Hi Abhi!
Thanks for the note.
WRT 1. YES. If you don’t understand how your ML system works, its RoI, you can’t put it out there with new customers. In a way, this is like any big new product feature you launch. The downside, though, is that ML features can be disruptive (in a negative) if rolled out wrongly. It can create incorrect data, can bias your systems and really hurt your business. So test extensively with a smaller base before rolling it out. Unless, it’s simple automation of workflows.
- Pricing entirely depends on value add. If your ML add-ons are providing incremental outcome of some kind, then customers will be willing to pay more. So test your hypothesis like you do with any other product feature. Talk to customers, build, iterate and evaluate value before pricing it. 2 things I’ve seen: customers pay for incremental growth in revenue or costs saved by AI.
Good luck!