The Forecasting System Behind $4B in GMV: How CTC Makes Profit Predictable

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

That is the founding principle behind the CTC Operating Model for Profit, the forecasting methodology Luke Austin, SVP of eCommerce Strategy at Common Thread Collective, breaks down in Episode 1 of the CTC Canon series. Built over 12 years and pressure-tested across hundreds of ecommerce brands, this is the system that produces 3% forecast accuracy across $4 billion in GMV, 32% average revenue growth, and 41% average contribution margin growth across CTC's client portfolio.

It is not a spreadsheet. It is a four-step operating model that starts with a different premise entirely.

Why Most Ecommerce Forecasts Fail Before They Start

The most common forecasting mistake is building from the wrong unit. Most brands start with channels: projected traffic, conversion rates, and year-over-year growth percentages stacked into a spreadsheet. The problem is that channels do not generate revenue. Customers do.

The CTC model starts with cohorts as the atomic unit of the business. Every customer who has ever bought from your brand belongs to a cohort. Every future customer will belong to a new one. When you model the retention behavior and lifetime value of those cohorts, you get a fundamentally more accurate picture of what your business is actually going to do.

The second failure mode is what Luke calls "optimizing for mice while tigers are abroad," a reference to statistician George E.P. Box. Bad contribution margin is a tiger. Bloated operating expenses are a tiger. Your next headline variation is a mouse. Great forecasting starts by building a system that puts the tigers in front of you every single morning, before anything else.

"The point of a forecast is not to predict the future perfectly. The point is to understand where you are wrong so you can course-correct before any damage is done."

The 4-Step Operating Model for Profit

The CTC forecasting system runs in four sequential steps. Each one feeds the next.

Step 1: Qualitative Planning

Before a single number enters a model, CTC builds a 12-month marketing calendar. Every promotion, product launch, sale event, and cultural moment gets plotted. This is not just a content schedule. It is the revenue architecture for the year.

Revenue is never linear. It comes in moments of elevated purchase demand, and the best brands design those moments in advance. When conversion rate spikes because of a well-timed promotion, it arbitrages the ad auction. Platforms price efficiency on expected performance. When you outperform that expectation through a planned marketing moment, the economics shift in your favor. That dynamic only happens if the moment was designed first.

Step 2: Quantitative Modeling

Three proprietary models convert the marketing calendar into a mathematical forecast.

The Spending Power (AMER) model analyzes the historical relationship between media spend and new customer acquisition efficiency. It produces a spend-to-efficiency degradation curve that predicts new customer revenue at any given spend level across three optimization modes: volume, efficiency, and blended.

The Cohort LTV model tracks every customer cohort's retention behavior over time, projecting returning customer revenue month by month. The key insight here is the active versus lapsed distinction. A brand with a large total customer file but a shrinking active cohort is in a fundamentally different position than one with a smaller but growing active base. The model surfaces that difference immediately.

The Event Effect model quantifies how each item on the marketing calendar impacts new and returning customer revenue on a given day. A promotion does not just increase overall revenue; it shifts the shape of the demand curve in ways that can be modeled and anticipated. This is what transforms a calendar from a planning tool into a mathematical input.

Step 3: Build the Map

With the three models producing outputs, CTC builds a full P&L forecast from the customer cohort level up, not from channel metrics down. Every dollar of planned spend, every campaign, every channel ladders up to the financial goal. Contribution margin is the north star metric throughout. Not ROAS. Not revenue. Contribution margin, because it is the only number that tells you whether the business is actually working.

"Tigers not mice. Bad gross margin is a tiger. Bloated OpEx is a tiger. Contribution margin pacing to forecast is what you look at first, every single day."

Step 4: Plot, Pivot, Profit

The most important step is the daily cadence. CTC profit engineers track 35 metrics inside Statlas every morning using a structured daily format called What / So What / Now What.

What: Yesterday's actuals versus the daily forecast expectation across revenue, contribution margin, new customers, and spend. So What: Is this a volume problem (not enough orders or revenue) or an efficiency problem (too much spend relative to revenue generated)? Which specific input is off? Now What: The specific action being taken today to course-correct before the month closes.

This daily cadence is what makes the forecast self-correcting. By day two of a month, if you can identify which model input is too high or too low, you still have time to act. By end of month, you are reporting on damage that could have been avoided.

Results Across the Portfolio

The outcomes CTC tracks across brands running this system: 3% average forecast accuracy across $4B in GMV, 32% average revenue growth, and 41% average contribution margin growth. Those numbers are the result of cohort-based modeling, daily operational discipline, and a relentless focus on contribution margin as the single most important measure of business health.

This is Episode 1 of the CTC Canon series, a complete codification of CTC's operating principles across forecasting, media buying, creative strategy, email, and measurement. The Canon represents 12-plus years of compounded learning across hundreds of ecommerce brands. If you want to go deeper on the forecasting methodology or explore how the Prophit Engine puts this system into practice for your brand, start there.

Frequently Asked Questions About Ecommerce Forecasting

Build a Forecast That Actually Drives Decisions

The CTC forecasting system turns a 12-month marketing calendar, three quantitative models, and a daily operating cadence into a compound advantage. If your current forecasting process is not telling you which problem you are facing before the month is over, we can help you build one that does.

Talk to Us