The Q4-to-annual-plan
framework.

Most DTC annual plans are built the wrong way around. A founder picks a growth number that feels ambitious, works backward to a revenue target, and hands the team a top-down goal with no lever attached. Then Q1 deviates from the plan, and there is nothing to pull, because the plan was a wish dressed as a forecast.

There is a better input sitting right in front of you: your Q4. November and December are roughly 19 percent of annual US retail, per the NRF, and the 2025 season crossed $1 trillion for the first time. That concentration is exactly why Q4 is the sharpest input for next year's plan. Peak volume stress-tests your CAC, your margin, your fulfillment, and your repeat behavior all at once, so the actuals reveal the real unit economics a quiet quarter never forces you to confront.

I built these models at WIN Brands, where the annual plan had to survive contact with a real Q4 and a real board. The version that worked was always driver-based: revenue built up from customers and repeat rate, planned on contribution margin, with acquisition sized to what the margin could actually absorb. The version that failed was always the top-down target with no driver logic underneath it.

This is the framework, step by step. Why Q4 is the input, how to deseasonalize it, how to build revenue from drivers, why you plan on contribution margin, how to set max-allowable CAC and the retention mix, how to plan the cash timeline and budget envelope, and how to run it as a rolling scenario forecast. Every benchmark is from a 2025 or 2026 source.

Why Q4 is the
best input for
next year.

Q4 is the input because it is where the year concentrates and where your economics get stress-tested. November and December are roughly 19 percent of annual US retail (NRF), and peak volume pushes CAC, contribution margin, fulfillment cost, and repeat behavior to their limits, which is exactly the data a plan needs. A brand's real unit economics show up under load, not in a slow quarter.

Two more reasons Q4 beats any other quarter as an input. Returns reality surfaces: with about 19.3 percent of online sales returned (NRF and Happy Returns), peak-season volume produces the return wave that reveals your true post-return contribution margin. And cohort quality reads fast, because 76 percent of repeat purchases happen within 90 days (BS&Co), so the Q4 cohort's early repeat signal is visible by late Q1, before you commit next year's acquisition budget.

The takeaway is to treat Q4 as a diagnostic, not just a revenue event. The BFCM revenue is nice, but the durable value is the data: what your real CAC was when the auction was hot, what your margin held at after discounts and returns, and how the peak cohort behaves. That is the raw material for a plan that holds, and it pairs directly with running the BFCM peak-season playbook well in the first place.

Deseasonalize, then
build revenue
from drivers.

Step one is to deseasonalize the Q4 actuals before you extrapolate anything. Q4 is a roughly 19 percent concentration of the year, so annualizing a December run rate overstates everything. Instead, pull your Q4 new-customer count, average order value, repeat rate, contribution margin, and CAC, and index each against a normal quarter. You are extracting the unit economics Q4 exposed, not projecting peak volume across twelve months.

Step two is to build next year's revenue bottoms-up from drivers, never top-down from a target. The model is straightforward: new customers per channel times new-customer AOV, plus returning customers times repeat rate times returning AOV. Anchor the repeat input to your category band rather than a blended average, since consumables run 22 to 44 percent and apparel closer to 10 to 17 percent (BS&Co). A driver model turns every growth assumption into a lever you can defend or cut when reality moves.

The reason this ordering matters is that a driver-based revenue line is auditable and a top-down one is not. When Q1 comes in soft, a driver model tells you exactly which assumption broke, new-customer CAC, repeat rate, or AOV, and what to adjust. A top-down number just tells you that you missed, which is the same feedback the whole team already has, and none of the diagnosis.

Plan on contribution
margin, not
gross margin.

The most common planning error is stopping at gross margin, and it is the one that hides the truth. The median public DTC gross margin is 56.6 percent (Eightx, SEC filings), but that is not the number that funds acquisition. Layer down to third-tier contribution margin, after shipping, fulfillment, and variable marketing, and it lands near 5 to 35 percent. That thin band, not the headline 56.6 percent, is the real fuel for growth.

Two adjustments make the margin honest. Fold a returns reserve into the plan, since 19.3 percent of online sales come back (NRF and Happy Returns), and that hits contribution margin directly. Then plan the P&L in those contribution-margin layers, not a single gross number, so the dollars you actually have for acquisition are visible. The DTC profitability tool models these layers if you want to see yours.

Set max CAC from
margin, and pick
your mix.

Step four is to set maximum allowable CAC from contribution margin and a payback window, not from a spend percentage. Your max CAC is the contribution-margin dollars a first order, plus expected repeat value, can absorb inside your target payback, typically the first order for pure new-customer economics or 90 days once repeat is included, since 76 percent of repeats land within 90 days. Separate blended CAC from true new-customer CAC, because retention revenue makes blended CAC look efficient and hides an overspend on acquisition.

Step five is to set the retention-versus-acquisition mix explicitly, as a decision rather than a residual. Established DTC brands run near 60 percent of revenue from returning customers (Eightx), so a brand far below that at scale is over-indexed on paid acquisition and exposed to CAC inflation. Fund lifecycle and retention as a line item with its own target repeat rate by category, not as whatever budget is left after acquisition.

Getting this mix right is what makes the plan durable in a market where acquisition keeps getting more expensive. A plan that leans on ever-cheaper new-customer CAC is planning on a trend that reversed years ago. A plan that funds retention deliberately compounds, which is the same logic behind the max allowable CAC calculator and the difference between brands that scale and brands that stall.

Plan the cash, then
set the budget
envelope.

Step six is to plan the cash timeline, not just the revenue line. DTC inventory runs a median of about 129 inventory days, and public DTC cash conversion cycles often run 75 to 120-plus days (Eightx), so a peak-season buy locks up working capital for a quarter or more. Build a month-by-month cash and inventory plan, the pre-peak buy, the in-season sell-through, and the post-peak liquidation, so a strong revenue plan does not create a January cash crisis the balance sheet cannot fund.

Step seven is to set the budget envelope against benchmarks, then allocate by driver. Marketing runs about 7.7 percent of revenue at large firms (Gartner 2025) and around 17.2 percent at growth-stage public DTC brands (Eightx 2026), and most scaling DTC brands plan toward the higher band. Set the envelope, then allocate it across the acquisition and retention drivers from the earlier steps, with paid media sized to your max-allowable CAC, not to a fixed percentage of revenue.

The discipline here is to let the drivers set the budget, not the other way around. A budget handed down as a percentage of a revenue target just re-imports the top-down problem you were trying to escape. A budget built up from the CAC your margin supports and the retention you are funding is a plan you can actually run, and adjust, as the year unfolds.

Run scenarios, and
avoid the usual
planning traps.

Step eight is to run the driver model as base, upside, and downside scenarios on a rolling forecast. Flex the two or three assumptions that move the plan most, usually new-customer CAC, repeat rate, and gross margin, and re-forecast monthly or quarterly against actuals. A rolling, driver-based forecast beats a static annual number the moment Q1 deviates, because it tells you which lever moved and lets you respond instead of just noting the miss.

The common traps are all versions of skipping a step. Planning off the peak month annualizes a December run rate that is a roughly 19 percent concentration of the year. A top-down revenue target has no driver logic to pull when reality shifts. Planning on gross margin instead of contribution margin overstates the cash for acquisition. Confusing blended CAC with new-customer CAC overspends into a real cost the model never isolated. Forgetting the cash cycle collides with 129-day inventory. And treating retention as a leftover leaves the brand exposed to rising CAC.

Avoid those six and the plan mostly builds itself, because each step feeds the next: deseasonalized actuals feed the driver model, the driver model feeds contribution margin, contribution margin sets max CAC, and the whole thing runs as a rolling forecast you can defend. That is the difference between a plan that survives the year and a target that gets quietly abandoned by February, and it is where your next growth inflection actually comes from.