In short
Publishing frequency is not a ranking factor. What matters is whether each piece is worth publishing and whether you can sustain the pace without quality falling. For most small teams that means one genuinely useful piece a fortnight or month, not several thin ones a week — and updating existing pages often returns more than adding new ones.
Frequency is not a ranking factor
Google does not reward a schedule. There is no threshold at which publishing more causes ranking to improve.
What is true, and gets misread as frequency mattering: sites publishing regularly tend to have more indexed pages, more opportunities to match queries, and more chances to earn links. That is a volume-of-opportunity effect, not a reward for consistency — and it only applies when the pages are individually worth having.
Publishing weak pages weekly is worse than publishing nothing, because site-level quality assessment means thin content drags down the pages that were working.
The real constraint
The question is not how often you should publish. It is how often you can publish something genuinely worth reading, given who is writing and what else they do.
For a small team where the writers are also doing the client work, that is usually one substantial piece every two to four weeks. Ambitious plans for three a week almost always degrade into thin posts within two months, and the archive of thin posts outlives the enthusiasm.
Be honest about capacity up front. A sustainable cadence you actually maintain beats an aspirational one you abandon — partly because abandoning it leaves a visibly stale site, which is worse than a small consistent one.
Depth usually wins
For most professional-services and product sites, one thorough piece beats four shallow ones covering the same ground.
A comprehensive article can rank for many related queries, earn links, be cited by answer engines, and remain useful for years. Four thin ones on subsets of the topic compete with each other, satisfy nobody completely, and often end up near-duplicates of one another.
The exception is genuinely distinct questions. "How much does X cost" and "how long does X take" are different searches deserving different pages, even though they cover the same project.
Updating often beats publishing
The most under-used move in content strategy: improve what already ranks.
A page on page two, given real improvement, can move to page one. A new page on the same topic starts from nothing. The existing page already has accumulated signals and history.
So build updating into the cadence. Every month, look at pages sitting just below where they should be and ask what would genuinely make them better — a missing section, out-of-date figures, a question the ranking pages answer and you do not.
This matters more for anything with dates, prices or version numbers, where a stale page is not merely old but wrong, which is a trust failure rather than a freshness one.
One caution: changing a date without changing the content is not an update. It fools nobody and is a habit worth avoiding.
Consistency, honestly considered
Consistency matters, but for practical reasons rather than algorithmic ones.
It keeps the habit alive. It gives returning readers a reason to return. It means crawling happens more predictably. And it prevents the pattern where a site publishes eleven pieces in a burst and then nothing for a year, which reads to a visitor as a business that stopped.
None of that requires weekly output. Monthly, sustained for three years, produces a substantial and coherent library. Weekly, sustained for two months, produces an abandoned one.
A cadence that works
What we would suggest for a small team:
One substantial piece a month, properly researched, answering a question people actually search.
One meaningful update a month to an existing page — real improvement, not a date change.
Occasional short pieces when there is something specific to say — a change in the field, a result from your own work. Unscheduled, and only when there is something.
That is roughly two pieces of work a month, which most teams can sustain indefinitely, and it compounds. Twenty-four substantial pages after a year is a real asset.
What actually determines success
Not the schedule. Whether each piece answers a question someone has, better than what already ranks, with something in it that only you could have written.
One piece a month meeting that bar beats twenty a month that do not — and the second approach carries risk the first does not.
If you are setting up a content programme and want a realistic cadence rather than an aspirational one, book a call.
Common questions
Does publishing more often improve rankings?
No — frequency is not a ranking factor. Sites that publish regularly tend to have more indexed pages and more chances to match queries and earn links, but that is a volume-of-opportunity effect and it only applies when each page is individually worth having.
How often should a small business publish?
As often as it can produce something genuinely worth reading — usually one substantial piece every two to four weeks for a team whose writers also do the client work. Ambitious plans for several a week reliably degrade into thin posts within a couple of months.
Is it better to publish new content or update old content?
Updating is under-used and often returns more. A page sitting on page two already has accumulated signals and history, so real improvement can move it up, whereas a new page on the same topic starts from nothing. Build a monthly update into the cadence alongside new writing.
Should I write one long article or several short ones?
Usually one thorough piece, which can rank for many related queries, earn links and stay useful for years. Several shallow pieces on subsets of the topic compete with each other and often become near-duplicates. The exception is genuinely distinct questions, such as what something costs versus how long it takes.
Does changing the date on an old post help?
No, and it is a habit worth avoiding — changing a date without changing the content fools nobody. Genuine updates matter most on pages carrying prices, dates or version numbers, where a stale page is not merely old but factually wrong, which is a trust problem rather than a freshness one.
