What Am I Paying For Every Month?

You get an invoice every month. You pay it every month. If someone asked you to explain, line by line, what each charge buys and what would break if it stopped, you probably couldn't. That isn't a failure of attention. It's a failure of how the invoice was written.

Four charges sit on most service businesses' monthly spend: domain, hosting, maintenance, and some version of a marketing or SEO retainer. Three of them are infrastructure with a defined job. One of them is where almost all the ambiguity lives, and almost all the money.

The three line items that are actually simple

Domain. You are renting a name. That's the whole product. A domain registration keeps your address pointed at your property and keeps someone else from taking it. It is annual, it is small, and it is the single most dangerous line item to let lapse, because a lapsed domain is the one failure in this list that can be permanent. Everything else can be rebuilt from a backup. Your address, once someone else grabs it, is gone.

What you should be able to verify in thirty seconds: who the registrar is, whose name and email are on the account, and when it renews. If the answer is "my web guy has it," you don't own your domain. You're borrowing it.

Hosting. You are renting the machine the site runs on, plus uptime, plus bandwidth. It's a utility bill. The relevant questions are whether the site is up, whether it's fast, and whether backups exist and have been restored at least once to prove they work. That's it. Hosting is not strategy, and nobody should be charging strategy prices for it.

Maintenance. This is where the first real ambiguity appears, because "maintenance" can mean two completely different things. It can mean keeping things from breaking: platform updates, plugin patches, security monitoring, SSL renewals, broken-link fixes. That's custodial work and it has a defined scope.

Or it can mean ongoing development: new pages, new functionality, changes to how the thing actually works. That is not maintenance. That is building, billed as custodial work so nobody has to scope it.

Here's a way to tell the difference. Real active development leaves a trail. Our own content platform has 674+ commits and is still shipping. A separate product we built has 893 commits across a sustained four-month stretch. Those are receipts. If a maintenance line is genuinely development, there's a changelog somewhere proving it. If nobody can produce one, you're paying a custodial fee and calling it progress.

The fourth line item is where the money disappears

Then there's the retainer. SEO retainer, marketing retainer, growth retainer, whatever the invoice calls it.

Every other charge on the list has an observable object attached. A name. A server. A patched plugin. The retainer's object is a set of activities: posts published, keywords tracked, reports generated, meetings attended. Activity is not an object. Activity can continue at full volume for twenty-four months while the business it's attached to stays exactly where it was.

That's the structural problem. The three simple line items fail loudly. Your domain lapses and the site vanishes. Your host goes down and the phone rings. A retainer that isn't working fails silently, and the monthly report is specifically engineered to make the silence sound like progress. Impressions up. Rankings improved on terms you didn't choose. Posting cadence maintained. None of those statements are false. None of them are answers either.

The three questions that make any retainer legible

You don't need to become a marketer to read your own invoice. You need three questions, asked every quarter, in writing.

One: what did you change this month? Not what did you produce. What did you change. Produced means twelve posts went out. Changed means something about how buyers find you, understand you, or decide about you is materially different than it was thirty days ago. A retainer can produce forever without changing anything.

Two: what number moved, and is it a number a bank accepts? Impressions aren't revenue. Reach isn't revenue. Rankings aren't revenue. Those are leading indicators and they're fine as leading indicators, but if twelve months of them haven't produced a trailing indicator, the leading ones were measuring the wrong thing. We've seen the gap from the inside: a two-month campaign for an arts festival with a fixed opening date produced 1M+ impressions and 600K+ reach, and the reason those numbers mattered is that $15,876 in ticket revenue was tracked through the same campaign and the event sold out for the first time in its fifteen-year history. The reach was only meaningful because the revenue line sat next to it.

Three: what stops if I stop paying? This one is uncomfortable and that's the point. If the honest answer is "the posting stops," you are renting activity. If the answer is "the inbound pipeline we built keeps producing for months while we decide what's next," you bought an asset. Both can be legitimate purchases. Only one of them should be priced like an investment.

The math nobody runs

Here is the arithmetic most owners never do, because the charges are small individually and arrive automatically.

Take whatever your monthly marketing retainer is. Multiply by twelve. Now multiply by the number of years you've been paying it. That number is your real spend. Put it next to the number of clients you can name who arrived because of that work, specifically, traceably, not "well, people see us around."

If the second number is small or unknown, the problem usually isn't that the work was lazy. The work was probably fine. It was aimed at a buyer nobody had explicitly named, which means every month of competent execution compounded in a slightly wrong direction.

Compare that to what aim looks like when it's correct. An education product was about five months from being shut down by its founder. The diagnosis wasn't a marketing problem: teachers didn't want to carry heavy printed materials, so the product itself had to change before any promotion could work. Moved to digital, repositioned around how teachers actually work, then promoted: 59K+ LinkedIn impressions, 1.4K+ new followers from a founder who had no audience at launch, 14.7K Meta ad reach, 3.2% CTR at $1.27 per click, and $39,378 in revenue in about five months. For a brand that was closing.

The retainer-shaped version of that engagement would have produced the same posts against the same heavy printed product and reported healthy impressions the entire way down.

Channel quality works the same way once you can see it. A 14.6% close rate on SEO-sourced inquiries and a 3.74% referral conversion rate are not the same business problem and should not get the same budget. You can only make that call if your invoice tells you which channel produced which outcome. Most don't.

"Isn't this just what a marketing agency does?"

No, and the distinction is the entire point of the invoice question.

An agency executes. That is a real job and a good agency does it well. But execution is downstream of a decision: who this is for, what it's worth, and what you're arguing against in the buyer's head. If that decision was never explicitly made, the agency inherits whatever assumption was lying around and executes against it, flawlessly, for years.

That's why swapping agencies so rarely changes the result. You changed the executor. You didn't change the aim. More budget doesn't fix a bad aim. A better retainer doesn't fix a bad aim. The only thing that fixes it is naming, in writing and based on evidence, who the work is actually for.

Which is also why a foundation engagement should end, and a retainer by design should not. One has a finish line. The other has a renewal date.

What the invoice should look like

Here's the version you should be able to reconstruct from memory:

• Domain: annual, in your name, in your account, renewal date known.
• Hosting: monthly utility, uptime and speed verifiable, backups tested.
• Maintenance: custodial scope defined in one sentence, or reclassified as development with a changelog attached.
• Strategy or growth work: a named objective, a dated outcome, and a number a bank recognizes.

Four lines. Each one with an object attached, not an activity. If any line on your current invoice can't be rewritten in that form, that's not a billing question. That's the line item to interrogate first, because it's the one absorbing budget on the strength of nobody being able to read it.

Stop paying for activity. Start paying for a decision that makes the activity worth something.

If you can't explain what one of your line items buys, bring the invoice. We'll tell you honestly which charges are infrastructure, which are execution, and whether what you actually need is a foundation rather than another monthly. Book a strategy call.