How long should a blog post be? What Google says about blog post length, and how to decide
By Thomas Lovas Lokoy · Published 16 October 2026 · 6 min read
Ask how long a blog post should be and you will find confident answers, usually a single number of words, usually attributed to a study of what ranks. Google's own answer is shorter. Its guidance on helpful content asks whether you are writing to a particular word count because you have heard that Google has a preferred one, and answers: "(No, we don't.)" Blog post length is therefore an editorial decision, not a ranking setting. This guide explains what Google does say, what actually determines the right length for a given post, and how to tell when a draft is too long or too short.
What Google says about blog post length
Three Google documents cover it, and they agree.
The helpful content guidance lists questions for assessing your own content. One of them is the word-count question above, placed among the warning signs of content made for search engines rather than people. The questions it wants you to answer yes to are about substance: whether the content provides original information, reporting, research or analysis; whether it gives a substantial, complete or comprehensive description of the topic; and whether someone will leave "feeling they've learned enough about a topic to help achieve their goal".
The SEO starter guide lists content length among the things it does not think you should focus on. It says "the length of the content alone doesn't matter for ranking purposes", and that there is no magical word count target, minimum or maximum, "though you probably want to have at least one word". It makes the same point about structure: there is no ideal number of headings either.
The search quality rater guidelines, the document Google gives the people who evaluate search results (the edition opened for this article is dated 11 September 2025), do not set a length either. What they describe is the opposite problem: filler. More on that below.
So the honest answer to "how long?" is that Google has published no number, and has said explicitly that it has no preferred one. Any figure presented as Google's requirement is not.
Why longer can look like it works
If length does not matter, why do so many people believe it does? A likely explanation is that thorough pages tend to be long, and thorough pages tend to satisfy readers. Length travels with the thing that matters without being that thing. Google's guidance asks whether content provides "a substantial, complete, or comprehensive description of the topic", and covering a broad topic completely takes words. Covering a narrow one completely does not.
The mistake is reversing the direction: adding words in the hope of acquiring thoroughness. That produces exactly what the rater guidelines warn against.
Filler: the real risk of writing to a word count
The rater guidelines describe filler as "low-effort content that adds little value and doesn't directly support the purpose of the page". They add that "Filler can artificially inflate content, creating a page that appears rich but lacks content website visitors find valuable", and that it can result in a poor experience, especially when placed ahead of the helpful content. They also say the most helpful main content should be placed near the top of the page so visitors can reach it immediately, using recipe pages as the example.
Every familiar padding technique fits that description: the introduction that restates the question, the history section nobody asked for, the paragraph that summarises the previous paragraph, the conclusion that repeats the article. A word-count target makes all of these tempting, because they are the cheapest words to produce.
At scale, padding becomes a policy matter. Google's spam policies define scaled content abuse as generating many pages "for the primary purpose of manipulating search rankings and not helping users", and list among the examples using generative AI tools to generate many pages without adding value for users. Length is not the offence there, but a process that inflates articles to a quota with whatever text is to hand is heading in that direction.
How to decide the right length for a blog post
Since there is no number, use a procedure. It replaces "how long should it be?" with questions you can answer.
- Write down the reader's question in one sentence. "How do I bleed a radiator?" and "How should a small charity choose accounting software?" are both blog posts, and they need very different amounts of text.
- List what the reader needs to act on the answer. For the radiator: the tools, the steps, how to tell it is done, what to do if water keeps coming. For the software: the criteria, how they trade off, the questions to ask a vendor, the mistakes to avoid. This list is your outline, and its size sets the length.
- Check what is already in the results. Read the pages that currently rank for the question. Google's guidance asks whether your content provides "substantial value when compared to other pages in search results". If they all cover the same five points, covering those five points again at greater length adds nothing; the length you need is whatever it takes to cover the point they missed.
- Account for the stage the reader is at. Someone just discovering a problem wants orientation; someone comparing options wants detail. The guide to content funnel stages for blog articles covers how that changes what a post should contain.
- Write it, then cut. Delete every sentence that only introduces the next one, every restatement, every section that exists for completeness rather than for this reader. What is left is the right length.
If you are briefing a writer or a writing tool, brief the outline rather than a number. A brief that says "2,000 words" invites padding; a brief that lists the six things the reader needs gets six things. The guide on how to brief an AI writing tool goes through what to put in instead.
Signs a draft is too long
- The answer to the reader's question appears after the halfway point.
- Removing the introduction makes the article better.
- Two sections could swap places without anyone noticing, because neither depends on the other.
- The conclusion repeats points rather than telling the reader what to do next.
- You cannot say which reader a section is for.
Signs a draft is too short
Short is not a virtue either. Google's questions include whether a reader leaves feeling they have learned enough to achieve their goal, and whether the content leaves readers "feeling like they need to search again to get better information from other sources". A draft is too short when:
- A reader following it would get stuck at a step it skipped.
- It names options without saying how to choose between them.
- It asserts something contested without showing why.
- It answers the headline question but not the obvious follow-up.
The fix for a short draft is a missing section, not longer sentences.
Length and quality checks
Length interacts with verification in one practical way: every additional claim is something else to check. A padded article contains more unverified statements than a tight one, and more chances to be wrong. If you are publishing AI-assisted drafts, the fact-checking routine is quicker on an article that says what it needs to and stops.