Long-Tail Keywords for SEO: What They Are and How to Find Them

A broad keyword like "running shoes" gets searched a lot, but many sites are competing for it. Big retailers and established sites already rank for it, and the search itself doesn't tell you much about what the person actually wants. A long-tail keyword like "best running shoes for flat feet marathon training" is much more specific. It may have fewer searches than "running shoes," and it may be easier to compete for, but you still need to check the results before assuming it will be easy to rank for.

What Are Long-Tail Keywords?

There is no single industry-wide cutoff for what counts as a long-tail keyword. A long-tail keyword generally refers to a search query with relatively low search demand compared with broader keywords about the same topic. It's often more specific, but word count alone does not define it.

A long-tail keyword isn't just another name for a low-volume keyword. A low-volume keyword is defined by its measured search demand. A long-tail keyword is described in relation to broader terms on the same topic and is usually more specific. The two overlap, but they aren't the same thing.

Type	Example	Typical intent
Short-tail	running shoes	Broad
Mid-tail	running shoes for women	More specific
Long-tail	best running shoes for women with flat feet	Highly specific

These labels are useful shorthand, not strict industry categories.

The name comes from the shape of this search-demand curve: a small number of keywords get very large numbers of searches, while a much larger number get relatively few searches each. Those lower-volume queries make up the "long tail."

Why People Target Long-Tail Keywords for SEO

There are a few practical reasons to target them:

• Often lower competition. Specific queries may have fewer competing pages, although this varies by topic. A long-tail keyword can still be hard to rank for when many strong pages already target the same search -- it's not automatically easy.
• More specific intent. The extra words in a long-tail query usually narrow down what the person wants, which makes it easier to write a page that actually matches what they're looking for.
• There are a very large number of low-volume search queries out there. A site can get traffic from many smaller searches instead of depending on one popular keyword.

Long-Tail Keywords Aren't Always Low Competition

It's easy to read "often lower competition" above and assume long-tail automatically means easy. It doesn't. Specificity, search volume, and competition measure three different things. A keyword can be high or low on each of these measures.

Query	Specificity	Search volume	Competition
SEO	Low	High	High
SEO tools	Medium	Medium	High
SEO tools for SaaS	High	Lower	Medium
SEO tools for early-stage SaaS	Very high	Very low	Could still be high

A highly specific query can still be difficult to rank for, particularly when it has real commercial value, like something someone searches right before buying. A broader query can sometimes have surprisingly weak competition. Check what's actually ranking before assuming a long-tail phrase will be easy just because it's specific.

The Different Types of Long-Tail Keywords

Long-tail keywords can take a few different forms. These aren't official categories; they're useful ways to think about the forms long-tail searches can take:

• Supporting long-tail keywords -- a specific variation of a broader topic. "running shoes" becomes "running shoes for flat feet."
• Topical long-tail keywords -- lower-volume queries that still represent the broader topic, just from a narrower angle. "best shoes for overpronation" is still about running shoes.
• Conversational, highly specific queries -- closer to a full sentence than a phrase. "what are the best running shoes for flat feet and knee pain?"

One topic can also produce many specific searches. "Google AI Overviews," for example: how they work, whether they reduce clicks, how to optimize for them, why a given site isn't appearing in one, how they compare to featured snippets. Long-tail research isn't always about finding one keyword to target. It can also reveal the range of questions people actually ask around a topic, which is useful when deciding what a page needs to cover, not just what to title it.

Should You Target Zero-Volume Keywords?

A keyword tool reporting "0 searches/month" doesn't necessarily mean nobody searches for that query.

Search-volume figures from keyword tools are estimates, not direct counts of every search. Different tools can also report different numbers for the same query. They can miss real demand for a few reasons:

• Fragmented demand. Thousands of closely related phrasings can each get a handful of searches, none large enough on its own to register.
• Emerging terminology. A new term doesn't have enough historical search data yet for a tool to report on.
• Seasonality. Demand can concentrate into a narrow window that a tool's averaging smooths over.
• Different grouping. Keyword databases can bucket closely related queries together in ways that hide one specific phrasing.

Don't confuse missing data with missing demand. This matters most for new, highly specific, seasonal, or fragmented queries -- a new product, feature or event can generate real searches (a "[product] pricing" or "[product] vs [competitor]" query, for example) before keyword tools have enough historical data to report meaningful volume.

That doesn't mean every zero-volume keyword is worth covering -- that would be its own kind of overreaction. Before spending time on one, check for a few signals: does it closely match a real problem your audience has, are there already search results for it, is the topic genuinely emerging rather than just obscure, and do closely related queries show broader demand? A zero-volume query that checks a few of those is worth a look. One that doesn't probably isn't.

How to Find (and Identify) Long-Tail Keywords

The full keyword research process is covered in the Keyword Research guide. Here, the focus is just on where to find and identify long-tail variations once you already have a seed term:

• Autocomplete. Start typing your seed keyword into Google and look at the suggestions. Google's own explanation of autocomplete predictions says these can reflect searches people have made, along with factors such as location, language, trends and previous searches.
• People Also Ask and related searches. Look at the questions and related searches that appear around your topic. They can reveal related questions and different ways people search for the same subject.
• Forums and communities (Reddit threads, niche forums, Q&A sites) in your category. People ask questions there in their own words, which tends to closely match how they'd phrase a search.
• Your own support tickets, sales calls and customer questions. Repeated questions can reveal topics worth researching. Check whether people are actually searching for those questions before treating them as SEO opportunities -- a repeated support question can just as easily point to a confusing product or a documentation gap as to a real search demand. A customer asking whether you have a tool that shows how their brand shows up when people ask ChatGPT for recommendations isn't a traditional keyword-tool discovery, but it's a real, specific search topic worth checking for.
• A keyword tool's "questions" or "phrase match" filters can help you find variations faster than checking the sources above one at a time. Use the reported search volume as an estimate, not an exact count -- estimates vary between tools.

Free ways to find long-tail keywords: most of the sources above are free to use. Google Search Console shows the long-tail searches your site already appears for, Google Trends shows whether relative search interest in a phrase is growing or falling, and Google Keyword Planner gives estimates of monthly searches, although you need a Google Ads account with billing set up to use it. An AI assistant like ChatGPT can work as a long-tail keyword generator if you give it a seed term and your audience, but it can't tell you whether anyone searches those phrases, so check them in the sources above. More free tools are in Keyword Research Tools: Free and Paid.

Use the Language Your Audience Actually Uses

You can understand a problem your audience has without using the words they'd actually use to describe it. That gap is worth closing, because it's often where long-tail search demand lives.

For example, a company might describe its own category as "AI SEO." Someone with the same underlying need might search "how do I get my company mentioned in ChatGPT," "why doesn't ChatGPT recommend my business," or "how to get my website into AI answers." These may have the same intent. The difference is how people describe the problem.

You can find this language in the sources above -- support tickets, sales calls, community discussions, reviews, comments, internal site search -- plus survey responses and chatbot conversations, if you have them. This can also improve your titles, headings, FAQs and examples, using the actual words your audience uses instead of forcing an awkward exact-match phrase into the copy for its own sake.

A Worked Example

Take the seed keyword "running shoes" again. Broad keyword: "running shoes" -- huge volume, huge competition, vague intent. Some narrower searches might be "trail running shoes" or "marathon running shoes." More specific examples include "best running shoes for flat feet marathon training," "running shoes for overpronation on pavement," and "lightweight trail running shoes for wide feet" -- these are more specific long-tail queries.

These phrases are much narrower than "running shoes," and they may have far fewer searches -- the actual search volume needs to be checked in a keyword tool. But someone searching one of them has given Google (or an AI assistant) more information about what they want, which makes the search intent easier to interpret. A page that directly answers the query can be a better match for that intent -- whether it actually ranks still depends on the competition, the quality of the page, the site's overall strength, and other factors.

Long-Tail vs. Short-Tail Keywords

Short-tail keywords are broad searches like "running shoes." Long-tail keywords are narrower searches like "best running shoes for flat feet marathon training." Neither is better on its own; they do different jobs.

	Short-tail	Long-tail
Example	running shoes	best running shoes for flat feet marathon training
Search volume	Higher	Lower per keyword
What the searcher wants	Often unclear	Usually clearer
Competition	Usually high	Often lower, but check
Best fit	Category and main topic pages	Specific guides, product details, FAQs

Most sites need both: a few pages built around broad topics, and many specific pages or sections that answer long-tail searches. The table at the top of this guide shows where mid-tail keywords sit in between.

Long-Tail Keyword Examples

The running shoes example runs through this guide. Here's how the same idea looks in other kinds of sites:

Seed keyword	Long-tail keyword	What it tells you
running shoes	best running shoes for flat feet marathon training	A runner with a specific foot problem and a specific goal
plumber	emergency plumber open on sunday near me	Someone with an urgent local problem, ready to call
pasta recipe	vegan pasta recipe without nuts	A cook with two clear limits on ingredients
project management software	project management software for small construction teams	A buyer comparing tools for one type of team
SEO tools	SEO tools for early-stage SaaS	Same example as the competition table above

Each long-tail keyword adds details: who the person is, where they are, what they need or what they want to avoid. Those details are what make the intent clearer. Check each one's real volume and the current results before you build a page for it.

How to Use Long-Tail Keywords

Finding a long-tail keyword doesn't mean it deserves a page. Before you use one, check five things: does the intent match what your page offers, is it your audience, are the pages already ranking ones you could realistically compete with, does it have value for your business, and do you already answer it somewhere else? Low volume alone isn't a reason to skip a specific query that matches a real problem your audience has. The full check is in Should I Target This Keyword?

Once you've picked them:

• Group related long-tail keywords that share one intent and answer them on one page, instead of one page per phrase.
• Use the most natural phrasing in a heading or a question, where it reads well.
• Answer the specific question early in the section, then add detail.
• Turn closely related questions into an FAQ on that page.
• Link the specific page to your broader page on the topic, and back, so both are easy to find.

Where to place keywords on a page is covered in On-Page SEO, and how many related pages build one topic is covered in Topical Authority.

Find New Content Opportunities in Your Existing Long-Tail Queries

Search Console can reveal another kind of opportunity: questions your existing page appears for but doesn't fully answer. This is different from a competitor keyword gap, and different from keyword cannibalization (covered in a separate guide, since that's specifically about two of your own pages competing for the same query). Search Console doesn't show every query (some are hidden for privacy), so treat the report as a useful sample of the searches your pages appear for, not a complete list.

Say you have a page targeting "keyword research." Search Console can start showing impressions for more specific queries around that topic: "keyword research for SaaS," "keyword research for AI search," "keyword research without paid tools," "how to find low-volume keywords." Those impressions are a signal, not a to-do list -- for each one, the real question is whether your existing page already answers it.

• If it already does, there's usually nothing to change.
• If it partially does, expanding the relevant section is often enough.
• If the query represents a substantial adjacent topic on its own, that's when a separate page is worth considering.

Don't turn every new query into a new page. Use these impressions to see whether your existing page actually answers the questions people are asking.

What Not to Do

• Chasing volume without checking intent. A long-tail phrase with decent search volume but mismatched intent (informational searches when you're selling something, or the reverse) may bring visitors who are not looking for what your page offers, even if you rank for it.
• Writing unnaturally to fit the exact phrase. A page should read like it was written for a person, with the long-tail phrase and its natural variations worked in where they fit -- not repeated verbatim for the sake of matching the search term.
• Assuming every long-tail keyword needs its own page. One page can rank for several closely related queries. Before creating a separate page for a long-tail keyword, check whether it represents a distinct search intent or can be answered well on a page you already have.

Long-Tail Keywords and AI Search

People can ask AI assistants questions in fuller, more natural language than a typical short search query. Those questions can be useful when researching topics and the language people use to describe what they want. But that does not mean long-tail keywords have a special ranking advantage in AI search, or that writing in question form guarantees visibility there.

A conversational question isn't automatically a long-tail keyword. A long question can still describe a broad, competitive topic. How a single query can expand into several related ones inside an AI system is covered in more detail in the Query Fan-Out guide.

A Quick Long-Tail Keyword Checklist

• Does the keyword clearly show what the searcher wants, not just a topic?
• Have you checked Autocomplete and People Also Ask for natural phrasings, not just guessed variations?
• Have you checked what your own customers actually ask, in their own words, and whether people are actually searching for that?
• Are you covering closely related searches on the same page when they share the same intent, instead of building a separate page for each variation?
• Does the intent behind the phrase actually match what your page offers?
• Does the page read naturally, or does it feel like it's repeating the exact phrase for its own sake?
• Before dismissing a zero-volume keyword, did you check for real signals of demand instead of trusting the number alone?
• Have you checked Search Console for specific queries your existing pages already get impressions for but don't clearly answer?

Frequently Asked Questions