Pipeline arithmetic: a quantitative framework for capacity, conversion, and the rate-versus-volume decision

1. The model

Let the monthly target be Q qualified meetings. Four rates govern the system: the qualify rate q (held meetings that qualify), the book rate b (conversations that become booked meetings), the conversation rate c (contacts reached that become conversations), and the reach rate r (contacts worked that are ever reached across the full sequence). Required monthly volume is then

V = Q / (q × b × c × r)

The structure matters more than any particular number: the rates multiply. A system of four rates at 70, 20, 15, and 40 percent transmits only 0.84 percent of worked contacts through to qualified meetings. Small rate changes therefore move the required volume disproportionately, which is the entire managerial content of the model.

Contacts worked 1,000

Contacts reached 400 r = 40%

Conversations 60 c = 15%

Meetings booked 12 b = 20%

Qualified meetings 8 q = 70%

2. The central consequence: rate beats volume

Differentiate the system informally. Doubling worked volume doubles output, costs double the hours, and saturates at the team’s capacity ceiling. Raising any single rate by a factor k divides required volume by k at zero marginal hours, and the benefit recurs every month after. Volume is an expense; rate is an asset.

The sensitivity is steepest where the rate is lowest, which gives the operating rule: the weakest stage is always the highest-return work. A team converting conversations to bookings at 15 percent should not buy more data; it should fix whatever makes its conversations end without a calendar commitment.

3. Where the rates come from

A rep’s first month runs on team averages; thereafter the rates must be the rep’s own, measured from the system of record rather than recalled. Self-reported rates drift optimistic, and the drift concentrates exactly where it does the most damage, in the qualify rate. The model is only as honest as its stage definitions, which is why the framework binds them: a conversation is a two-way exchange, a booked meeting is calendar-accepted, and a qualified meeting requires a problem in the buyer’s words, an impact with a timeline, and a committed next step. Definitions loose enough to flatter the funnel destroy the math that plans the quarter.

Reference distributions matter for a second reason: a rep cannot know which of their rates is weak without something to compare against. Team averages serve first; the annual benchmark series provides the cross-company distributions.

4. Dynamics the static model omits

Two corrections matter in practice. First, signal decay: reach and conversation rates are not constants but functions of recency, falling within hours of a trigger event (Oldroyd et al. 2011). The same sequence run against a fresh signal queue and a stale list will report different funnel constants. Second, capacity interaction: volume pushed past a rep’s attention budget degrades message quality, which lowers c and b; teams that respond to a weak month by mandating more activity often buy the opposite of what they intend. Orchestrated execution changes the binding constraint from rep hours to judgment hours, which raises the ceiling but does not repeal the arithmetic.

5. Managerial use

The framework yields one number per rep (the daily worked volume their own rates require) and one priority per rep (their weakest rate against the reference distribution). Goal-setting research is specific that targets of this shape, specific and proximal, outperform exhortation (Locke and Latham 2002). The weekly inspection then has exactly two questions: did the volume happen, and what moved the weakest rate.