Pipeline conversion math — the per-stage funnel and unit economics.

The funnel from send to closed-won

A B2B outbound motion decomposes into five conversion stages. Each has an empirical range, and each has a single dominant variable that determines where in that range a given team lands. The cumulative product of the five rates determines the revenue outcome of the upstream investment.

The product of those midpoints — 50% × 2.5% × 20% × 50% × 22.5% — yields a send-to-closed-won conversion of roughly 0.028%. The product of the floor values yields 0.018%. The product of the ceilings yields 0.108%. A roughly 6x spread across operators is the steady-state observation, and the spread compounds disproportionately at the middle stages.

Per-stage variance and what drives it

Each stage's range is wide because the underlying variable is not a property of the market but a property of the operator. A 40% open rate and a 60% open rate represent the same prospects receiving mail through differently-configured sending estates.

Stage 1: open rate (40–60%)

The dominant variable is deliverability. The relevant inputs are SPF, DKIM, DMARC posture, the warmup ramp completed before sequence activation, per-domain volume relative to its warmup ceiling, the bounce rate at sequence entry, and suppression discipline against role-based addresses. A team with a deliverability floor below 40% is running a fundamentally different campaign than the one its operator intends — the open metric reported in the sequencing tool reflects pixel fires on delivered mail, not the larger denominator of mail that never reached the inbox.

Stage 2: reply rate (1–4% of open)

The dominant variable is copy quality measured against segment specificity. A generic prospecting message to a generic list converges to the 1% floor. A segment-specific message with a credible point of view on the recipient's stated problem moves toward the 4% ceiling. The variable is not the operator's skill in writing prose — it is discipline in maintaining a list narrow enough that a single message can be specifically relevant. ICP discipline (cluster reference: ICP chapter 03 hypothesis testing) is the upstream determinant.

Stage 3: reply-to-qualified-meeting (15–25%)

The dominant variable is the reply-handling discipline established across the prior six chapters. Classification accuracy determines which replies receive which response. Routing latency — the four-hour window — determines whether the reply is engaged while intent is still present. Objection-handling depth determines whether soft-pass and objection-with-signal replies convert at the rate of positive replies or at the rate of cold outreach. At operators with no reply-handling discipline this rate sits below 5%. At operators with the full stack — Slack routing, four-hour SLA, per-objection response library, calendar-link handoff — the rate clears 20%.

Stage 4: meeting-to-opportunity (40–60%)

The dominant variable is AE quality and discovery rigor. A meeting that produces an opportunity is one where the AE qualified budget, authority, need, and timeline with sufficient specificity for forecasting. The 40–60 range is largely a function of how the team defines "opportunity" — under strict definition (signed mutual action plan, identified economic buyer, defined evaluation criteria) the rate sits at 40%; under loose definition (verbal next step) the rate clears 60%.

Stage 5: opportunity-to-closed-won (15–30%)

The dominant variable is ICP fit. An opportunity sourced from a prospect outside the ICP closes at the floor; an opportunity inside a sharp, hypothesis-validated ICP closes at the ceiling. The secondary variable is ACV-appropriate cycle length — running an enterprise deal on a mid-market clock, or vice versa, depresses close rates by 30 to 50% from segment baseline. Weak ICP discipline compounds: it produces low reply rates and low close rates on the replies that do convert.

Cumulative funnel — a worked example

A team sending 10,000 sequenced messages per month against a sharp ICP, with disciplined infrastructure and disciplined reply handling, produces the following:

A 10,000-send month at typical rates produces between 1.5 and 3 closed-won deals — and the variance is not random. The same upstream send volume, run at the funnel-stage floors throughout, produces under 1 deal per month. Run at the ceilings, the same volume produces 6 deals. The 6x spread between floor and ceiling at the cumulative output is the central operational fact of outbound.

Per-channel funnel comparison

The same five-stage funnel applies at each outbound channel, with different per-stage rates and a similar cumulative outcome at well-run motions.

The downstream stages converge — once a reply exists, the channel that produced it ceases to matter for conversion. The upstream stages diverge sharply. A 10,000-touch month on phone, at typical rates, produces 600 conversations of which 180 are replies; the same month on email produces 5,000 opens of which 125 are replies. Per-touch economics differ; per-reply economics do not.

The multi-channel funnel — 1.4–1.8x, not 2x

A common operator error is to model multi-channel as additive: email funnel plus LinkedIn funnel equals 2x pipeline. The empirical observation is that overlap and prospect saturation produce a compounding closer to 1.4–1.8x rather than 2x. Two reasons. First, the same prospect that responds to a LinkedIn touch would, with non-trivial probability, also have responded to the email touch — the second channel cannibalizes engagement the first would have captured. Second, prospects that respond only to multi-channel are systematically harder to convert downstream: they engaged later, with less intent, and convert at the funnel floors rather than the ceilings.

The implication: multi-channel is correct for reasons of coverage (reaching the 30% of an ICP that email alone misses) rather than for raw multiplier. A two-channel motion at 1.6x single-channel output is the planning anchor.

ACV vs volume — the tradeoff

The economic structure of an outbound motion is determined by the relationship between average contract value, per-stage conversion, and per-touch cost. High-ACV segments justify slower funnel velocity and higher per-touch investment; low-ACV segments require volume to make the unit economics work.

The send-volume requirement scales linearly with the inverse of ACV, and at SMB ACV the required volume crosses the threshold where individual personalization is mechanically impossible and the funnel must be optimized for templated scale. The opposite is true at enterprise: 36,000 sends is the output of a single SDR over a quarter, and the marginal hour is better spent on per-account research that lifts stage-2 reply rate from 2% to 4%, doubling effective output without touching upstream volume.

Per-segment unit economics

A worked example, at typical operator cost structure:

• Enterprise ($100K ACV). 36,000 sends produces 90 replies, 18 meetings, 9 opportunities, 2 closed-won deals over a 90-to-180-day cycle. At $200K of loaded cost across SDR salary, infrastructure, and tooling, the gross margin per deal supports the motion at the first closed-won. Payback inside one quarter.
• Mid-market ($20K ACV). 180,000 sends produces 450 replies, 90 meetings, 45 opportunities, 10 closed-won deals over a 30-to-90-day cycle. At $300K of loaded cost (1.5 SDRs, more infrastructure), the motion breaks even at deal six or seven of ten — payback inside two quarters.
• SMB ($5K ACV). 720,000 sends produces 1,800 replies, 360 meetings, 180 opportunities, 40 closed-won deals over a 7-to-30-day cycle. At $400K of loaded cost (3 SDRs, substantial infrastructure, marketing-automation overlay), the motion requires all 40 deals to clear payback — and is exposed to any deterioration in stage-3 conversion. The SMB motion does not survive a reply-handling discipline failure.

AE-load math

A productive AE handles approximately 25 to 35 first meetings per month, producing 12 to 18 opportunities per month, producing 3 to 5 closed-won deals per month at typical conversion. The implication for hiring is direct: an outbound program producing more than 35 qualified meetings per month exceeds single-AE capacity and either (a) requires a second AE or (b) requires meeting throttling at the SDR layer to match AE bandwidth. Operators routinely produce stage-3 success that exceeds stage-4 capacity, and the resulting bottleneck is observed as a sharp degradation in meeting-to-opportunity conversion — the AE under load cannot run discovery at the rigor required to qualify, and falls back on accepting prospect narratives.

A two-AE team supports approximately 60 to 70 qualified meetings per month, mapping to approximately 250 to 350 replies, mapping to a send program of 100,000 to 175,000 messages per month at mid-market funnel rates. Beyond that volume, AE hiring leads SDR hiring.

Time-to-close patterns and payback

The empirical cycle lengths from first meeting to closed-won, observed across segments:

• Enterprise: 90 to 180 days. Multiple stakeholders, formal procurement, security review, legal redlines.
• Mid-market: 30 to 90 days. Single economic buyer, lightweight procurement, occasional security review.
• SMB: 7 to 30 days. Single buyer, no procurement, credit-card or self-serve close.

Combined with the per-segment unit economics above, the payback period for the outbound investment runs roughly one quarter at enterprise, two quarters at mid-market, and three to four quarters at SMB. The SMB motion compounds the unit-economics fragility with the slowest payback at the program level — a fact rarely surfaced in initial planning because the per-deal cycle feels short.

The highest-leverage funnel-improvement decision

At most operators, the highest-leverage stage to invest in is stage 3 — reply-to-qualified-meeting. The reasons are structural. Stage 1 (open) is capped near 60% and the marginal lift from full deliverability optimization is bounded. Stage 2 (open-to-reply) responds to copy and ICP investment but the per-month gain from a copy refresh is 0.5–1 percentage point. Stage 4 and stage 5 are downstream of upstream output and respond slowly to investment because the AE feedback loop runs at the cycle length of the segment.

Stage 3, by contrast, sits between 3% at undisciplined operators and 25% at disciplined operators — an 8x range, almost entirely a function of operational pattern rather than hire quality. A team converting replies at 5% that invests in the four-hour SLA, the Slack routing, and the per-objection response library typically moves to 15% within a quarter. At a 125-reply month, that is 12.5 additional meetings, 6 additional opportunities, and 1.4 additional closed-won deals — at no additional upstream send volume, at no additional headcount, at marginal infrastructure cost.

Stage 3 is the leverage point because it is the stage at which the upstream investment in domains, deliverability, ICP, copy, and sequencing is either captured or wasted. The reply already exists. Whether it converts is operational.

The operational metric stack

The dashboard the operator needs to actually see this funnel is narrow. Per stage, per segment, per channel, per month, the following six measures:

• Volume entering the stage
• Volume exiting the stage
• Per-stage conversion rate, with trailing-three-month moving average
• Cumulative conversion from send to current stage
• Variance from segment baseline
• The dominant variable's current value (deliverability score for stage 1, reply-handling SLA hit rate for stage 3, etc.)

A funnel observed at this granularity makes the next investment decision mechanical: identify the stage with the largest variance from baseline, identify the dominant variable, and invest. Most operators do not run this dashboard. Most operators see send volume, reply rate, and closed-won — three numbers that, together, are insufficient to identify which stage is producing the gap from target.

Cross-reference to the rest of the stack

Three upstream chapters anchor the math on this page. ICP chapter 03 in the segmentation cluster — the hypothesis-testing discipline that produces the narrow lists required to land at the stage-2 ceiling and the stage-5 ceiling simultaneously. The email cluster's authentication and warmup chapters — the deliverability discipline that produces the stage-1 floor at 50% rather than 30%. And the prior six chapters of this reply-handling cluster — classification, routing, triage, objection handling, meeting booking, nurture — which collectively produce the stage-3 ceiling at 25% rather than 5%.

The bounce-rate ceiling discussed in the email cluster (3% sustained, with intra-month spikes triggering provider-level throttling) is the upstream gate on stage 1. A sender at 4% sustained bounce rate observes stage-1 conversion collapsing toward 20–30% as the provider reputation system applies penalties. The funnel does not recover at downstream stages.

Common operator failures

• No per-stage measurement. The team reports send volume and closed-won and infers everything in between. The funnel-improvement decision is consequently made on intuition rather than on observed variance, and the team invests at the wrong stage by structural overweight on what is most visible (the email tool dashboard) rather than on what is most leveraged.
• No per-segment unit economics. The team runs enterprise, mid-market, and SMB through the same dashboard at the same cadence, and the slower-cycle enterprise pipeline is interpreted as failing when it is in fact maturing on schedule. Resource allocation degrades toward the segment with the fastest visible feedback rather than the segment with the highest economic return.
• No funnel-improvement prioritization. The team invests at every stage equally, or invests at the stage that is easiest to instrument (typically stage 2, reply rate, because the sequencing tool surfaces it). Investment at stage 3 — where leverage is largest — is structurally underfunded because stage 3 lives in Slack and CRM rather than in the sequencing tool.
• Treating all stages as equal investment. The team applies the same per-month engineering and operational investment to every funnel stage, which produces gains exactly where they are not needed and fails to produce gains where they are. A funnel-aware roadmap concentrates investment at the lowest-conversion stage relative to baseline.
• Modeling multi-channel as 2x. Planning assumes additive channel output; actual output is 1.4–1.8x; pipeline targets miss by 20–30% in the first quarter of multi-channel activation.
• AE-load blind spot. The team scales SDR output without scaling AE capacity, stage-4 conversion collapses, and the team interprets the collapse as a sourcing-quality problem when it is a capacity problem.

Pre-deployment metric checklist

• Per-stage conversion measured at each of the five stages, broken down by segment and channel
• Trailing-three-month moving average per stage, with variance from baseline flagged
• Per-segment unit economics computed monthly — loaded cost, deals required to break even, current payback timeline
• AE-load tracked against per-AE meeting and opportunity capacity, with throttling triggers defined at the SDR layer
• Multi-channel uplift measured against single-channel baseline at 1.4–1.8x, with variance investigated
• Bounce-rate observed at sequence entry and at trailing-30-day windows, with the 3% sustained ceiling enforced upstream
• The stage with the largest variance from baseline identified each month, with the next investment decision attached to it

Where the pipeline conversion math fits

The seven chapters of this cluster, read end-to-end, describe the operational layer between a reply existing and revenue existing. The six chapters that precede this one describe the patterns; this chapter is the arithmetic that makes those patterns load-bearing. Without per-stage measurement, the discipline established in chapters one through six is invisible to the operator — and what is invisible is not maintained. With per-stage measurement, the entire upstream stack — the domains catalogued in the email cluster, the segments validated in the ICP cluster, the sequences orchestrated in the channel clusters, the replies handled in this cluster — produces a single legible quantity: the per-month closed-won output of the program, decomposed into the five rates that each upstream investment moves.

That decomposition is where the entire reference stack realizes its return as measurable revenue. The arithmetic is unforgiving; it is also where the leverage is. An operator who runs the dashboard described in this chapter is making the next investment decision on the basis of where the funnel actually leaks, and the funnel-improvement decisions on that basis compound faster than the funnel-improvement decisions made without it.