1) It would be helpful to have ability to pivot on any enumerated field in the dataset. This would allow someone to get better transparency on the segment makeup and can also be used as a QA tool to ensure that counts match user expectations. For example, current capabilities provide a high level count but if I wanted to get a count of contacts by job title or by region then I would have to export data and analyze it offline. 

2) Additionally, it would be great to allow user to dedupe a segment based on different columns in the dataset. For example, let's suppose I have a dataset of contacts from my CRM. If I am sending dataset to an email platform then I will want to dedupe on email address but if I am sending dataset for a postal campaign then I will want to dedupe on mailing address. I would envision giving user ability to select 1 or more dedupe columns and then also allowing someone to select 1 or more columns to prioritize. This could then be used to build out a cte column where you partition by the columns selected for dedupe and order by the columns used for priortization.