Ask HN: Updating ML Models for changing market conditions?

I'm wondering what ML teams across companies who are building/updating models to make them perform well in volatile markets too? Especially if the predictions of the model might be directly impacted by a changed market? Some of my models were built with data over the last 10 years which was a bull market. For example, using sales data from 2010-2020 (but not using any financial/macro economic data directly as features). Any advice for best practices to update such models to make their predictions valid even in uncertain market conditions? Thanks in advance!