When would a columnstore index be unsuitable for an OLAP environment with large and wide tables?

Suppose that your SQL Server 2019 machine has a query that hits a disk-based table that is both long (let's say 3,000,000 rows) and wide (let's say 150 columns). Suppose that you retrieve most of the rows and most of the columns. Finally, suppose that the table is infrequently changed. Consider it a typical reporting server table that is only updated once a day. **Are there any circumstances at all where the above conditions apply and a traditional rowstore index would be preferable to a columnstore index, clustered or otherwise?** I worry that I have been sold too easily on columnstore indexes and will begin using them on all of my large and wide tables.