On Compounding

Most marketing budgets are not marketing budgets. They are rent.

You pay for impressions, and the impressions stop when you stop paying. You pay for creative, and the creative ages out in a quarter. You pay for agency hours, and the hours leave with the person who billed them. The money moves through the business without leaving anything behind.

This is not a failure of any particular spend. It is the default shape of the function. Marketing, as most companies practice it, is designed to be rented. The platforms are rented. The audiences are rented. The expertise is rented. When the check stops, the results stop. The business owns almost none of what it bought.

I want to propose a different way of thinking about this, because I think the next decade will be won by the companies that figure it out.

There is a different kind of marketing spend. It builds systems that keep working after you stop paying for them.

An owned email list is not rented. A brand that customers actually remember is not rented. A measurement layer that tells you the truth about your business is not rented. A team of senior operators who understand the category is not rented. An AI system trained on your proprietary data is not rented. A community that exists because of something you built is not rented.

These things compound. The email list is more valuable in year three than in year one. The brand gets stickier every quarter. The measurement layer answers harder questions over time. The team gets better at seeing around corners. The AI system gets smarter the more it runs. The community becomes self-sustaining.

The interesting thing about compounding assets is that they are usually not the things that show up in the quarterly report. A founder looking at the P&L sees acquisition costs and agency fees, because those are the line items. The owned distribution layer does not have a line item. The brand does not have a line item. The measurement layer does not have a line item. The compounding happens off-balance-sheet, which is part of why most companies do not invest in it.

The companies that will win the next decade are the ones that shift their marketing spend from rented to owned. From expensing to compounding. From renting attention to building the systems that make attention come to them.

This is harder than it sounds, and the reason it is hard is the reason it is worth doing.

Rented marketing has fast feedback. You run an ad, you see the click-through, you know within a week whether it worked. The feedback loop is tight, which is good, but it is also part of the trap — because the things that show up in a tight feedback loop are the things that are easy to measure, and the things that are easy to measure are rarely the things that compound.

Owned distribution takes eighteen months to matter. Brand takes years. A real measurement architecture takes a quarter to build and another quarter to trust. An AI system that actually runs the function is a six-month project minimum, and most of its value accrues in year two. None of this feels like marketing is supposed to feel. None of it gives you the hit of the weekly optimization meeting.

The operators I have seen make this shift successfully share a specific trait: they are comfortable with the gap between when they spend and when they see the result. They do not mistake activity for progress. They do not confuse the weekly report for the business. They know what the owned layer is worth even when it does not show up in the ad account.

The companies that do not make this shift, in my experience, are not blocked by strategy or resources. They are blocked by the discomfort of building something they cannot measure in a week. The path of least resistance is the path they already know. The easy path is the one that got them here. And the easy path does not build the business they actually want to be running in five years.

I want to be careful here not to overstate this. I am not arguing that companies should stop spending on paid acquisition. Most of the businesses I work with still run meaningful paid media, because paid media still works and because the owned systems take time to carry the weight alone.

What I am arguing is that the allocation question is the most important question in marketing right now, and most companies have it backward. Most spend most of their budget on the rented layer and treat the owned layer as an afterthought. The allocation that compounds is the opposite shape — most of the investment goes into the systems that stay, and the paid layer becomes the distribution engine for the assets you have built, not the asset itself.

A company that runs this allocation for three years looks very different from a company that runs the default allocation for three years. The compounding version has a list it can ship to. It has a brand that makes acquisition cheaper because customers come in warmer. It has a measurement layer that tells it which experiments to run. It has a team that has seen every problem before. Its marketing function becomes a flywheel. The default version still has to pay for every customer, every quarter, forever.

The difference between these two shapes, ten years in, is not incremental. It is categorical. One business owns its marketing function. The other is permanently renting it.

The thing I would tell any operator thinking about this: the decision to build compounding marketing is not a strategic choice. It is a temperament choice. It requires you to be willing to spend money today on things that will not show up in the dashboard until next year. It requires you to say no to the weekly optimization meeting and the quarterly campaign launch and the constant hit of small, legible wins. It requires you to believe that the things that are hard to measure are sometimes the things that matter most.

If you cannot make that trade, the compounding path will never feel right. You will keep going back to the rented layer because that is where the feedback is.

If you can make that trade, the compounding path is the only marketing decision that matters in the next decade. Everything else is optimization on top of a function that will be commoditized.

The brands that figure this out first will own their categories. The brands that do not will be renting forever, and the rent keeps going up.

Pick carefully.