← Back to BlogRevenue Architecture

    The Pipeline Math That Keeps CFOs Up at Night

    That Works Team13 min read

    Most pipeline reports are fiction. Here's the math your CFO actually cares about, and how to build a model that doesn't collapse under scrutiny.

    Your CRO says pipeline is 4x target. Your CFO says the company is going to miss the quarter. They're both looking at the same dashboard. How?

    Because pipeline math, the way most B2B companies calculate, report, and forecast their revenue pipeline, is fundamentally broken. Not because the tools are bad, but because the assumptions underneath are wrong.

    The 3x Coverage Myth

    Somewhere along the way, "3x pipeline coverage" became gospel. Need $1M in bookings? Build $3M in pipeline. Simple.

    Except it's not. The 3x rule assumes:

    • Consistent win rates: They're not. They vary by segment, source, deal size, and rep.
    • Consistent deal sizes: They're not. One enterprise deal can swing the entire quarter.
    • Consistent cycle times: They're not. Q4 deals close faster. New segments close slower.
    • Pipeline quality is uniform: It absolutely is not.

    A pipeline full of early-stage, poorly qualified opportunities at 3x coverage is worth less than a pipeline of late-stage, champion-confirmed deals at 1.5x coverage. But the dashboard doesn't know the difference.

    The Numbers That Actually Matter

    1. Weighted Pipeline by Stage and Age

    Raw pipeline is vanity. Weighted pipeline adjusted for stage and age is the number that matters.

    Here's why age matters: An opportunity that's been in "Discovery" for 90 days is not the same as one that entered Discovery yesterday. The longer a deal sits in a stage without progressing, the lower its actual probability of closing, regardless of what your stage-based probability model says.

    Build a decay function:

    • On-pace deals (within expected stage duration): Use standard probability
    • Stalling deals (1.5x expected duration): Reduce probability by 30%
    • Zombie deals (2x+ expected duration): Reduce probability by 60% or remove

    2. Source-Adjusted Win Rates

    Your overall win rate is meaningless. What matters is win rate by source:

    Source | Avg Win Rate | Avg Deal Size | Avg Cycle

    Inbound demo request | 25-35% | Varies | 30-45 days

    Outbound SDR | 10-18% | Varies | 45-75 days

    Partner referral | 30-45% | Higher | 30-60 days

    Event/conference | 8-15% | Lower | 60-90 days

    PLG conversion | 15-25% | Lower | 20-40 days

    When your pipeline is 60% outbound-sourced, applying a blended 22% win rate massively overstates likely bookings. Segment the math.

    3. Commit vs. Best Case vs. Pipeline

    Your forecast should have three tiers:

    • Commit: Deals with verbal/written commitment, late-stage, champion confirmed. This should be 90%+ accurate.
    • Best case: Deals in late stages with positive signals but not yet committed. Weight at 40-60%.
    • Pipeline: Everything else that's qualified. Weight at 10-25%.

    If your Commit number doesn't cover 80%+ of target by mid-quarter, you have a problem. No amount of "best case" optimism fixes a weak Commit.

    4. Pipeline Creation Rate vs. Pipeline Required

    This is the number that keeps good CFOs up at night. Not "how much pipeline do we have?" but "how fast are we creating new pipeline, and is that rate sufficient?"

    The formula:

    • Pipeline required = (Target - Commit) / Win Rate
    • Pipeline creation rate = New qualified pipeline created per week/month
    • Weeks remaining = Time left in the quarter
    • Gap = Pipeline required - (Creation rate × Weeks remaining × Win rate)

    If the gap is positive, you're short. And unlike pipeline coverage, this number actually tells you how short and whether the trend can close the gap.

    5. Cost Per Pipeline Dollar

    Marketing loves to report cost per lead. Finance cares about cost per pipeline dollar, and ultimately cost per revenue dollar.

    • Cost per pipeline dollar = Total GTM spend / Pipeline generated
    • Healthy range: $0.10-0.30 per pipeline dollar for mid-market B2B
    • Cost per revenue dollar = Total GTM spend / Closed-won revenue
    • Healthy range: $0.30-0.80 depending on ACV and segment

    If you're spending $1 to generate $1 of pipeline, your unit economics are broken regardless of growth rate.

    Building a Model That Survives the Board Meeting

    Step 1: Segment Everything

    Stop reporting blended numbers. Break pipeline by:

    • Source (inbound, outbound, partner, PLG)
    • Segment (SMB, mid-market, enterprise)
    • Product line (if applicable)
    • New business vs. expansion

    Each combination has different win rates, cycle times, and deal sizes. Your model needs to reflect that.

    Step 2: Build Rolling Forecasts

    Quarterly targets are important. Rolling 3-month forecasts are more useful. They smooth out the hockey-stick pattern and give you earlier warning signals.

    Every Monday, update:

    • What closed last week
    • What's in Commit for the quarter
    • What moved stages
    • What was created
    • What was lost or pushed

    Step 3: Track Leading Indicators

    Lagging indicators tell you what happened. Leading indicators tell you what's about to happen:

    • Meeting volume this week: Leading indicator for pipeline creation next month
    • Stage progression rate: Leading indicator for close rate this quarter
    • New logo vs. expansion mix: Leading indicator for future growth sustainability
    • Rep activity patterns: Leading indicator for individual performance

    Step 4: Create a Pipeline Council

    The best-run revenue organisations hold a weekly pipeline council, not a forecast call, a pipeline council.

    Participants: CRO, CMO, VP Sales, VP Marketing, RevOps lead.

    Agenda:

    • Pipeline health by segment (5 min)
    • Creation rate vs. required (5 min)
    • At-risk deals review (10 min)
    • Coverage gap plan (10 min)

    30 minutes. Every week. No slides longer than 3 pages.

    The Conversation With Your CFO

    CFOs don't care about pipeline. They care about predictability. When they push back on pipeline numbers, they're really asking:

    1. "Can I trust these numbers?": Show them your methodology: weighted, source-adjusted, age-decayed.
    2. "Will we hit the quarter?": Show them the Commit number, not the total pipeline.
    3. "What's the risk?": Show them the creation rate gap and the deals at risk.
    4. "What do we need to change?": Show them the specific segment or source that's underperforming and the plan to fix it.

    If you can answer those four questions with data every week, your CFO becomes your biggest ally instead of your toughest critic.

    The Bottom Line

    Pipeline math isn't hard. But it requires honesty, about win rates, deal quality, and creation velocity.

    Stop hiding behind 3x coverage ratios. Start measuring what actually predicts revenue: weighted pipeline, source-adjusted win rates, creation velocity, and the gap between where you are and where you need to be.

    The companies that forecast accurately aren't the ones with the best tools. They're the ones willing to look at the uncomfortable numbers.