The sponsor of a multi-division rollout sent us the framework before we sent him ours.
He laid out four options for deal stages across eight divisions: common stages for everyone; a few anchor stages that every division shares, with division-specific stages in between; independent pipelines per division with a canonical-stage property that maps each to a shared reporting stage; or fully independent pipelines with no shared reporting. Then he asked the three questions that actually decide it. His framework is better than most partners', and this decision gets made badly in almost every multi-brand implementation I've seen, so here it is.
Three questions decide it
How consistent do stages need to be across divisions for portfolio reporting and stage-to-stage conversion? If leadership wants one funnel view with conversion rates that mean the same thing everywhere, stages have to line up somewhere. If division-level flexibility is the priority, there's room. You can't have both by accident, only by design.
How does stage-skipping work for transactional deals inside a project-heavy division? One division sells engineered projects that run through months of stages. The same division also takes small book-and-turn orders that go from quote to close in a day. Force those through project stages and reps skip stages, which corrupts every conversion metric. Give them a separate path and now one division has two pipelines.
Are internal handoffs stages at all? "Cost accounting," "cost ready," "engineering review." They're steps, so they feel like stages. They're work queues. Put them in the sales pipeline and your funnel reports measure internal throughput, not customer progression.
Why the canonical-stage property usually wins
Common stages are the cleanest reporting and the right answer when divisions genuinely sell the same way. Most don't. In this group one division is mostly purchase-order business, another averages large engineered deals, and a third turns small orders daily. Common stages would be a fiction for two of the three.
Fully independent pipelines respect the differences and destroy portfolio reporting, which for a PE-owned group is the point of the project.
The canonical-stage option, each division with its own pipeline and a workflow that writes a shared "reporting stage" property based on where the deal sits, gives divisions their real process and gives leadership one funnel. The cost is a mapping table and the discipline to maintain it when a division adds a stage. That's a small price for not forcing an engineered-projects team and a transactional team into the same stages.
Anchor stages are the compromise when you want the shared milestones visible in the pipeline itself rather than in a property. They work, and they get messy when a division's process legitimately doesn't pass through an anchor.
To be fair to common stages: if you're rolling out to divisions that were all on the same CRM with the same process before, they may already be aligned, and adding a canonical property would be overhead for nothing. The framework is: answer the three questions honestly before you pick. Sometimes the honest answer is common stages.
Decide this before the build, not during. Every report, every dashboard, every conversion metric inherits it. Which option are you on, and did you choose it or inherit it?
This came out of a client project. The story behind it, and what to check in your own portal, goes to subscribers on Thursdays. Get Thursday's note.
by Perry Nalevka on October 06, 2026
CEO of Penguin Strategies
