Publishing velocity: when speed matters, and when it doesn't

Almost every content plan built in the last decade rests on one number. Companies publishing 16 or more posts a month get roughly three and a half times the traffic of companies publishing four or fewer. It gets quoted in pitch decks, in board reviews, in the paragraph of the strategy doc that justifies the headcount. Follow it back to HubSpot's post, which is where it lives, and the sentence hyperlinks to a second HubSpot page presented as the source. That page has no data in it. No sample size, no methodology, no mention of the figure at all. It is a first-person essay about blogging frequency. Secondary accounts trace the original analysis to a study of thirteen thousand-odd companies somewhere in the 2011 to 2015 window, but no live primary report appears to exist. So the doctrine that shaped a decade of content operations is a decade-old finding with a broken citation chain, from a period before mobile-first indexing, before BERT, before AI Overviews existed. That doesn't make it wrong. It does mean nobody has checked.

Google wrote its position down, and it is not subtle

Search Central's guidance on helpful content contains a self-assessment list. Three of the questions are aimed squarely at volume, and they are phrased as warnings rather than encouragements.

Verbatim from Google Search Central, "Creating helpful, reliable, people-first content," last updated 10 December 2025. The parenthetical is Google's, not ours.

The scaled content abuse policy, introduced in March 2024 and last revised in May 2026, makes the same point with enforcement behind it: generating many pages "for the primary purpose of manipulating search rankings and not helping users… no matter how it's created." That last clause matters. The policy is indifferent to whether a human, a model, or a team of freelancers produced the pages. Volume plus thin value is the trigger. Google estimated ahead of the March 2024 rollout that the combined changes would cut low-quality, unoriginal content in results by 40%. When it finished shipping in April, it revised that upward, to 45%. Most coverage still quotes the pre-launch estimate.

The arithmetic of one more post

Set policy aside for a moment and just look at the odds.

New pages1.74%

Reach the top 10 within a year

Same figure, 2017 study

Down from 5.7% eight years earlier. 1M URLs tracked for twelve months.

Incumbents5 yrs

Top-10 pages older than 3 years

Top-10 pages under 1 year old

Average age of a #1-ranking page, up from two years. 1.3M US keywords.

Fewer than two new pages in a hundred reach page one inside twelve months. Nearly three-quarters of the pages already sitting there are more than three years old. Separately, Ahrefs' index-wide analysis of roughly fourteen billion pages puts the share receiving zero organic traffic from Google at 96.55%, and Backlinko's nine-hundred-million-post study found 94% of blog posts earn no external backlinks at all. Every additional post is a lottery ticket in a game where incumbency compounds. That's not an argument for publishing nothing. It's an argument that the marginal post is a weaker instrument than the marginal hour spent on a page that already ranks.

Where speed genuinely earns its keep

Freshness is real. It is just narrower than the folklore. Google runs what it calls query-deserves-freshness systems, "designed to show fresher content for queries where it would be expected." The examples in the documentation are breaking news and newly released films. It is query-scoped. It is not a site-wide reward for a busy publishing schedule, and Google's own byline-date guidance explicitly warns against manufacturing recency by changing dates. The AI-answer picture is more interesting, and it cuts against the obvious assumption. Ahrefs measured just under 17 million cited URLs across seven AI platforms and found the average AI-cited page was 1,064 days old against 1,432 for organic results — about 26% fresher in aggregate. But break it out and Google AI Overviews cites content 16 days older than the organic average. The freshness effect is concentrated in the standalone assistants, and even there the typical cited page is nearly three years old. A later Ahrefs study went further. Looking at 1.4 million ChatGPT prompts and comparing pages that were cited against pages that were retrieved but passed over, the cited pages skewed older. Median age around 500 days, with some cited pages past 2,700. Being new appears to help you get retrieved. Being established is what gets you quoted. News queries are the genuine exception, and there the gap is sharp.

What the one survey everyone quotes actually found

Orbit Media has run an annual blogger survey since 2014. The 2025 edition, fielded that August across 808 content marketers, is the closest thing to a real dataset on cadence, and it deserves better reading than it usually gets. Yes, higher-frequency publishers more often report strong results. But the two effects the survey quantifies most sharply aren't about cadence at all. Bloggers spending six or more hours per post reported strong results 26% of the time, against 10% for those spending under two hours. Those publishing pieces over 2,000 words reported strong results 39% of the time, against a 21% benchmark across everyone. And the headline number is going the wrong way. Twenty-one per cent report strong results now. Five years ago it was thirty. The frequency correlation is also the most obviously confounded thing in the study. Organizations that can publish weekly have bigger budgets, older domains, more distribution, and more people. Cadence is a symptom of resourcing as much as a cause of results, and a self-selected survey with self-reported outcomes cannot separate the two.

The version of this that holds up

Velocity is not a ranking input, and the only party in a position to know has said so in writing. What velocity actually buys you is surface area — more chances to be the page that happens to match a long-tail question, and more raw material for the topical depth that does correlate with authority. That's a real benefit. It is not three and a half times the traffic, and it turns negative the moment volume outruns the effort per page, because at that point you're producing exactly the thing the scaled content abuse policy was written to catch. The practical read: pick the cadence you can sustain at six-plus hours a page, then spend the rest of the budget on the pages that already rank. Publishing faster than you can do that isn't ambition. It's a rounding error with a calendar attached.

Sources

Related reading

• Real movement versus measurement noise — how to tell a genuine gain from a number that was always going to wobble.
• A single content score can't tell you what to fix — the same failure mode as the 16-posts-a-month stat, in metric form.
• Retrievable, extractable, citable, reusable — where the hours go if they are not going into more posts.
• How it works — tracking competitor publishing velocity against your own, run over run.