Statement of null hypothesis Exercise 5.9

First, I want to say thank you for providing the PDF version of the presentation and for your availability to answer and clarify concerns. 

I am still confused about the best way to state H0. In exercise 5.9, I calculated with  “Hypothesized diff =0”  and drew the same conclusions (“we reject the null, so Extreme Ajax+ is a significantly better product”). I investigated pulling the same conclusion by changing the “Hypothesized diff” to a different value.  With a Hypothesized difference of 4, “H0: μExtremeAjax+ - μAjax ≤ 4” can be rejected at 5% significance, and we can draw the same conclusion as before. However, for Hypothesized diff = 5 or higher, H0 can no longer be rejected at 5% significance (p_value = 0.06), that is, accept "H0: μExtremeAjax+ - μAjax ≤ 5.” This indicates that our initial conclusion is not all-inclusive. 

My point is that  I am not clear on how to state H0 effectively.

I am not sure if I am conveying my concerns adequately. But I want to understand how to state a good H0 effectively.

Best, Charlemagne