You are buying an archive, not a series of posts

Most companies decide whether to blog by imagining a single post. They picture one article, estimate what it might plausibly bring in, and conclude the arithmetic does not work. The arithmetic does not work. The conclusion is still wrong, because the post was never the thing being bought — the asset is the archive, and an archive behaves nothing like the sum of its parts.

Per-post return is a question with no useful answer

Ask what a specific post earned and you get a number that is technically true and practically useless. Traffic across a working archive is lopsided in a way that surprises people the first time they look. A small minority of pages carry most of the arrivals, a long middle does modest steady work, and a tail does almost nothing you can measure. Sort your own analytics by sessions and the shape is obvious in about ten seconds.

The skew is not the problem. The problem is that it is unknowable in advance. Ask anyone who has run a blog for three years which page brings the most traffic, and the answer is almost never the one they were proudest of. It is usually something narrow that somebody eventually searched for, and only one page on the internet had bothered to answer it properly.

That makes per-post evaluation actively harmful rather than merely imprecise. Judged one at a time, on a horizon short enough to report on, most posts fail. Cut the failures early and you remove the tail that contained the winners, because on publication day nothing distinguishes them.

Coverage is a property of the set, not of any member

One post about a subject ranks for a handful of phrasings around that subject. Thirty posts about adjacent subjects rank for a surface: the odd, specific, low-volume questions that individually bring four visitors a month and collectively bring most of the qualified traffic a blog produces. No single post creates that surface. It only exists once there are enough pages for the gaps between them to close.

This is also why the thirtieth post outperforms the first without being better written. It arrives into a structure. There are existing pages to link it from and link it to, an established subject it deepens rather than introduces, and a domain that has published in that territory long enough for Google to have formed a view. The first post has none of that available to it, and no amount of effort supplies it. We have argued separately that internal linking is the cheapest SEO work most teams skip, and it is cheap precisely because the archive has already paid for it.

So the early posts are not underperforming. They are doing the only job available to them, which is to exist so that later posts have somewhere to arrive.

An archive is what you still own when you stop

Compare it with the alternative. Paid acquisition is a rental agreement. The traffic is real, it arrives the day you pay, and it stops the day you stop. An archive is closer to a capital asset: it depreciates slowly, needs maintenance, and keeps working through quarters when nobody touches it.

That durability is the actual argument for owning a blog, and it is an unfashionable one, because it pays out on a timescale longer than most marketing plans run. A company that publishes for two years and then goes quiet still has two years of pages answering questions. A company that runs ads for two years and stops has whatever brand memory it built and no pages at all.

An archive of pages nobody needed is a liability

None of this makes volume valuable on its own, and the honest version of this argument has to say so. Thirty pages that restate each other do not produce coverage. They produce thirty pages competing for the same phrase and a site that looks, to readers and to Google, like it is filling space. Google names this directly in its spam policies under scaled content abuse, and the test it applies is whether pages were made to help people rather than how they were produced — which is the same distinction we made in our post on why Google does not penalise AI content.

The constraint that matters is whether each page answers a question somebody actually has. That gets harder to satisfy as the archive grows, and this is the real cost of the argument: post thirty is worth more than post one and also harder to justify writing, because the obvious subjects are gone by then. Sourcing becomes the binding constraint long before writing does.

What changes if you accept this

Two things. You stop asking whether a given post paid for itself and start asking whether the archive is compounding: whether coverage is widening, whether older pages still get found, whether new posts have somewhere to link. Those questions have answers. The per-post one does not.

And the decision about starting changes shape. It is not whether one article is worth writing. It is whether owning thirty pages that answer your buyers' questions better than anyone else does would be worth having in two years. If that answer is no, do not start — it is a legitimate answer and it fits plenty of businesses honestly. If it is yes, then the first ten posts are an infrastructure cost rather than a campaign, and judging them like a campaign is how most company blogs quietly stop at month four.