Inside View vs. Outside View: Auditing Overconfident Campaign Forecasts
Objective: Audit three multi-million-dollar marketing launch forecasts, identifying inside-view cognitive biases, unanchored conversion assumptions, and improper reference class selection.
You are the VP of Marketing Strategy at Freshworks reviewing three quarterly campaign budget proposals submitted by different product marketing teams: an outbound enterprise ABM campaign, an organic inbound content hub, and a self-serve freemium conversion redesign. Each proposal promises aggressive revenue ROI, but several are classic examples of the planning fallacy.
Audit the three forecast proposals. For each specimen, identify which claims violate reference-class forecasting principles (e.g. Reference Class of One, adjusting for subjective hopes, assuming top-decile performance across all funnel stages) vs. which claims are properly anchored in historical base rates.
Before you start
What you'll need
Free path (everything below is enough to finish)
Provides structured document layout for strategic teardowns and evaluations without paid tools.
No access? Google Docs or any markdown/text editor
The process
Specimens to review
Which specific forecasting assumptions in Campaigns A, B, and C suffer from the planning fallacy or inside-view distortions? Identify the defects, explain why they violate reference-class principles, and identify which assumptions are legitimate base-rate anchors.
Q3 MARKETING BUDGET PROPOSALS — Executive Summary PROPOSAL A: Outbound Enterprise ABM (Target: Global 2000 IT) - Proposed Budget: $250,000 - Forecast: 60 Enterprise Deals Closed ($1.8M ARR) - Reasoning: 'Last quarter we ran a pilot targeting 10 accounts and closed 2 ($60k ARR each). If we scale outreach to 500 accounts, we will easily close 60 deals at a 12% win rate because our product-market fit in IT service management is proven.' - Cold Outbound Response Rate Anchor: Model assumes a 1.8% cold meeting booking rate, which matches our historical median across 45 previous outbound campaigns. PROPOSAL B: Product-Led Organic SEO Content Hub - Proposed Budget: $180,000 (100 in-depth comparison articles) - Forecast: 50,000 Monthly Organic Visits within 60 days of publishing - Reasoning: 'The target keywords have a combined search volume of 180,000/mo. At a top-3 ranking CTR of 28%, we will capture 50,000 visits by month 2, generating 1,200 product trials.' PROPOSAL C: Self-Serve Freemium Onboarding Overhaul - Proposed Budget: $90,000 (UX redesign + interactive product tours) - Forecast: +80% increase in free-to-paid conversion (from 1.5% to 2.7%) - Reasoning: 'The new onboarding UI is significantly more modern, intuitive, and visually stunning. Team enthusiasm is unanimous, justifying an 80% conversion uplift.' - Testing Plan: 14-day 50/50 holdout A/B test with pre-registered sample size to measure actual vs forecast conversion lift.
Specimen: synthetic, realistic
Final deliverable
A written audit report classifying the forecast defects across all three campaign proposals, citing the specific base-rate principle violated and providing a calibrated baseline for each.
See a reference example
Worked example for an email platform (Klaviyo campaign audit): An ecommerce brand's email team submitted a Black Friday forecast predicting a 45% open rate and 8% click-to-purchase conversion based on their top performing flash sale from July. Audit finding: Defect flagged under Mistake 1 (Reference Class of One) and Inside-View overconfidence. The historical base rate across 12 Q4 peak periods for this industry sector shows median open rates drop to 18.2% (due to inbox congestion) and conversion averages 2.4%. The passing audit replaces the 8% assumption with the 2.4% baseline, adjusting upward only by +0.5% for proven SMS segmentation, resulting in a realistic revenue forecast of $320k rather than an unachievable $1.1M.
Success criteria
You're done when you can:
- Correctly identifies all 3 forecasting defects (Reference Class of One, Funnel Stacking Overconfidence, Hopes-Based Adjustment)
- Leaves the 2 sound base-rate and holdout calibration elements unflagged as distractors
- Explains the exact statistical or behavioral flaw in each defective forecast assumption
- Provides a calibrated outside-view correction for each flagged proposal