Why content compounds: the math behind long-term SEO growth

Every channel you can buy works on the same terms: you pay, visitors arrive, you stop paying, visitors stop. Search content is the one channel where the output piles up instead — this month's traffic comes mostly from things you published months or years ago. People in marketing call this compounding, usually without showing how it works. This article shows the mechanism, the data behind it, and the math that turns a flat monthly effort into a rising traffic line.

It's the mechanism article, deliberately. When the money comes back is the payback question, and how many articles you need is its own answer too.

Google's results are full of old pages

In May 2025, Ahrefs studied the age of pages ranking in Google and found that 72.9% of pages in the top 10 are more than three years old — up from 59% when they ran the same study in 2017. The average page holding the #1 spot is five years old; in 2017 it was two. Only 13.7% of top-10 pages are less than a year old.

Read those numbers as a description of what a ranking is: an asset that gets harder to dislodge as it ages. A page that ranks gets read, linked to, and clicked, and each of those signals helps it keep ranking. Some of this is survivorship — the pages that lasted are by definition the ones you see — but the loop is real, and it's the first half of compounding. The article you published in 2024 isn't just still around in 2027. If it held its position, it's more entrenched than it was.

Old posts do most of the earning

The second half is what an archive does to a monthly traffic report. HubSpot analyzed its own blog and found that 76% of monthly page views came from posts published in earlier months, and 92% of the month's leads did. Whatever was published that month barely registered against the accumulated library.

The same analysis found the earning heavily concentrated: 46% of monthly leads came from just 30 posts, out of roughly six thousand. In a separate study of its archive, HubSpot found about one post in ten becomes what they call a compounding post — traffic that grows over time instead of spiking at publication and fading — and that minority generates 38% of the blog's total traffic. So the library does the work, and inside the library, a small group of veterans does most of it.

The arithmetic of the rising line

Here is the whole trick in one model. Suppose you publish four articles a month, and suppose — realistically — that only one of the four ever ranks for anything meaningful, settling at 300 visits a month. Both numbers are assumptions; swap in your own. The shape is what matters.

Month one gives you almost nothing. By the end of year one you have twelve ranking articles sending roughly 3,600 visits a month. By the end of year two, twenty-four articles and 7,200 visits. Your effort never changed. The line rises anyway, because each month's traffic is the sum of every article still ranking — the new ones are only the newest layer.

This month's traffic is the sum of every article that still ranks, which is why flat effort can produce a rising line.

Now run the same budget through ads. It buys about the same number of visits in month twenty-four as it did in month one, and the month you stop, it buys zero. That isn't a reason to never run ads — they're immediate, and content is slow. It's the reason the two curves cross: one channel's output resets every month and the other's accumulates.

Why compounding is never automatic

This is where the story usually gets oversold, so here are the three conditions doing the quiet work in that model.

• Most new pages never join the curve. The same Ahrefs study found only 1.74% of newly published pages reach Google's top 10 within a year, down from 5.7% in 2017. Much of that reflects the internet's enormous volume of abandoned and thin pages, but the direction is real: the compounding math applies only to pages that rank, so choosing topics you can actually win decides everything before the first word is written.
• Rankings decay without maintenance. Competitors publish, information goes stale, and a neglected page drifts down the results. The repair is cheaper than it sounds: when HubSpot went back and updated its old posts, their organic search views rose by an average of 106%. A compounding library needs a maintenance schedule, not just a publishing schedule.
• You can't pick the winners in advance. If roughly one post in ten compounds, the strategy is a portfolio: publish consistently on one subject, watch which pages show movement, and put your update time behind those. Publishing four posts and waiting to see which one wins beats trying to write the winner on the first attempt.

What this changes about planning

Budget in years, not months — the curve's early flat stretch is the price of the later rise, and the break-even math tells you how long your own flat stretch should be allowed to last. Keep the topics narrow enough to build authority, since depth is what lets articles reinforce each other. And once the library is real, judge the program by the cumulative line, not by any single month. A blog with ninety articles that adds four more this month didn't grow 4% — it grew by whatever those four will still be sending you in three years.