Calculate them on a cohort, not on a snapshot. Take every opportunity created in a month, follow that same set forward, and measure what share reached each later stage. Snapshot ratios (stage two divided by stage one today) mix cohorts moving at different speeds and will mislead you. Once you have the curve, do three things: find the single worst step, since one stage usually leaks far more than the rest and that is where your attention belongs; measure time in stage alongside conversion, because a stage where deals sit for eleven weeks is a qualification problem wearing a conversion costume; and use the rates to work backwards into how much pipeline each rep needs, which turns a vague target into an activity plan.
3 hand-picked resources, 3 link-checked. Pick how you want to dig in.
Why we picked it The clearest single document on what a revenue data model should look like end to end, from lead through renewal and expansion. It is the vocabulary most modern RevOps teams have quietly standardised on.
From Winning by Design by Winning by Design 46 page PDF
Open winningbydesign.com →Why we picked it It walks through building a Stage Management Guide, which is the single artefact that turns a messy pipeline into a forecastable one. Directly usable as a template.
From First Round Review by Derek Draper 15 min read
Open review.firstround.com →Why we picked it Explains why the 3x rule of thumb is wrong for most teams and shows how to derive your own coverage number from your actual win rate. Weighted versus unweighted coverage is handled properly too.
From Clari by Jess Richter 11 min read
Open clari.com →How forecasting and gtm metrics reads from a different seat.
