• You should identify which stats need updating rather than updating all the stats. You probably have some queries that perform poorly due to stats out of date and you should start your investigation from those queries.

    Take the query and look at the actual execution plan: do you see wildly inaccurate estimates? Then you need to understand where they come from. Do you have new data in the tables which is not accounted for? Maybe you should enable trace flag 2389 and/or 2390 to enable statistics branding (useful for ever increasing columns).

    If you have huge tables with stats that are not updated due to the inability to reach the 20% threshold of modified rows, maybe enabling the trace flag 2371 will help you (it enables an adaptive threshold for big tables).

    If you have stats that are not representative of the data unless based on a 100% sample, you should disable auto updating them by using NORECOMPUTE and set up a manual update process using a SQL Server Agent job.

    You can read more about TF 2371, 2389 and 2390 here: https://www.simple-talk.com/sql/database-administration/statistics-on-ascending-columns/

    -- Gianluca Sartori