Every marketing budget is a bet.
Most people just don't admit it. They build a number, call it a plan, and treat it like it's safe. It isn't. The moment you commit real money against an outcome you can't fully control, you're exposed — to media costs that move, conversion rates that drift and a market that doesn't care what's in your spreadsheet.
I found out exactly how exposed while running a monthly budget north of $3M.
There was a stretch where I adjusted our targets five times in one month. Not because the market was unusually volatile. Because the targets had no structure to begin with — just a number, divided by twelve. At that size, every adjustment isn't a tweak. It's real money moving on no real logic.
That's what an unhedged bet actually looks like from the inside.
After that, I stopped building budgets as projections. I started building them as hedged bets.
Three anchors. In this order. Every time.
1. Seasonality Before Anything Else
Before I touch a single number, I pull two to three years of data and map the peaks and valleys.
Not because history repeats exactly — it doesn't. But because the curve is real, and ignoring it is expensive.
Every business has a season. Sometimes obvious: Q4 retail, back-to-school, summer travel. Sometimes subtle: a spike after a specific holiday, a dead zone in February that the team always misreads as a channel problem.
One of the most consistently underallocated windows I've seen is Q5 — the stretch from post-holiday through mid-January. Intent stays high after the holiday season. CPMs soften. Most competitors have mentally checked out, already focused on Q1 planning. Brands that lean into this window often buy their best CAC of the year.
The goal isn't to predict the curve perfectly. It's to build a budget that rises when the season is working for you and preserves cash when it isn't — instead of spending evenly each month.
Push when seasonality is with you. Pull back when it's against you. Most teams never make this distinction explicit. They pay for it.
2. A Realistic Baseline, Not an Aspirational One
Once I understand the shape of the year, the next question is blunt: what's the actual probability we hit this target — 85%? 95%?
To get there, I break the target into three variables:
Media cost × Conversion rate × Average order value
Each one gets a real working assumption — not a hope, an assumption I can defend.
Will media get more expensive this year, and by how much? Historically, yes. Platform CPMs rise. Auction competition increases. If you're planning for flat media costs, you're planning to be wrong.
Can we realistically improve CVR — and what would actually drive that?
Can we grow AOV, and through what mechanism? Bundling? Price increases? Upsell flows?
If I can't put a number on my own confidence in these three assumptions, I don't believe the forecast. If I don't believe the forecast, the target isn't ready — regardless of what the board deck says.
This is the step most teams skip. They accept the number, build backward, and spend the rest of the year wondering why reality keeps diverging from the model.
3. Quarters Before Months
The instinct is to take the annual target and divide it into monthly numbers — either equal slices or seasonally adjusted ones. Then manage month by month.
The problem: monthly targets are too short to absorb variance and too granular to hold strategy.
One bad week in a monthly budget feels like a crisis. In a quarterly budget, it's information.
My sequence: annual target → quarterly breakdown → monthly targets → channel allocation.
Quarterly planning sets the direction and gives you room to move. Monthly planning runs the execution. If something shifts mid-quarter — a channel reprices, a creative format burns out, a macro event disrupts intent — you have room to respond without blowing up the whole plan.
The quarterly layer is where strategy lives. The monthly layer is where execution lives. When both are running together, you have a system. Not a guess.
Where Your Exposure Actually Is
You can't take the bet off the table. What you can control is how exposed you are when it doesn't go your way.
Seasonality, the baseline, the quarterly structure — pick whichever one you've never actually stress-tested. That's where your exposure is.
Which of the three anchors leaves you most exposed in your org right now — and why?
Originally published in The Growth System newsletter on LinkedIn.