The Revenue Stack

There is no one thing.

I said a version of that in the first note of this series, and it's worth repeating now that the pieces are on the table. Not subscriptions, not advertising, not philanthropy, not events, not AI, not a better CMS or one rich donor who believes in democracy. Any model that depends on one thing eventually gets held hostage by it. A newsroom funded mostly by ads is hostage to the ad market. Mostly by subscribers, hostage to churn and household budgets. Mostly by grants, hostage to someone else's priorities. And a newsroom funded mostly by platforms isn't funded at all. It's rented.

The durable version is a stack: several lines of revenue, each built on the asset you actually own, each doing a different job, none of them carrying the whole weight by itself. This note is the map of the series in one place. If the early layers look familiar, that's because they are. The point here is how they fit together, and the two layers nobody talks about.

The floor is habit

The first layer is the everyday: weather, closings, scores, road work, election results, the names people know. Mostly free, on purpose, because free utility builds the behavior everything else depends on. That was the whole argument of the second note, so here I'll only add the stack view: without habit, every other layer has to work too hard. Ads underperform because the audience is thin. Subscriptions stall because there's no routine to convert. The data loses its signal, the events lose their pull, and the services lose the proof that you understand the place. Habit is the floor, and the floor gets built first.

Two engines on the floor

Advertising isn't dead. It just can't be the whole business anymore. In the stack, its job is narrower and more honest: monetize the reach. Sponsorships around the utility, ownership of the local categories, the email, the alerts, the video, the surfaces where the audience actually shows up. It funds the front door. It doesn't carry the house.

Reader revenue does the opposite job. It funds the depth: the investigations, the beat expertise, the work people would miss if it disappeared. The habit earns the right to ask. The paywall isn't the starting point, it's the conversion point. A reader who already opens you three times a week is close to paying. The stranger who landed once from search never was.

The lines you're not running yet

Then come the lines the middle of this series argued for, each one a full note on its own.

Market intelligence sells what you know: the quarterly brief, the dashboard, the custom research for the bank, the developer, and the hospital making expensive decisions in the dark. Aggregated and privacy-safe, always. The trust pays for everything else in the stack, so no single product gets to spend it.

Services sell the relationship and the knowledge behind it: putting your read of the market to work for the businesses that already return your calls, stacking whatever service they need next on an understanding no platform has, with a person who picks up the phone, and with the editorial firewall standing every single time.

Verticals and events sell the gathering: the prep sports site, the housing product, the room where the right people show up because of whose name is on the invitation. Don't chase every product you could build. Build where your market already gathers, and skip the rest.

Retention is the layer everyone underbuilds

Now the part the strategy decks skip, and the part twenty years of carrying a number won't let me skip. Creating revenue is half the job. Keeping it is the other half, and almost everyone underinvests in it, because retention never looks heroic next to a launch.

But a stack only compounds if the layers hold, and holding them is unglamorous operating work. Scope clean enough to renew without a fight. Pricing that reflects what the work is actually worth. Knowing which clients are wobbling before the cancellation email arrives. Giving a subscriber a real reason to stay on the day they decide to leave. Knowing the true margin on every line, including the ones everybody loves. I've watched more revenue die from sloppy renewals than from bad ideas. A new line that churns out the back isn't growth. It's motion.

So measure every line, give every line an owner, and make each one prove, in numbers, whether it's creating revenue, protecting it, or quietly eating more than it returns. Run them as business units, not initiatives. That's the difference between a stack and a list of ideas.

Sequence is the strategy

The other thing the decks skip: order. A stack isn't built by believing in all the layers at once. It's built by choosing the next one, making it pay, and then adding the next without breaking the operation underneath it.

Habit almost always comes first, because everything leans on it. After that, the right next build isn't the most exciting one. It's the one with the clearest path to revenue, the least fantasy in the operating model, and the best odds of proving to a tired team that the business can still learn. Get two builds paying and the third gets easier, because now you have margin, momentum, and a staff that believes.

And don't run the sequence alone. The previous note argued the publisher one county over was never your competition. Their flopped pricing test and the newsletter that took off two markets away are tuition you didn't have to pay. Compare notes with the markets that look like yours, and every build on this list gets cheaper to choose.

Where AI earns its place

Same rule as everywhere else in this series: the tools earn their place where they change the math, and nowhere else. In the stack, that's the everyday utility carried without burning out the newsroom, the century of data made answerable, and the hundred small accounts served at a margin you can keep. And one place nobody mentions: the retention layer, where the client reporting, the renewal prep, and the early-warning signals are exactly the kind of repeatable work the tools are good at. AI doesn't add a layer to the stack. It makes the layers affordable.

The part you can copy, and the part you can't

The model isn't the secret. I just handed it to you, and I'd do it again, because the map was never the hard part. Build the habit. Monetize the reach. Charge for the depth. Sell what you know, sell your help, gather your people, and keep what you earn.

The hard part is the sequence inside your market: what belongs, in what order, at what price, run by which people, against which constraints. That's where a framework becomes a business. It's also where most of these plans die, not because the thinking was wrong, but because nobody ran it like they meant it.

The last two notes in this series are about exactly that: first the systems you run the model on, then the seriousness you run it with.