Four posts a month, forever, measured by whether they arrived.
That is the standard content retainer, and it is a good business to sell because the deliverable is countable and the outcome is not. Three years in, the client has a hundred and forty posts, no idea which twelve produce anything, and two pages competing for every term that matters.
Two posts on the same topic split the signal between them, and both rank worse than one would. This is the single most common defect in a library over fifty pages, and volume-based retainers create it structurally.
A post that ranked in 2022 and slipped in 2024 usually needs updating, not replacing. Refreshing an existing page is cheaper and works faster than writing a new one, and it is almost never in the plan.
Deleting a page feels like destroying an asset you paid for. Sometimes it is the asset that is costing you, and there is no incentive for the agency who wrote it to say so.
Six steps, and two of them are subtraction.
This is the loop, not a launch plan. It runs continuously, and the useful version of it spends as much attention on the library you already have as on anything new.
Every page you have, with a keep, merge, or retire verdict and the reason written down. This is the deliverable most clients have never seen and the one that changes the plan.
What people actually search, which of it you can realistically win, and which questions your buyers ask that nothing on your site answers.
Competing pages consolidated into the one that should win, with redirects. This happens before anything new is written, because writing into a cannibalized library wastes the new page too.
A schedule you can actually staff, and briefs specific enough that a writer who is not you produces something you would put your name on.
Structured so a person can read it and a machine can lift a clean answer out of it. Those two goals agree more often than people expect.
A quarterly read on what moved, what decayed, and which pages are one update away from ranking again. Refresh beats replace almost every time.
Six years, 18 markets, one editorial calendar.
Content operations for a global blog program at a Fortune 100 technology company, in house. Keyword research through brief, publication, tracking, refresh, and retirement, on a library large enough that the retirement decisions mattered more than the new ones.
I will not name the company, because promoting my own practice on my employer's name would be the wrong thing to do while I work there. And I will not publish that program's traffic or revenue figures, because they are not mine to publish.
If a consultant shows you an impressive number from a job rather than a client, ask them which of those two rules they broke to do it.
What people ask about content.
What is content SEO?
The part of search work that decides which pages should exist and what each one says, as distinct from technical SEO, which decides whether a search engine can read them at all. In practice it is demand and intent research, the page structure that follows from it, the writing itself, and the continuing decisions about what to refresh and what to remove. That last part is the one usually missing, and it is the reason libraries stop working after about fifty pages.
Can I do this myself?
For a site under about twenty pages, genuinely yes, and I would rather tell you that than sell you a retainer you do not need. Pick the ten questions your customers actually ask, answer each one properly on its own page, and check Search Console every quarter to see which ones landed. Where it gets hard is judging cannibalization across a larger library, and knowing when a page that used to work needs updating rather than replacing. That is the point where an outside read pays for itself.
Do you write the content, or plan it?
Both, though the planning is worth more and I would rather sell you that. I write where the subject needs someone who understands search, and I brief where you or your team knows the subject better than I do, which is more often than agencies admit. A brief specific enough for someone else to execute is a deliverable in its own right.
How many posts a month?
I do not sell it by the post, because volume is the metric that produced the problem. A typical first quarter is heavier on inventory and refresh than on new writing, and it is common for the first month to publish nothing new at all while three existing pages get merged and two get retired. If you need a fixed monthly count for a budget line, say so and we will set one, but I will tell you what it costs you in results.
Will deleting pages hurt my traffic?
Retiring a page means redirecting it to the page that should have had that traffic all along, not deleting it into a 404. Done that way it usually helps, because the two pages stop competing. It can go wrong, which is why the inventory states a reason for every verdict and why the merges happen before anything is removed.
Can AI write this for me instead?
It can produce the draft, and for some formats that is genuinely fine. What it cannot do is decide which page should not exist, which is the expensive judgment. It also cannot supply the thing that makes a page worth citing, which is something true that only you know. Assistants are increasingly the audience for this work, and they reward specifics they can attribute, which is the opposite of what generated filler provides.
How long before content pays back?
Refreshing pages that already rank can move within weeks. New pages typically need four to six months to compete, and anyone promising faster is guessing. Content is the slowest of the five services on this site and the one that compounds hardest, which is why it is a poor choice if you need something to move this quarter and the right choice if you intend to still be here in three years.