Loading
Stages are the one decision that is expensive to change later, because every deal, report and automation depends on them. Write exit criteria for each: something the buyer has done that you could verify.
Five or six is plenty. Every extra stage is a decision somebody makes every week for the rest of the year.
Bring the open pipeline and its accounts. Closed history can be archived somewhere readable rather than migrated, and old data is exactly where duplicates and dead contacts come from.
Run the import with a diff you approve before it commits, and resolve the rejects rather than ignoring them.
Days of work spread over weeks. The bottleneck is deciding, not configuring.
Usually not. Open pipeline and accounts; archive the rest somewhere readable.
Starting with reporting requirements, so the first experience of the system is unpaid admin.
Half an hour with your own data usually saves reading three of these. The guides will still be here afterwards.