives I would assume but perhaps not always.
Thus it would still be useful to the discussion to determine a theoretically effective cycle of use, even if one must ignore for now what appears may be a potential for a very short window of action left open, though I am not quite sure of that yet without further insight from a greater mind on what that window would be exactly. However, I must say: ugh math, surely someone here enjoys it more than I do? There may be some variables to figure in to the equation, and perhaps it would be fair to give a reasonable example as well as the extremes if such turns out to be possible. However variables such as how times of fertility would effect a cycle of usage are beyond me, thus I suggest the extreme example of a simpler cycle of usage would be easier to figure out though of less utility. I am poorly equipped to understand such variables intimately, and so with my current knowledge I would be limited to a less functional example of calculating how often the testing would need to overlap to achieve efficiency without including things such as times of fertility; and on top of that I would likely fudge my math along the way somehow since I don't do math very often, thus my request for a mathematically inclined person to advise us.