Should you cut upper-funnel budget before Peak? I Fospha Academy

Upper-funnel spend gets cut first because it doesn’t show an immediate return, and most reporting is built to only credit the last thing someone clicked before buying. Cutting it doesn’t save that money, it just turns it into a tax. That tax gets collected later, during peak, at the highest rate of the year, once you’re rebuilding a cold audience and a bidding algorithm that lost its training data.

Most of us have already sat through this exact budget meeting more times than we can count. Someone pulls up a spreadsheet, points at the awareness line, and asks why it doesn’t have a clean ROAS next to it like the conversion campaigns do. Everybody in the room is reading the report in front of them correctly. The report just wasn’t built to answer that question.

What the report isn’t telling you

Last Click attribution gives credit to whatever a customer touched right before they bought. If someone watched a brand video in June and bought in October after searching your name directly, that sale gets filed under search. The video that got them there shows up as nothing. Not underperforming, just invisible. We call this the halo effect, real lift in direct traffic and branded search that never gets traced back to what caused it. Once you know to look for it, you start reading those flat-looking upper-funnel numbers a lot differently.

That’s the first trap. The second one is slower, and it’s the one that really costs money.

The tax you don’t see coming

You must have heard about ad stock or the carryover effect. It describes something most of us already sense but rarely name: an ad’s effect on a customer doesn’t switch off the moment you stop paying for it. It fades on a curve. Cut the spend today and the damage doesn’t show up today. It shows up in three weeks, then six, as that audience quietly cools off and forgets you exist.

Now add what’s happening on the platform side. Most peak campaigns run on some version of automated bidding, Advantage+, Performance Max, whatever your brand already uses. These systems need a steady stream of real conversion data to get good at finding buyers. Pause a campaign for two months and you don’t just lose the audience, you lose the training the algorithm was doing. Turn it back on in November and it’s not picking up where it left off. It’s starting cold, in the most competitive, highest-priced week the platform sees all year.

That’s the tax. You didn’t avoid the cost by cutting the spend, you deferred it, and peak is when it’s collected, with the audience and the algorithm both starting from zero at the worst possible prices. It’s the same idea Niket Shah, co-founder of Acceler8 Labs puts to agencies working through restructures before peak: every delay has a tax, and you want to pay it while rates are still low, not after they’ve spiked.

What top brands protect

We want to be clear about what this means, because it’s easy to hear “protect upper-funnel spend” as a nice-sounding idea that doesn’t survive contact with a real budget review. In Fospha’s client data, brands in the top quartile for Q4 ROAS held at least 20% of their funnel budget in Awareness and Consideration every quarter of the year, peak included. The other 75% sat at 9 to 11%. That gap didn’t open up under pressure in Q4. It was already there in January, and it just never closed.

We think about it a bit like training for a race. You wouldn’t stop running for two months and expect to show up on race day at the same pace. The fitness doesn’t pause quietly in the background waiting for you. It fades, and getting it back takes longer than it took to lose it. Upper-funnel spend works on the same clock. It’s not the flashy part of the plan, but it’s the part that determines how hard your peak budget has to work later.

The 20% gut check

Before your next budget review, pull up your last four quarters and answer these questions:

What share of your funnel budget went to Awareness and Consideration each quarter? Write down the actual number, not your best guess.
Did any quarter drop below 20%? If so, which one, and what was happening in the business at the time?
If you cut this spend further right now, when would you need to turn it back on, and what would that cost at peak pricing versus today’s?
What’s the smallest number you’re willing to protect through peak, even when someone asks you to find savings?

Bring actual numbers into the room. A benchmark only means something once you’ve measured yourself against it.

None of this is about spending more, it’s about choosing when you pay. Hold the 20% now, while the audience is still warm and the algorithm still remembers who buys from you, and peak gets easier. Cut it now, and you’re just moving the tax to the one week you can least afford it.

FAQs

Why is upper-funnel spend the first thing cut from marketing budgets?

Because it doesn’t show up as a direct return in the same week it’s spent, and most attribution models only credit the last click before a sale, so upper-funnel campaigns look weaker than they really are.

What happens if you cut upper-funnel spend for a quarter?

The audience you built starts cooling off gradually rather than all at once, and any automated bidding tied to that campaign loses the data it needs to keep performing, so restarting it later costs more than pausing it saved. It’s a tax deferred, not a cost avoided.

What is the halo effect in marketing?

It’s the lift a channel creates in other parts of the funnel, like direct traffic or branded search, that doesn’t get credited back to the original channel because last-click reporting only counts the final touchpoint.

What percentage of budget should go to upper-funnel spend?

Fospha’s data shows the top 25% of brands by Q4 ROAS kept at least 20% of their funnel budget in Awareness and Consideration every quarter. It’s a benchmark from that client base, not a fixed rule for every business.

Is this based on real data or just a theory?

It’s based on a consistent pattern across Fospha’s client base rather than a controlled experiment, so treat it as a strong planning signal, not a guarantee for every brand.