How many words does a page actually need to rank?

The word count myth that won’t die

I still get asked this once a week: “what’s the ideal word count for SEO?” There isn’t one. I run a portfolio of sites across a handful of niches, and the pages that rank best on some of them are under 600 words. On others, the top page is 4,000 words because the topic genuinely needs that much explaining. Content length seo advice that gives you a flat number, like “always write 1,500+ words,” is a rule of thumb dressed up as a ranking factor, and it’s been dead for a while.

The number itself was never the thing search engines cared about. It was a rough stand-in for something else: does this page actually cover what the searcher needs. Somewhere along the way, tools and templates turned the stand-in into the target, and people started padding pages to hit a count instead of answering the question.

Why word count became a proxy in the first place

Longer pages used to correlate with better rankings because, on average, a page that answers a question thoroughly needs more words than a thin page that doesn’t. If you’re comparing a 300-word page that lists three bullet points against a 1,800-word page that walks through the process, compares options, and answers follow-up questions, the second one is probably more useful, and it’s also longer. The correlation is real. The mistake is treating it as causal.

Once “long content ranks better” became conventional wisdom, a lot of writers started hitting word counts by restating the same point three different ways, adding a “brief history of X” section nobody asked for, or answering questions nobody was actually searching. That doesn’t help a reader, and it doesn’t fool anything that’s evaluating whether the page satisfies the query. It just makes the page slower to read and more expensive to write.

What the query actually needs, not the number

The better question to ask before writing anything is: what does someone searching this term need to walk away with? That answer sets the length, not the other way around.

A page targeting “what time zone is Tokyo in” needs a sentence and maybe a small table for daylight saving edge cases. Padding that to 1,200 words to “compete” with a competitor’s overwritten article just buries the answer the visitor came for, which hurts your bounce rate and your conversion, whether or not it hurts your ranking.

A page targeting “how to set up a Cloudflare Worker for image resizing” needs prerequisites, a step by step walkthrough, code, common errors, and probably a note on cost. That’s naturally going to run long, not because you’re chasing a count but because each of those sections is doing real work for someone trying to get unstuck.

I look at length as an output of scope, not an input. Define what the page has to cover to be the best answer available for that specific query, write that, and stop. If it’s short, it’s short.

Look at what’s already ranking, but read it, don’t count it

Checking the current top results for your target term is still useful, but not for their word counts. Read them and ask what they cover that you’d otherwise miss, what they get wrong or leave vague, and what format they use, whether that’s a table, a video embed, a numbered list, or a wall of prose. If every top result answers a comparison question with a table and you’re planning a page of paragraphs, that’s a signal about the format the searcher wants, not about hitting a specific length.

I’ve had pages beat a competitor’s 2,500-word guide with 700 words because the competitor’s guide buried the actual steps under three paragraphs of “why this matters” preamble. Nobody searching a how-to query wants preamble. They want the steps.

When longer genuinely helps

There are real cases where more length matters, but for reasons that have nothing to do with hitting a target:

The topic has multiple sub-questions. A buyer’s guide for, say, proxy providers naturally needs pricing, use cases, limitations, and alternatives. Cutting any of those to save words just means the reader bounces to a competitor’s page that covers it.
You’re differentiating on depth. If the existing top results are shallow, a page that actually explains the mechanics, not just the steps, can earn links and repeat visits that a thin page won’t. That’s a content length seo strategy that works because it changes user behavior, not because a crawler is counting words.
The format demands it. Comparison tables, FAQs answering real long-tail variants, and code samples all add length as a side effect of being genuinely useful, not as padding.

When longer hurts

You’re restating the same point to look thorough. Readers notice repetition faster than any algorithm does, and they leave.
You’re answering questions nobody asked to hit a target. A “history of X” section on a transactional page is dead weight.
The extra length delays the answer. If someone has to scroll past 400 words of context before finding what they searched for, you’ve made the page worse, not better.
You’re writing for a keyword density target. That died over a decade ago. Repeating a phrase to hit a percentage reads badly and doesn’t help.

I’ve cut pages from 1,800 words down to 900 and watched rankings hold steady or improve, because what came out was the filler, not the substance.

The process I actually use

Before I write a page now, I do three things, in this order:

Write down what the searcher needs to leave with. Not a keyword, an outcome. “Knows which of three tools fits their budget” is different from “knows the term exists.”
List the sub-questions that outcome implies. For a comparison page, that’s price, limitations, setup time, and who it’s not for. For a definitional page, it might just be one clear paragraph and an example.
Write to cover that list, then stop. I don’t check the word count until after the draft is done, and when I do check it, it’s just a sanity check, not a target. If it’s 400 words and covers everything on the list, it ships at 400 words.

This also means two pages targeting similar-looking keywords on the same site can have wildly different lengths, and that’s fine. A “what is X” page and a “how to do X” page are different jobs even if the keyword tool groups them together.

Bottom line

Content length seo advice that hands you a number is optimizing for the wrong variable. Length is a byproduct of how much a topic actually requires to be answered well, not a lever you pull to rank. I’ve eaten the cost of writing to a template before, padding pages that didn’t need it, and rewriting them shorter later. The pages that have held up over time are the ones sized to the question, not to a word count target pulled from a tool.

Figure out what the page needs to do for the person reading it. Write until that job is done. Stop there, whether that’s 400 words or 4,000.

If you want more breakdowns like this on keyword research, technical SEO, and link building without the recycled advice, check out more from us at The SEO Desk.