How Many Blog Posts Per Month Do You Actually Need?

Summary

• There is no frequency that ranks. Google’s John Mueller put it plainly in 2022: “No, more content definitely does not mean more traffic.”
• Four to eight posts a month is a realistic working range for a service business, and the right number inside that range depends on how many clusters you are trying to own.
• Count clusters, not posts. One money page plus six to twelve supporting posts that answer every pre-purchase question is a cluster. Finish one before starting the next.
• Google explicitly warns against volume for its own sake, asking whether you are “producing lots of content on many different topics in hopes that some of it might perform well.”
• Sometimes the right number is zero new posts. If two of your pages target the same query, merging them is worth more than publishing a ninth.

Table of Contents

• How many blog posts per month do you actually need?
• Why publishing frequency is not a ranking factor
• What actually determines how fast you see results
• The cluster math: calculating your real number
• When fewer posts is the correct answer
• Common mistakes
• Frequently asked questions

How many blog posts per month do you actually need?

For a Canadian service business starting from a thin site, four to eight posts per month for six to nine months is enough to own one or two topic clusters. Below two posts a month you cannot finish a cluster inside a year. Above twelve, quality control becomes the constraint and you start competing with yourself. The range is wide because the correct number is derived from your target, not chosen from a benchmark.

The honest answer by goal

Notice the last row. It is the most common real situation and the one most content retainers get wrong, because a monthly post count is easier to sell than an audit. If you are hiring help, the first question worth asking a content production partner is not how many posts they will write. It is what they will consolidate.

Why publishing frequency is not a ranking factor

Google has never published a posting cadence, and its own guidance points the other way. There is no mechanism by which “four posts a month” is better than “three” in the ranking systems. What there is: a quality assessment applied per page, and a topical assessment applied across your site.

In February 2022, asked whether publishing more quality pages per day would help, Google’s John Mueller answered: “Not necessarily. Does writing 4 ‘good quality’ pages per day make a book a best-seller?” The following day he was more direct: “No, more content definitely does not mean more traffic.” Those are not ambiguous statements, and nothing in Google’s documentation since has contradicted them.

Google’s own guidance on creating helpful content goes further and treats volume as a warning sign. Among the questions it tells publishers to ask themselves is whether they are “producing lots of content on many different topics in hopes that some of it might perform well.” It also says plainly that there is no preferred word count, and warns against adding or removing large amounts of content purely to look fresh. Taken together: the system is not counting your posts.

So why does cadence still show up in every content proposal?

• It is billable and measurable. Eight posts is a deliverable. “Finished the cluster” is a judgement call.
• It correlates with the real driver. Sites that publish consistently tend to cover more ground, and coverage does matter. The cadence gets the credit for the coverage’s work.
• It creates the appearance of progress during the six to nine months before rankings move, which is a real client-management problem with a bad solution.

If you want the honest timeline rather than the cadence, we laid it out in how long SEO actually takes in Canada. The short version: the first three months buy coverage, months four to nine buy rankings, and nothing you do with frequency compresses that much.

What actually determines how fast you see results

Four things, in order of leverage. None of them is a number of posts.

1. Cluster completeness

A cluster is one commercial page plus every post a buyer needs before they contact you. For a web design service that is cost, timeline, process, ownership, what to ask before hiring, what goes wrong, and the comparison against the alternative. Seven posts. When all seven exist and link to each other and to the service page, the service page starts ranking. When four of seven exist, nothing much happens, no matter how often you published them.

2. Query coverage, not post count

One thorough post can rank for dozens of long-tail variants. Three thin posts split across the same intent rank for none of them well, and may actively compete. Count the distinct questions you have answered completely, not the files in your CMS. This is also the unit AI answer engines work in: they cite the page that answers the question directly, not the site that published most often.

3. Internal link structure

A post that nothing links to is a post that does almost nothing for the site. Every new piece should link up to its money page, sideways to at least two siblings in the same cluster, and across to one post in a different cluster. Without that graph, a hundred posts behave like a hundred separate pages, which is exactly what we found when auditing a site with more than two hundred and fifty of them. The mechanics are in internal linking for topical authority.

4. Whether the post is better than what already ranks

This is the brutal one. If the current top result answers the query more completely than your draft does, publishing your draft changes nothing. The only honest test before you write: open the top five results, list what they miss, and make sure your piece supplies it. If you cannot name what you are adding, do not publish. That single filter will cut a proposed cadence in half and improve the outcome.

The cluster math: calculating your real number

Stop choosing a cadence. Derive it. Five steps, and you can do this in an afternoon.

1. List your money pages. The pages that sell something. Usually three to eight for a service business.
2. Pick one. The service with the best margin and the clearest buyer, not the one with the most search volume.
3. Write down every question a buyer asks before hiring. Cost, timeline, process, risk, ownership, how to choose, what goes wrong, the alternative, the local or regulatory angle. Use the words they use. Pull them from your own sales calls and from the forums your buyers actually post in.
4. Subtract what you already cover well. Not “have a post about.” Cover well, meaning it would survive a comparison against the current top result.
5. Divide by your deadline. Eight missing posts in four months is two a month. Eight in two months is four. That is your cadence, and it expires when the cluster is done.

A worked example

A Calgary contractor sells two things: renovations and new builds. The renovation buyer asks nine questions before calling: what it costs, how long it takes, do I need permits, how do I know the quote is real, what happens if it goes over, who is liable, do you handle design, what about financing, and how do I check you are legitimate. Four of those are already answered on the site, two of them thinly. That leaves five new posts and two rewrites, seven pieces. Targeting a spring season four months out gives a cadence of roughly two a month. Not six. Not eight. Two, with a defined finish line.

Once that cluster is finished, the number resets for the next one. This is why a flat monthly retainer measured in posts is usually mispriced in both directions: too much work in the quiet phase, not enough in the build phase. A better structure is a content system with a defined cluster target, reviewed quarterly.

When fewer posts is the correct answer

If your site already has a hundred or more posts and traffic is flat, more posts will not fix it. Three things will, and all of them reduce your post count.

• Merge the pages competing with each other. Two posts targeting one query split their own signals. Keep the stronger URL, fold the better parts of the weaker one into it, and redirect. Finding these is mechanical, and the process for finding and fixing cannibalisation is worth running before you commission anything new.
• Refresh what is already ranking on page two. A post sitting at position eleven to fifteen with impressions already has demand proven. Improving it moves revenue faster than a new post starting at zero. The method is in how often to update old blog posts.
• Repurpose instead of writing new. One finished cluster post has a month of social, email and video inside it. That is distribution, not production, and it is usually the cheaper half of the system. We broke it down in turning one piece of content into a month of marketing.

There is a hard version of this worth saying out loud: if you are publishing weekly and getting no leads, the problem is almost never cadence. It is usually that the posts answer questions nobody buying from you is asking, or that the site does not convert the traffic it already gets. That second failure has its own diagnostic in why your website gets quote requests that never close.

Common mistakes

• Choosing a cadence before a cluster map. This guarantees scattered coverage. The map produces the number.
• Publishing on topics your buyer does not research. Industry news, company updates and trend roundups almost never sit in a buying path.
• Measuring output instead of coverage. “We published 24 posts” is not a result. “We now rank for every question in the renovation buying path” is.
• Ignoring the existing library. Most sites have more unrealised value in the posts they already published than in the ones they have not written.
• Thin posts to hit a number. A post you would not send to a prospect should not be on your site. It dilutes the cluster and competes with your better page.
• No internal links at publish time. Adding them “later” means never. It takes four minutes at publish and it is the difference between a page and an asset.
• Treating AI drafting as a volume multiplier. The constraint was never typing speed. It is whether the piece is better than what ranks, and editorial judgement does not scale with word count. Our view on where the line sits is in does Google penalise AI-generated content.

Pros and cons of a high publishing cadence

Where high cadence genuinely helps

• It finishes clusters faster, and a finished cluster is the thing that ranks.
• It produces more raw material for email, social and video, which is where most of the near-term revenue actually comes from.
• It shortens the feedback loop. More published pages means more Search Console data, sooner.
• In genuinely competitive niches, depth is table stakes and you need volume to reach it.

Where it backfires

• Quality control fails first. Editorial attention, not writing capacity, is the real ceiling.
• Overlapping posts start cannibalising each other, which costs you rankings you already had.
• The library becomes too large to maintain, and stale pages lose trust and citations.
• It hides the real problem when the issue is conversion or offer clarity rather than traffic.

Frequently asked questions

How many blog posts per month is good for SEO?

Four to eight is a reasonable working range for a service business, but the number should come from arithmetic rather than a benchmark. Count the buyer questions in one cluster that you have not covered well, divide by the months until your deadline, and that is your cadence. Google does not reward frequency, so a cadence with no cluster target behind it is just a production schedule.

Is publishing frequency a Google ranking factor?

No. Google has never named one, and its guidance warns against publishing volume for its own sake, asking whether you are producing lots of content on many topics hoping some of it performs. In 2022 John Mueller stated directly that more content does not mean more traffic. Frequency correlates with coverage, and coverage is what matters.

Is one blog post a week enough?

Usually yes, if every post belongs to the same cluster and links into it. Four focused posts a month finishes a seven-post cluster in under two months. One post a week across four unrelated topics finishes nothing, and that is the version most businesses are actually running.

How many posts before I see results?

Think in clusters, not counts. A money page tends to move once the supporting posts around it cover the full buying path, commonly six to twelve pieces, and once those pages have been indexed and linked for a few months. Expect three months before anything meaningful and six to nine before the service page itself competes.

Should I publish less and write longer posts?

Only if length is what the query needs. Google states it has no preferred word count. Match the depth to the intent: a pricing question needs numbers and ranges, a process question needs steps, a comparison needs a table. Padding a post to hit a length target makes it worse, not more rankable.

Does posting consistently matter at all?

It matters operationally more than algorithmically. A consistent schedule is how clusters actually get finished and how a team keeps editorial standards. A site that publishes six posts in one week and nothing for five months usually has an abandoned cluster, and that is the real cost.

What if I have no time to publish at all?

Then publish nothing and fix something else. For a local service business, an optimised Google Business Profile, reviews and a site that converts are worth more than a stalled blog. Content earns its place when you have the capacity to finish a cluster, not before.

The bottom line

Pick the service you most want to sell. Write down every question a buyer asks before they hire. Subtract what you already cover properly. Divide by the months until you need results. That is your posts per month, and it changes when the cluster is finished. Anyone selling you a flat monthly post count without doing that arithmetic first is selling you a production schedule, not a result.

Want the cluster map before you commit to a cadence?

We start with what your buyers search before they hire, map it against what you already rank for, and tell you the real number of posts that gets you there, including the ones you should merge rather than write. Start with the intake form and tell us which service you want to own. You can also see how we structure content production engagements.