Where Marketing Actually Moves EBITDA

In a value creation plan, marketing typically appears under growth. That’s the slowest and least certain of the four ways it actually affects earnings, and treating it as the only one leaves the faster ones untouched.

Lever one: spend that stops

The fastest and most certain, and it requires no new capability.

Nearly every company in the lower middle market funds marketing it cannot defend with evidence. Channels running since before anyone measured. Retainers nobody has audited against output. Overlapping tools. In practice this is commonly 15 to 30 percent of the marketing line.

It drops straight to EBITDA and it takes a quarter. The reason it persists in most companies isn’t ignorance, it’s that stopping spend requires overruling whoever advocated for it, and nobody below the executive level has the standing to do that.

Lever two: mix, without changing the total

Reallocating the same budget toward what actually produces customers, rather than what reports well.

This is where honest measurement pays for itself. A company that has been optimizing to cost per lead is almost certainly overfunding a channel producing volume that doesn’t close, because cost per lead is the easiest metric to improve and the least connected to revenue. Moving that money at constant total spend improves contribution without a budget conversation.

The prerequisite is reporting that reconciles to booked revenue. Without it you’re reallocating on the same fiction that produced the current split.

Lever three: conversion, which is the underrated one

Improving the rate at which existing demand becomes revenue.

Most companies have never measured stage-to-stage conversion honestly, so nobody knows which step is worst. Fixing the follow-up sequence, the routing, the offer, or the page everyone lands on regularly returns more than a media increase, at a fraction of the cost, and it improves the economics of every channel simultaneously.

A ten percent improvement in close rate on existing volume beats a ten percent increase in traffic, costs less, and doesn’t require ongoing funding to maintain.

Lever four: growth, which is real and slow

New demand does move earnings. It’s just the slowest of the four and the least suited to a short hold period if you start it late.

Owned demand compounds: organic search, an engaged list, review inventory, referral flow. It also survives a budget cut, which is why it matters disproportionately at exit. But it takes quarters, which means it belongs in year one of a five-year hold rather than year four.

Paid media is faster and it’s rented. Turn it off and it stops. That’s sometimes exactly right, and it shouldn’t be mistaken for building an asset.

The multiple, which isn’t EBITDA but matters more

The fifth effect, and the one usually left out of the marketing conversation entirely.

A buyer pays for the probability that revenue continues without the current owners. A marketing function with documented allocation, reconciled reporting, diversified sourcing and real owned demand underwrites that probability. One that lives in a founder’s judgment does not, and gets discounted for it. Why founder dependency costs you at exit.

That effect operates on the multiple rather than on earnings, which means it can be worth considerably more than the operating improvements, and it takes the longest to build.

Sequencing across a hold period

Roughly: subtract in the first two quarters, fix mix and conversion through year one, build owned demand from year one onward so it has time to compound, and make the function legible to a buyer starting eighteen months out.

Doing it in the other order, growth first, produces activity that nobody can evaluate against a baseline that was never established. Which is why the first hundred days are for instrumentation.