"The client of an email process has specified that the % of A or B quality transactions should exceed 85%. The program is current operating at 75% of A or B scores. The client can tolerate a measurement error of 2% and would like to be 95% confident of the quality scores. The program is handling 16000 transactions / day in 16 hours with 120 agents.What should be the sample size for quality measurements? To what level should the quality scores improve to ensure that not more than 5% of transactions are inspected for quality?"

For the sample size i used the formula n = N*X / (X + N - 1),

where,X = Zα/22 *p(1-p) / MOE2,and Zα/2 is the critical value of the Normal distribution at α/2 (e.g. for a confidence level of 95%, α is 0.05 and the critical value is 1.96) and came up with 2088 sample size

I'm not sure how to answer the second question about to what level should the quality schores improve to ensure that not more than 5% of transactions are inspected for quality. I'd like to know if I answer the first part of the question correctly and also how to approach the second part of the question. Any help would be gladly appreciated! Thanks!