I like how easy it was to bring in new data into my cluster, as well has updating existing tables with various changes easily without downtime or issues of stability/reliability (as long you are doing your homework ahead). It's also great when wanting an ETL tool for processing big loads from csv files coming through APIs etc.. Being able update our main table fairly quickly! Sometimes there can be some confusion about what needs syncing across servers vs just local copies which could cause errors if not fully understood at first glance - especially after having used other DBMS' tools over time where this wasn't such issue due their workflows already built around that model; however clickhouse does its job very nicely so i think most people will find they don’t really have anything bad say against them here either way once using enough times myself.