A partner book is not a CRM full of logos. It is a ranked list of companies with a workflow, an owner, and a next step. If those three fields are empty, you do not have a book. You have a spreadsheet of hope.
Mine before you invent
Open last quarter’s call notes, tickets, lost-deal reasons, and feature requests. Every time a tool is named, tag it. Count. The tools that show up more than twice are the start of the book.
Then look at the marketplaces your buyers already shop. A category with a weak listing and a workflow you can own is a distribution opening.
Do not start with a brainstorm on a whiteboard of “strategic logos.” That list is what everyone else is also chasing, and it is usually not what your customers asked for.
Five fields, no more
For each candidate: company, workflow in one sentence, demand (count of customers or deals), distribution (marketplace or not), next step. If you cannot write the workflow, delete the row.
Cap the active book. A team of one can honestly run five to twelve live relationships. The rest is a watchlist.
Owners or it does not count
A row without an owner on their side is a lead, not a partner. Your first job after scoring is to find the partner manager, the PM, or the founder who can put engineering on a date. Until that name exists, do not promise sales a listing.
Keep it ugly
A Notion table or a HubSpot list is enough. Do not wait for a partner portal. Portals become where information goes to die if the book is not already clean.
Revisit scores when the product moves. The book is a living ranking, not a plaque.