Most small firms ask how many posts they need and get told to publish more. The better answer starts with what those posts cover.
The short answer
- A firm with 5 to 20 staff and no search presence needs 20 to 30 posts on one subject, published within 6 months, to see results.
- The first measurable movement usually shows up around month 4, and almost never before month 3.
- Post count is a proxy for coverage. 12 posts that cover one subject completely will outrank 30 posts on 30 unrelated subjects.
- The real target is the list of 25 to 35 questions a buyer asks between first awareness and signing.
- Check progress at month 2, month 4, and month 8 to 12, and decide at month 4 whether to continue.
- A firm with 30 or 40 existing posts usually needs 8 or 10 new ones plus consolidation, not 30 more.
How many blog posts before SEO works
For a firm with 5 to 20 staff and no existing search presence, the working number is 20 to 30 posts covering one subject, published within 6 months, with the first measurable movement around month 4.
That number is close to useless on its own, because post count is not what produces the result. 30 posts on 30 unrelated subjects does nothing. 12 posts that cover one subject completely will outrank it. The count only matters as a proxy for coverage, and coverage is the thing you are actually buying.
So the useful version of the question is how much of a subject you have to cover before a search engine treats you as a source for it. That has an answer, it is checkable at 3 specific points, and it tells you whether to keep spending.
Why post count is the wrong unit
Here is how SEO works in one sentence. A search engine decides whether your site is a credible source on a subject, and it reads that from the shape of what you have published, not the volume.
Think about what a founder searching “AI readiness audit” needs before hiring anyone. What it costs. What happens in it. How long it takes. What they get at the end. What to prepare. What to do with the findings. Whether to do it themselves. That is 7 questions, and a site that answers all 7 looks like a source. A site with 30 posts that answers 2 of them, plus 28 posts about unrelated productivity tips, looks like a blog.
This is why publishing volume advice fails so consistently. “Publish twice a week” is a cadence with no target attached. Two posts a week for a year is 104 posts and can still leave every commercial question half-answered, because nobody was tracking which questions were covered.
Count the questions, not the posts. Write down every question a buyer asks between first awareness and signing. For most service businesses that list runs to 25 or 35 items. That list is your real target, and it does not grow as you publish.
The number for a small firm with no existing traffic
Here is what the first 6 months should contain for a 5 to 20 person firm starting from zero.
One pillar post covering the whole subject. Long, structured, answering the subject at the level someone would want before hiring. This is the page everything else links to.
12 to 20 posts answering one buyer question each. One question, one page, a title that matches how the question gets typed. These are what actually rank, because they match specific searches with low competition.
3 to 5 posts on the adjacent decision. The comparison, the alternative, the cost. These catch people further along and convert better than anything else on the site.
That is 16 to 26 posts. At one per week it is 6 months. At two per week it is 3 months, and the second pace is only better if quality holds, which for most small teams it does not.
The sequence matters more than the pace. Publish the 12 question-answering posts before the pillar if you have to, but link them all together as you go. Posts that never link to each other do not build anything, which is the failure mode covered in the internal linking system.
How long before the blog posts start working
Three checkpoints, and each one has a decision attached.
Month 2, after roughly 8 posts. You should see impressions in Search Console for your target subject, with average positions somewhere between 20 and 60. Zero impressions at month 2 means the pages are not being indexed or the subject has no search demand. Both are worth 20 minutes to check, and both are fatal if ignored for another 4 months.
Month 4, after roughly 16 posts. Average position on your best 5 pages should have moved into the teens, and 1 or 2 of the long-tail posts should have real clicks. This is where honest programs and doomed ones separate. Movement in the right direction, even small, means continue. Flat impressions and flat positions mean the coverage is too thin or the pages are too similar to what already ranks.
Month 8 to 12. Top 10 positions on several of the question posts, and an inquiry that you can trace to an organic search. One inquiry is the signal. It is not a revenue result yet, and treating it as one is how people talk themselves into another year of publishing with nothing attached.
| Checkpoint | Posts published | What you should see | Decision |
|---|---|---|---|
| Month 2 | About 8 | Impressions on the target subject, positions 20 to 60 | Zero impressions means check indexing and search demand now |
| Month 4 | About 16 | Best 5 pages in the teens, 1 or 2 long-tail posts with clicks | Any movement means continue, flat means coverage is thin |
| Month 8 to 12 | 20 to 30 | Top 10 on several question posts, 1 traceable organic inquiry | The inquiry is the signal, not yet a revenue result |
Movement almost never starts before month 3. A page that ranks in week 2 is ranking for something nobody searches. Anyone promising results in 60 days is selling either a paid channel or a page that will not hold.
What makes the number smaller
Three things move the number down, and they are worth more than doubling your publishing rate.
Existing domain history is the largest. A site that has been live for 8 years with real backlinks may see movement at 8 posts where a 6-month-old domain needs 25. This is not fair and cannot be bought quickly.
Narrowness is the one you control. “AI for small business” is a subject with a thousand competing sites. “AI readiness audits for professional services firms” is a subject with maybe 12, and you can cover it completely in 15 posts. Picking the narrower subject changes the required post count more than anything else you decide, and it gets decided once, at the start, usually in about 5 minutes.
Genuine first-hand material is the third. A post containing something you learned doing the work, with real numbers from a real engagement, will outperform 5 posts assembled from other people’s articles. This is also the part that AI-assisted writing is worst at supplying on its own, which is why the input matters more than the drafting speed. A content brief that carries your actual experience into the draft is the difference between 20 usable posts and 20 that read like everyone else’s.
What makes the number bigger
Publishing without a subject. 40 posts spread over 12 loosely related topics builds nothing, and this is the most common pattern in small business blogging by a wide margin.
Rewriting the same post repeatedly. Four posts that all argue AI saves time compete with each other and none of them wins. If two posts would have the same title in plain language, they should be one post.
Thin pages published to hit a number. A 600-word post that restates the first result on the page adds nothing to your coverage and dilutes the average quality of the site.
No internal links. An unlinked post is a page nobody, including a crawler, has a reason to reach.
If your program has been running 9 months with no movement, the cause is almost always in this list rather than in the post count. Running an SEO content audit will name which of the 4 it is in about 2 hours.
Counting the blog posts you already have
Most firms asking this question are not starting from zero. They have 30 or 40 posts published over 3 years by whoever had time, and nothing to show for it. The instinct is to add more. The correct first move is to count what the existing posts cover.
Take your list of 25 buyer questions and put each existing post against one of them. In every audit I have run, the result looks the same: 6 or 7 questions have 3 posts each, roughly half have none, and a third of the posts do not answer a buyer question at all.
That gives you a real backlog instead of a publishing schedule. The questions with nothing against them are the posts to write. The questions with 3 competing posts get consolidated into one, with the other 2 redirected into it. The posts that map to no question get left alone or removed, and either is fine.
A firm doing this usually finds they need 8 or 10 new posts rather than 30, and that the consolidation moves rankings faster than the new writing does. Consolidating 3 thin posts into 1 complete one is the cheapest ranking gain available to a site that has been publishing for years without a plan.
Decide at month 4 rather than month 12
Set the review before you start, because a program with no review date runs until the person doing it leaves.
At month 4 you are asking one question: did the impressions and positions move at all. Not whether there is revenue. Not whether it feels worthwhile. Movement is the leading indicator, and it appears long before money does.
Movement means continue to month 12 and expect results. No movement after 16 competent posts means the subject is wrong, the site has a technical problem, or the competition is far stronger than the plan assumed. Publishing 20 more posts into that will not fix it, and it is the most expensive mistake available in content marketing because the cost is 12 months of someone’s time rather than a line on an invoice.
20 to 30 posts on one subject in 6 months, checked at month 2, month 4 and month 12. That is the whole answer. The part that decides whether it works is choosing a subject narrow enough that 25 posts covers it, and then building the production system that gets those 25 posts published without depending on one person having a good month. The system is covered in AI-enabled SEO operations, and it matters more than the writing does.
Common questions
How many blog posts do I need to rank on Google?
For a small firm starting from zero, plan on 20 to 30 posts covering one subject over 6 months. The count matters less than whether those posts answer the questions your buyers actually ask.
How often should a small business publish blog posts?
How often to post a blog for SEO comes down to what the team can sustain. One post a week gets you to 25 posts in about 6 months. Two a week is only better if the quality holds, and for most small teams it does not. Pick a pace the team can keep up in a busy month, not just a quiet one.
How long does it take for a blog post to show up in Google?
Expect impressions in Search Console by month 2 and real movement around month 4. A page that ranks in week 2 is usually ranking for something nobody searches.
Should I keep blogging if I am not getting any traffic?
Check at month 4 whether impressions and positions moved at all. If they did, keep going to month 12. If nothing moved after 16 competent posts, the subject, the site, or the competition is the problem, and more posts will not fix it.
Is it better to update old blog posts or write new ones?
If you already have 30 or 40 posts, map them against your buyer questions first. Merging 3 thin posts into 1 complete one often moves rankings faster than new writing. Clean the garage before you buy more boxes.
