"How many posts should I publish per week?" has no universal answer, and any article that gives you a single number is probably selling you a schedule rather than a method. A publishing cadence is a capacity decision: it depends on how much real demand you can cover, how much of your topic you have already written, how much review capacity your team has, and how much value is still locked in the posts you have already published. Get those four inputs right and the number falls out; skip them and you will either burn out or post filler.
Short answer
There is no universal number of posts per week, and this article will not invent one. Google does not publish a required publishing rate and does not reward raw volume. What you can do is calculate a cadence from your own two constraints — the number of distinct real questions you can still answer, and the number of posts your reviewers can verify — instead of copying a target. The worked example below shows how the number falls out of those inputs.
Worked example: calculating a cadence from capacity (illustrative)
This is a teaching example with made-up numbers, not a benchmark or a recommendation for your site. Replace every figure with your own.
| Input | Example value | How you get it |
|---|---|---|
| Uncovered questions you can answer with real expertise | 26 | Count distinct buyer questions without a matching page (Search Console + sales/support threads) |
| Posts a reviewer can verify with product knowledge | 3 per week | The slowest necessary step: how many pieces the person who confirms facts can actually check |
| Updates competing for the same review time | 1 of every 3 slots | Share of the calendar you reserve for refreshing existing pages |
If 3 posts could be verified a week but 1 of every 3 slots goes to updates, new pages run at about 2 a week. 26 uncovered questions then cover roughly 13 weeks of new pages — after which the cadence should shift toward updates, not toward publishing filler to hit a number. If your reviewer can only confirm one post a week, the cadence is one a week regardless of how many questions you found.
The output is a schedule that matches your capacity and your remaining real questions. It changes when either input changes: a new product adds questions, a smaller review team lowers the ceiling. That is why a fixed "posts per week" figure is the wrong thing to remember.
The four inputs that set your cadence
1. Demand you can actually cover
Start from the questions your buyers and customers already type. Google Search Console query data, sales and support questions, and on-site search are better sources than a generic keyword list, because they are specific to your business. Count the distinct questions with real search intent that you can answer with genuine expertise. That count is your first ceiling: publishing more posts than you have distinct real questions produces thin pages that repeat each other.
2. Overlap with what you already have
Before scheduling a new post, check whether you already cover the intent. Two pages that answer the same question compete with each other, and neither gets the full signal. A useful rule is to settle the intent map before the calendar: one page per intent, and expand a page rather than write a near-duplicate of it. If a topic is already covered, the work is usually an update, not a new post.
3. Review and production capacity
Every post needs a subject-matter check, not just a spelling check. If your cadence outruns the people who can confirm a specification, a price rule or a technical claim, the extra posts will either contain unverified statements or sit unpublished while a reviewer catches up. Match the schedule to the slowest necessary step — often review, not writing. If one reviewer can handle two posts a week, four posts a week is not four posts of value; it is two verified posts and two risks.
4. The value still sitting in old posts
For most established sites, updating an existing page beats publishing a new one. Old posts lose accuracy, lose rankings as competitors improve, and hold queries you have already earned the right to answer. Because updates are often cheaper than new production and can recover traffic faster, they should be a fixed share of your monthly schedule, not an afterthought. A simple split works well: a set number of new posts for genuinely new intent, and a set number of updates chosen from pages that are declining, outdated or thin.
A scheduling model instead of a magic number
Rather than fix "posts per week", build the month from these buckets:
| Bucket | Decision | Typical trigger |
|---|---|---|
| New pages | Only for intent you do not cover yet | A real question with no matching page |
| Updates | Fix facts, add depth, refresh examples | Traffic or rankings slipping; content is outdated |
| Consolidation | Merge overlapping pages and redirect | Two or more pages chasing the same query |
| Support | Product/landing copy, internal links | New product or a page that needs to be found |
Then size each bucket to the review capacity you have that month. A small team might run one new post and two updates a week and make steady progress; a larger content operation might run more. The steady, reviewable cadence beats a spike followed by silence, because both publishing and indexing reward consistency.
What changes your cadence
- New site vs established site. A young site usually needs to establish topical coverage first, so it may publish more new pages early; an older site with broad coverage should shift weight toward updates.
- How competitive the topic is. Harder topics need deeper, better-supported pages, which take longer to produce. That naturally lowers cadence.
- Whether you can promote it. A post that no one links to, shares or mentions has to earn everything from search alone. If you have no distribution, quality per post matters even more.
- Whether you have subject-matter reviewers. No reviewer means you should publish less and verify more, not publish more and hope.
How to tell if the cadence is working
Do not judge a posting schedule by volume. Watch whether the work is producing durable results:
- Are new pages getting indexed, and how long does it take?
- Are they picking up impressions for the intended queries, or for unrelated ones?
- Are updated pages recovering, holding or continuing to slide?
- Is organic traffic and qualified activity trending up without a matching rise in thin or duplicate pages?
If pages are published but not indexed, or indexed but never shown, the problem is not that you published too few — it is usually quality, intent or a technical issue. Adding volume on top of an unresolved problem just multiplies the problem.
When and how to bring in outside writing help
Outsourcing writing can raise cadence, but only if the brief, the facts and the review process are already under control. The company still owns the product facts, the technical claims and the final approval; a writer organises and expresses them. If you buy articles, prepare product source material, define the buyer questions each piece must answer, and keep a reviewer for accuracy. You can read our English SEO article service for how we scope single pieces and batches, and see how the pieces fit the wider site in our Google SEO service.
Common questions
Do I need to publish every week? No. Consistency helps, but a reliable twice-a-month cadence beats an erratic weekly one you cannot sustain or review.
Should I update old posts or write new ones? Do both, with updates as a fixed share. For most sites the fastest wins are usually in updating pages that already have some visibility.
Is there a posting frequency that Google rewards? Google does not publish a required publishing rate, and search results are not awarded for volume. It rewards helpful, reliable content, however often you can produce it well. See Google's guidance on creating helpful, reliable, people-first content.