Modeling
The four proprietary models that power CTC's forecasting and planning system. Each model answers a specific question about the future of your business, and together they produce the daily P&L forecast that drives every operational decision.
A Connected Forecasting System
Before diving into each model, it is important to understand how they work together. The four models form a connected system where each output feeds the next.
The Spending Power Model answers the first question: given a level of ad spend, how efficiently will that spend convert into new customer revenue? It produces a spend-to-efficiency curve that tells the Profit Engineer exactly how many new customer dollars to expect at any budget level.
Those new customers then flow into the Retention Model. For every cohort of new customers acquired in a given month, the Retention Model predicts how much revenue they will bring back in each subsequent month. The output is a complete picture of returning customer revenue built from the bottom up, cohort by cohort.
The Event Effect Model adds daily precision. On any given day, a marketing event (a sale, a product drop, a VIP campaign) shifts the natural pattern of revenue. The Event Effect Model learns these shifts from historical data and applies them to future event days on the marketing calendar. Critically, it does this without breaking monthly budget targets. Event days run hotter. Surrounding days compress proportionally. The month always reconciles.
Finally, the Creative Demand Model ensures the ad account has enough fuel to execute the plan. Based on the brand's current ad portfolio health and the planned spend level, it calculates exactly how many new ads need to be produced each month. This closes the loop: the business plan determines the spend, the spend determines the creative demand, and the creative enables the spend.
Spending Power
Forecasts new customer revenue given any level of ad spend.
Retention
Forecasts returning customer revenue from every past cohort.
Event Effect
Distributes revenue accurately across days with marketing events.
Creative Demand
Ensures the ad account has enough creative fuel to sustain the plan.
Together: Spending Power forecasts new revenue. Retention forecasts returning revenue. Event Effect distributes revenue across days. Creative Demand ensures the ad account can sustain the plan. The sum is a daily P&L forecast that every Profit Engineer operates against.
Forecasting New Customer Acquisition Efficiency
The Spending Power Model forecasts new customer acquisition efficiency. It answers: given a level of ad spend in a future month, how efficiently will that spend convert into new customer revenue?
The Core Metric: ACONS
The model is built on ACONS (Ad Cost of New Sales), defined as total ad spend divided by new customer revenue. ACONS is the inverse of aMER (acquisition Marketing Efficiency Rating). It represents how many cents of ad spend it takes to generate one dollar of new customer revenue. Lower ACONS means higher efficiency.
The model produces a linear equation for each month (a slope and an intercept) that describes how ACONS changes as spend increases. The slope captures diminishing returns: the more you spend, the less efficient each incremental dollar becomes. The intercept represents baseline efficiency at near-zero spend.
Spending Power
A related output is Spending Power, defined as the inverse of the slope (1/slope). Spending Power tells you how many additional dollars of monthly spend it takes to degrade ACONS by one point.
High Spending Power
The brand can absorb a million dollars of incremental spend before efficiency drops meaningfully. Strong creative, broad TAM, efficient machine.
Low Spending Power
Efficiency falls off quickly with scale. Narrow audience, creative fatigue, or market saturation. Scale requires caution and creative investment.
How the Profit Engineer Uses It
In the Statlas Planning section, the PE inputs a planned spend level for a future month. The model applies that month's forecasted slope and intercept to compute expected ACONS, from which aMER, new customer revenue, new customer count, and CAC are all derived. This is the first building block of the monthly revenue forecast.
The model also establishes valid spend ranges based on historical data. Predicting efficiency at spend levels well outside the historical range should be treated with caution.
Scenario Planning: Three Points on the Curve
The spend-to-efficiency curve is not just a forecast. It is a decision tool. Every point on the curve represents a different financial outcome, and the Profit Engineer's job is to help the business choose which outcome to optimize for.
By layering in the customer's LTV and gross margin over time, the model allows the PE to identify three critical optimization points:
Maximize Month One Contribution
The most conservative point. Spend only to the level where every new customer acquired generates positive contribution margin in their first month. No reliance on future purchases. Highest efficiency, lowest volume.
New Customer Revenue at Breakeven
The middle ground. Spend to the point where new customer acquisition breaks even on first purchase, with all future returning revenue as pure upside. Higher volume than M1, with the bet that retention will deliver the profit.
Maximize Lifetime Contribution
The most aggressive point. Spend to the level where the total lifetime gross margin of the acquired customers (factoring in all future cohort revenue from the Retention Model) exceeds the acquisition cost. Maximum volume, longest payback period.
Each of these points represents a fundamentally different business strategy. A cash-constrained brand may need to optimize for Month One Contribution. A brand with strong retention and available capital may choose to push toward Lifetime Contribution, accepting short-term losses for long-term customer file growth.
The model makes these trade-offs explicit and computable. The PE does not guess which point to target. They model each scenario in Statlas, show the business the financial implications of each choice, and align the media plan to the outcome the business selects. This is how CTC turns a spend-to-efficiency curve into a strategic conversation.
What Gets Delivered
Each model delivery includes:
The Spending Power trend over time — is the brand becoming more or less efficient at scale?
Predicted vs. actual accuracy metrics across all forecasted months
Platform-by-platform allocation recommendations with suggested spend share ranges
Incrementality estimates per platform, enabling apples-to-apples channel comparison
A comparison of the brand's trend against CTC's cross-brand DTC Consumer Index
Multiple model versions are uploaded to Statlas (primary ensemble, last-year baseline, and alternative ensemble) enabling the PE to evaluate different scenarios during planning.
Cadence: The model is rebuilt when meaningful changes in spend level, efficiency trends, or business conditions warrant it. New data continuously refines the forecast.
Forecasting Returning Customer Revenue
The Retention Model forecasts returning customer revenue. It answers: given the new customers we acquired each month, how much revenue will they bring back in future months?
The Core Metric: LTV Lift
LTV Lift is the percentage of a cohort's first-time order revenue that comes back as returning revenue in each subsequent month. If a cohort spent $100,000 in their first month and the Month 3 LTV Lift is 2%, we expect $2,000 in returning revenue from that cohort three months later.
LTV Lift follows an exponential decay pattern: highest in the first month after acquisition, dropping sharply over the first several months, and eventually flattening to a low, stable terminal rate that persists indefinitely.
How the Forecast Builds
For any future month, returning revenue is the sum of all active cohort contributions. Each past cohort contributes its new revenue multiplied by the predicted LTV Lift for that cohort's age.
January 2026 returning revenue equals January runout plus December 2025 new revenue × the M1 Lift, plus November 2025 new revenue × the M2 Lift, plus October 2025 new revenue × the M3 Lift, and so on back through every active cohort.
The new revenue figures for each cohort come from whatever plan is currently active in Statlas, which means the Retention Model is directly connected to the Spending Power Model's output.
How the Profit Engineer Uses It
The PE sees a cohort retention chart showing predicted LTV Lifts over time, an actual vs. modeled lift table comparing real data against predictions, and the regression results. This gives visibility into whether the brand's customer retention is improving, declining, or stable.
A growing active customer file predicts revenue growth. A shrinking one signals trouble, no matter how large the total email list looks. Many of those customers are lapsed. They are not coming back.
When Something Looks Wrong
R-squared Below 0.6
The model is flagged as underperforming. The data team investigates whether the curve shape or the level needs adjustment.
Curve Shape Right, Level Off
The data team adjusts the prediction scale. The pattern of decay is correct but the absolute values need recalibration.
Curve Shape Wrong
Retention behavior has fundamentally changed. The model's decay and terminal rate parameters are updated to reflect the new reality.
Cadence: Quarterly review, with off-cycle retunes triggered by major promotional events, significant changes in acquisition volume, or sustained gaps between modeled and actual returning revenue.
Daily Precision Through Marketing Events
The Event Effect Model makes daily forecasts more accurate by learning how marketing events shift revenue patterns. It answers: on a day with a planned marketing event, how much should we expect new and returning revenue to deviate from the baseline?
Two Core Signals
For every historical marketing event, the model measures two things:
Return Ratio
How much the event skewed revenue toward returning customers relative to a normal day. A ratio above baseline indicates the event activated the existing customer base more than usual — common in loyalty campaigns, VIP drops, and seasonal events.
AMER Lift
How much more (or less) efficiently paid acquisition performed during the event compared to baseline. A lift above 1.0 means paid acquisition was more efficient — common in high-intent sale periods where conversion rates spike.
The Budget Constraint Guarantee
This is critical: event effects are applied as relative lifts, not absolute overrides. If a month is planned at $30,000 spend at 3.0 aMER, the aggregate daily outcomes across that month will always reconcile back to those targets. The model redistributes when revenue lands within the month, not the total amount. Event days run hotter. Surrounding days compress proportionally. The month-level constraint is never violated.
How the Profit Engineer Uses It
When building the marketing calendar in Statlas, each event the PE adds carries an expected performance multiplier derived from the model. The PE sees what the model predicts for each event type based on historical data, and the daily forecast adjusts automatically.
Seasonal events get special treatment. Rather than using the type-level average, the model looks for last year's equivalent event and uses that performance directly, preserving year-over-year patterns.
Requirements
The model requires at least 10 historical tagged events. Below 20 to 30 events, type-level summaries may rest on small samples and uncertainty will be wide. Forecast quality improves with more history.
This is why tagging events consistently in the marketing calendar from day one is essential. Every untagged event is a missed data point that degrades future forecast precision.
Ensuring the Ad Account Has Enough Fuel
The Creative Demand Model ensures the ad account has enough creative fuel to execute the media plan. It answers: how many new ads does this brand need to produce this month?
The Creative Score (0 to 100)
The model evaluates the current health of the brand's ad portfolio using five metrics, each benchmarked against all managed stores:
Zero Revenue Rate — The percentage of active ads that never converted. High values mean many ads are not working.
Ad Concentration — The percentage of total spend in the top 5 ads. High values mean over-reliance on a few ads, creating fatigue risk.
ROAS Degradation — The change in efficiency after initial launch week. Negative values mean ads are losing effectiveness quickly.
Spend Degradation — The change in spend after initial launch week. Negative values signal rapid fatigue.
Evergreen Share — The percentage of ads running consistently for 30+ days. Higher evergreen share means more stable performance and lower creative demand.
Each metric is converted to a percentile against all managed stores, then weighted and combined into a single Creative Score. A score above 50 means the brand can create fewer ads than last year. A score below 50 means they need to create more to compensate for lower portfolio health.
How the Recommendation Works
The recommended number of new ads is calculated from four inputs:
How many ads the brand created in the same month last year
How much planned spend has changed year-over-year
The Creative Score (as a health multiplier)
The evergreen share (how much existing creative carries forward)
A less healthy account gets a higher multiplier, meaning more new ads are needed. A healthier account gets a lower multiplier. The model also includes sanity checks: if last year's ad volume was clearly insufficient for the spend level, the model adjusts upward rather than anchoring to a bad baseline.
Smart Insights
The metrics and Creative Score are passed to an AI explanation layer that generates a plain-English summary called Smart Insights. This gives the growth strategist something they can walk clients through without interpreting raw numbers. For example:
“High Evergreen Stability Shows Promise Despite Concentration Risks.”Smart Insights, AI-generated summary
How the Profit Engineer Uses It
The PE sees a specific target: “You need X new ads this month.” This number is not a guess. It is derived from the intersection of the media plan (how much spend needs creative support), the current portfolio health (how fast existing ads are fatiguing), and the historical baseline (what worked at this spend level before).
The Creative Demand Model connects directly to the Four-Phase Expansion Framework in the Meta methodology. When the system is underspending, Phase 1 (Creative Expansion) is always the first lever. The Creative Demand Model tells you exactly how much creative is needed and why.
Tools, Not Oracles
All four models are maintained by the CTC data team and tuned per brand. Each model has configuration controls that allow the data team to adjust for brand-specific behavior without rebuilding the model from scratch.
The key principle: models are tools, not oracles. They are built to be useful, not to be perfectly right. The value of the model is not in its point estimate but in its ability to surface deviations quickly so the PE can course-correct before the damage is done.
When a model's predictions deviate meaningfully from reality, the data team investigates. The question is not “is the model broken?” but “what changed in the business that the model has not yet captured?” Sometimes the answer is a retune. Sometimes the answer is that the business is in a genuinely new operating environment and the model needs new data before it can be trusted again.
Rebuild Cadence
| Model | Standard Cadence | Off-Cycle Triggers |
|---|---|---|
| Spending Power | As-needed | Meaningful changes in spend level, efficiency trends, or business conditions |
| Retention | Quarterly | Major promotions, acquisition volume shifts, sustained model-actual gaps |
| Event Effect | As-needed | New event types, calendar changes, systematic prediction errors |
| Creative Demand | Monthly | Significant spend changes, portfolio health shifts, creative strategy pivots |
Models are rebuilt on their regular cadence and outside of schedule when triggered by business changes. The operating principle is simple: when the map no longer matches the territory, update the map.
The Engine Behind the Forecast
Every number in CTC’s forecasting system runs through these four models. The result is a daily P&L forecast with channel-level precision that every Profit Engineer operates against, every single day.
