Running a pilot space
A pilot is not a smaller migration. It is a rehearsal with a review at the end, and the review is the part that matters.
Pick a space with real content, migrate it, look at the result properly, and roll it back if you do not like it. Then plan the rest with evidence.
#The loop
#Choosing the space
Real content, not a test space. A sandbox tells you the tool works; it does not tell you what your estate does.
Something with diagrams in it if you have any. Diagrams are where the surprises live.
A space with an owner who will look at the result and tell you if it is wrong.
#What to check afterwards
Check | Why |
|---|---|
Open ten pages and read them | Rendering problems are obvious to a human and invisible in a log |
Look at a page's history | Confirm the previous version is there with the original macro |
Check a diagram opens and edits | A diagram that renders but will not open is a different problem |
Check the failure groups | Every group with a hint is a decision you will have to make at scale |
Keep source macros is worth switching on for a pilot and switching off for everything else. It puts the old and new macros side by side so you can compare them, at the cost of leaving duplicates behind.
#A few things that catch people out
Roll back the pilot if you want a clean slate, using the log from that run. It restores content, not assets.
Time the run and note the pages per minute. That figure is how you budget the real migration.
Do the review with the space owner. They will spot things you will not.
#Related
A rehearsal with a review. The review is the point.
