Sorry about that, I'll explain.
In their testing they used one group of protozoa (the same animal group as crypto and giardia) as test animals in laboratory conditions. They used a certain amount and it did deactivate (kill) them. Therefore they assumed that because those protozoa were killed, others would be as well. Also, it might work fine in a lab with a known, set quantity of "washed" organisms in a sterile background (the carry water the protozoa are living in), but they did little testing that approximates real world conditions. Other factors, such as water turbidity and composition will greatly impact how ultraviolet units work, and to not test using that real-world background makes me nervous. Something else that glared at me while I was reading the material was they never brought the unit to failure to see how much it could handle, which is standard for anything engineered, including every ultraviolet unit I have operated (low, medium, and high pressure units).
An anology would be that every tent keeps you dry when it's inside a garage with a light sprinkler on it. But get it in a wind and rainstorm on the top of a mountain and it might not be up to the task.
But, they do something similar with car alternator testing. They check the output at 1,000 rpm, then 1,500 rpm, then 2,000 rpm, and plot those numbers on a graph. Then a progression is used to plot the curve the rest of the way out, and the output of the alternator is determined from that curve, not from actually running it to see its actual output. Some companies do this, but the higher end ones don't, I'm just using it as an example.
While the testing data they generated looked fine, these are the items that jumped out at me, coming from a water treatment/backpacker standpoint. It may work fine and I love the concept, these are just the items I was concerned about.