Start earlier than feels necessary. The data work is the long pole.
Most teams treat planning as a series of decisions made in a room. The decisions take days. Getting the data into a state where the decisions can be made takes weeks, and that is the part people underestimate.
If the CRM cannot tell you win rate by segment and source cleanly, that is where planning actually starts.
Ten to twelve weeks out: audit the data. Can you produce win rate by segment, real cycle length, and territory potential without anyone cleaning a spreadsheet by hand? If not, fix that first.
Eight weeks out: build the market model. Six weeks out: territory design. Four weeks out: capacity and quota. Two weeks out: comp reconciliation and the sensitivities. One week out: the plan is presented, not assembled.
Territory design slips when account scoring is not ready, which slips when the data is dirty. Quota slips when territory slips. Comp slips last and gets rushed, which is why comp and the operating plan so often disagree.
The whole chain traces back to whether the underlying data was usable in week one.
Sales leadership for territories, finance for capacity and quota, and whoever owns the CRM for the assignment rules. If the CRM owner is only brought in at the end, the plan becomes a document rather than a system.
Ten to twelve weeks before the year turns, and the first two weeks should be spent on whether your data can support the decisions rather than on the decisions themselves.
An honest audit of whether you can produce win rate by segment, cycle length and territory potential from the CRM without manual cleanup. Everything downstream depends on it.
Two to three weeks once account scoring is ready. The scoring is what takes time, not the drawing.
The data is not ready, so every stage slips, and comp reconciliation gets compressed into the last few days. That is why comp and the operating plan so often pay for different things.
It can be done, but the output is an estimate presented as a plan. The cheaper path is to fix the few fields the model depends on rather than all of them.
We use cookies to run the site and to understand how it is used. You can accept all or necessary only. See our privacy notice.