How We Forecast $4B in GMV to 3% Accuracy

Every forecast is wrong. The question is whether yours is useful.

Luke Austin, walks through the CTC Methodology series opener: a complete operating model for making profit predictable across ecommerce brands. This is not a spreadsheet. It is a four-step system built on 12 years of experience and $4 billion in GMV, combining proprietary data science with daily operational discipline to hit 3% forecast accuracy at scale.

Topics covered in this episode:

• Why all models are wrong and what makes the best ones useful

• The "tigers not mice" framework for prioritizing what actually matters

• Qualitative planning: how a 12-month marketing calendar becomes a mathematical input

• The Spending Power (AMER) model and three optimization modes for new customer spend

• Cohort LTV modeling: why active vs. lapsed customer distinction changes everything

• The Event Effect model: how marketing moments get quantified, not just scheduled

• Building a full P&L forecast from customer cohorts up, not channel metrics down

• Why contribution margin is the north star metric, not ROAS

• Plot, Pivot, Profit: the daily cadence that makes forecasts self-correcting

• The "What / So What / Now What" daily operating format used by CTC profit engineers

• Results: 3% forecast accuracy across $4B GMV, 32% avg revenue growth, 41% avg CM growth

This is Episode 1 of the CTC Canon Series. The Canon represents CTC's cumulative operating principles across 12-plus years and hundreds of brands, covering forecasting, media buying, creative strategy, email, and media measurement.

Show Notes:

• Go to https://bit.ly/4aiEz79 to start your free migration with Omnisend today

• Explore the Prophit Engine: https://commonthreadco.com/pages/prophit-engine

• The Ecommerce Playbook mailbag is open — email us at podcast@commonthreadco.com to ask us any questions you might have