The variance of the attribute is a function of p*(1p), so taking a purposely large value of p is fine, as taking p=.5 to obtain the largest variance is too conservative (labor intensive).
Better idea: construct a loss function balancing the cost of labor and the potential loss of sales, profits, upon misestimation.
Another approach: A two stage sample size estimator (Stein). The 1st sample estimates p and incorporates this value into the second stage sample size estimate.
