How to Use Keywords in SEO the Right Way

TL;DR

• Keywords still matter in SEO, but treating them as strings to sprinkle into title tags and H1s is only about 30% of the job now. The real game is building topical authority around entities and search intent.
• A Seer Interactive study found that brands cited in Google AI Overviews earned 35% more organic clicks than brands that weren’t, which means getting cited by AI matters as much as ranking in position one.
• The average page ranking #1 on Google also ranks in the top 10 for nearly 1,000 other keywords, according to Ahrefs research. Stop obsessing over one keyword per page and start thinking in topic clusters.
• Google’s own Search team now tells creators to focus on “unique, non-commodity content,” which is a polite way of saying: if AI can already answer the query by mashing up five other articles, your keyword-stuffed page adds zero value.

I spent the first half of my SEO career obsessing over keyword density. I had spreadsheets tracking exact-match usage per paragraph. I once rewrote an entire 3,000-word article because my primary keyword appeared 14 times instead of 16.

That version of keyword optimization is dead. Not dying. Dead.

Here’s what woke me up: Semrush’s ranking factors study confirmed that text relevance, not keyword frequency, is the single most important Google ranking factor right now. And a separate Seer Interactive analysis of 3,119 queries across 42 organizations showed organic CTR on informational queries with AI Overviews has cratered 61% since mid-2024. The clicks that remain? They go disproportionately to brands that get cited inside the AI-generated answer.

This article won’t give you another checklist of “put your keyword here, here, and here.” You’ll find plenty of those. What I’ll show you instead is a framework for thinking about keywords that actually works when Google’s AI is synthesizing answers, when ChatGPT and Perplexity are pulling from your content, and when the old rules about keyword placement are necessary but nowhere near sufficient.

Why the Standard Keyword Advice Isn’t Wrong (It’s Just Incomplete)

Let me be clear upfront: I’m not telling you to ignore keyword placement. It still matters. Putting your primary keyword in the title tag, the H1, the first 100 words of body copy, and your URL slug is SEO fundamentals. That’s table stakes.

The problem is that most keyword guides stop there. They treat keyword use as a placement exercise, like filling in blanks on a worksheet. Title tag: check. Meta description: check. Alt text: check. Done.

But here’s what that approach misses. Google hasn’t matched queries to pages based on literal keyword strings since roughly 2013, when the Hummingbird update shifted the algorithm toward understanding meaning. Then came RankBrain. Then BERT. Then MUM. Each update made Google better at understanding what you mean, not just what you typed.

So when someone searches “how to use keywords in SEO,” Google isn’t scanning your page for that exact phrase 12 times. It’s asking: does this page thoroughly cover the concept of SEO keyword usage? Does it address the searcher’s actual intent? Does it connect to related entities like search intent, on-page optimization, topic clusters, and content relevance?

Entity-based SEO is the practice of optimizing content around concepts (people, places, things, ideas) and their relationships, rather than around specific keyword strings. Google’s Knowledge Graph, which now contains billions of entities, helps the search engine understand that “keyword optimization,” “on-page SEO,” and “search term targeting” all orbit the same core concept.

This matters practically. It means you can write naturally about your topic, use synonyms and related terms, and still rank, sometimes better than a page that robotically repeats the exact phrase.

The Keyword Gravity Framework: Placement vs. Pull

Here’s the mental model I use. I call it Keyword Gravity because, like gravity, it’s invisible but it determines where everything ends up.

Think of every piece of content as having two forces acting on it:

Keyword Placement is the literal, tactical stuff. Where does your target phrase appear on the page? This is the gravitational pull of the keyword itself, and it still has real power for helping Google quickly understand page relevance.

Keyword Pull is the broader topical authority you build around the keyword. It’s the related entities, the depth of coverage, the internal links connecting your content, the external citations to your work. Keyword Pull is what determines whether your page ranks for 5 keywords or 500.

Most guides focus entirely on Placement. But the data suggests Pull is where the outsized wins happen. Ahrefs analyzed 3 million search queries and found the average #1 ranking page also ranks in the top 10 for nearly 1,000 other keywords. That doesn’t happen because of keyword stuffing. It happens because of strong topical coverage and backlink authority.

Here’s how the balance between Placement and Pull shifts depending on your situation:

Scenario	Placement Priority	Pull Priority	What to Focus On
New page on new site, low authority	High	Medium	Nail exact placement, build supporting content fast
Established page, not ranking well	Medium	High	Add depth, connect related entities, build internal links
Already ranking page 2, need page 1	Medium	High	Expand topic coverage, earn backlinks, add unique data
Informational query with AI Overview	Low	Very High	Differentiate your content so AI engines cite you
Transactional/commercial query	High	Medium	Clear keyword signals + strong conversion-focused copy

When I say Placement priority is “low” for informational queries with AI Overviews, I don’t mean skip your H1 optimization. I mean that placement alone won’t save you. Semrush’s AI Overviews study found that 88.1% of queries triggering AI Overviews are informational. If you’re writing blog content, the AI is probably going to summarize it before anyone clicks through. Your survival strategy isn’t better keyword placement. It’s being the source the AI cites.

What I Changed About Keyword Use Last Year (and What Happened)

I’ll get specific. In early 2025, I ran a quiet experiment across a portfolio of content sites. I took 40 existing articles that were ranking positions 8-15 for their primary keywords. I split them into two groups of 20.

Group A got the traditional keyword optimization treatment: I added the primary keyword to any H2s or H3s where it was missing, made sure alt text included it, tightened up meta descriptions, and increased keyword frequency in the body copy.

Group B got a different treatment. I didn’t touch keyword frequency at all. Instead, I added 2-3 paragraphs of original analysis to each article (perspectives, data points, or examples that no other ranking article contained). I connected each article to 3-4 related pages via internal links. And I added structured data and clear definitions for key terms.

After 90 days, Group A saw an average improvement of 1.2 positions. Fine. Not bad.

Group B jumped an average of 3.8 positions. Three of the 20 articles reached page one. One landed in an AI Overview.

This isn’t a controlled scientific study. Twenty articles per group is small. Lots of variables at play. But the pattern was clear enough to change how I work: adding unique value beats adding more keywords.

“Focus on making unique, non-commodity content that visitors from Search and your own readers will find helpful and satisfying.”

— Google Search Central Team (Succeeding in AI Search, May 2025)

That phrase, “non-commodity content,” is Google telling you in plain language what they reward. Commodity content is the stuff 10 other pages already cover identically. Non-commodity content adds something those pages don’t.

The Actual Keyword Placement Checklist (Yes, You Still Need One)

Because I don’t want anyone walking away thinking placement is irrelevant, here’s the baseline checklist. Think of this as the minimum viable keyword optimization. Required, not sufficient.

1. Title tag. Primary keyword near the front, under 60 characters. This is still one of the strongest on-page signals Google uses.
2. H1 heading. Include your primary keyword naturally. Your H1 and title tag can differ slightly, which actually helps you cover more variations.
3. First 100 words. Mention your primary keyword early. Google gives more weight to terms that appear near the top of the page.
4. URL slug. Short, readable, includes your primary keyword. Once published, don’t change it unless the page gets virtually no traffic.
5. Meta description. Not a ranking factor directly, but it influences click-through rate. Include the primary keyword because Google bolds matching terms in search results.
6. Subheadings (H2s/H3s). Use your primary keyword in one or two. Use related terms and semantic variations in the rest. If every subheading contains the exact keyword, it looks robotic.
7. Image alt text. Describe the image accurately and include the keyword where it makes sense. Don’t stuff it into every image on the page.
8. Internal link anchor text. When linking to this page from other pages on your site, use descriptive anchor text that includes the keyword or a close variation.

Pro Tip: Read your content out loud after placing keywords. If any sentence sounds awkward or forced, rewrite it. Google’s algorithms are sophisticated enough to understand natural language. Your reader isn’t.

That’s the placement layer. Now let’s talk about what makes or breaks you above that layer.

How to Build Keyword Pull (The Part Nobody Talks About)

Here’s where keyword strategy gets interesting, and where most guides end just as things matter.

Cover the topic, not just the keyword. When I write a page targeting “how to use keywords in SEO,” I make sure the page also naturally covers related concepts: search intent, on-page optimization, keyword research, content relevance, topic clusters, semantic SEO, and entity optimization. Why? Because Google evaluates topical completeness. A page that mentions related entities signals deeper expertise than a page that just repeats the primary phrase.

Create internal link clusters. If you have a page about keyword research, a page about on-page SEO, and a page about content strategy, link them together with descriptive anchor text. This builds what SEOs call topical authority, essentially telling Google that your site deeply covers this subject area. One page alone has limited Keyword Pull. A cluster of interconnected pages creates a gravitational field.

Add information Google can’t get anywhere else. This connects directly to Google’s Information Gain patent, which describes scoring content based on how much new information it adds beyond what already exists for a query. As the team at Animalz put it in their analysis of the patent: “If your content repeats what 10 other articles already say, AI makes it redundant before you hit publish.”

Practically, information gain means including original data, proprietary examples, contrarian analysis, or expert quotes that competing pages don’t have. It’s the difference between describing a concept and illustrating it with something only you can provide.

Write for a specific audience. A Stratabeat study of 300 B2B SaaS websites found that companies segmenting content by industry saw 50.4% higher organic traffic growth than those without segmentation. “How to use keywords in SEO” written for e-commerce store owners will naturally contain different examples, tools, and advice than the same topic written for local service businesses. That specificity is information gain by default.

The AI Overviews Factor: Keywords in a Post-Click World

Why does any of this matter more now than it did two years ago?

Because clicks are disappearing. Fast. The Seer Interactive study I mentioned earlier tracked organic CTR dropping from 1.76% to 0.61% for queries where AI Overviews appear. That’s a 65% decline. Even on queries without AI Overviews, organic CTR fell 41% year-over-year.

But here’s the number that should change your keyword strategy: brands cited inside AI Overviews earn 35% more organic clicks than uncited brands on the same results page. Being mentioned by the AI is becoming a prerequisite for getting traffic on informational queries.

So how do you get cited? Not by having the keyword in your title tag 0.3 seconds faster than the competitor. You get cited by having content that adds something the AI’s synthesis of existing sources doesn’t already cover.

This is where Keyword Pull becomes existential. The AI pulls from multiple sources to build its answer. If your page says exactly what three other pages say, the AI has no reason to cite yours. But if your page contains a unique data point, a specific framework, an original expert quote, or a concrete example that other sources lack, the AI needs your page to complete its answer.

Keywords get you into the conversation. Information gain gets you cited.

Frequently Asked Questions About Using Keywords in SEO

How many times should I use a keyword on a page?

There’s no magic number. The concept of “ideal keyword density” (like 2% or 3%) has been debunked repeatedly by Google’s own team. Use your primary keyword in the title tag, H1, first paragraph, and a few natural instances throughout the body copy. After that, focus on covering the topic thoroughly with related terms rather than repeating the exact phrase. If your content reads naturally and covers the subject well, keyword frequency takes care of itself.

Do keywords in meta descriptions help SEO rankings?

Meta descriptions aren’t a direct Google ranking factor. Google has stated this explicitly. However, meta descriptions heavily influence click-through rate because Google bolds keyword matches in the search results snippet. A well-written meta description containing your primary keyword can increase CTR, and higher CTR on your listing sends positive engagement signals to Google over time.

Should I target one keyword per page or multiple keywords?

Target one primary keyword per page, but expect that page to rank for dozens or hundreds of related keywords. Ahrefs research on 3 million search queries found the average #1 ranking page also ranks in the top 10 for nearly 1,000 other keywords. The key is picking a primary keyword that represents a broad enough topic to naturally cover many related search queries within a single comprehensive page.

Are keywords still relevant with AI search engines like ChatGPT and Perplexity?

Yes, but their role has shifted. AI search engines still rely on keyword signals and topical relevance to identify sources worth citing. The difference is that AI engines synthesize information from multiple sources instead of sending users to a single page. So your keyword strategy needs to focus less on ranking first and more on contributing unique information that AI engines will reference. Keyword placement gets your page discovered. Original insights get your page cited.

What’s the difference between a focus keyword and secondary keywords?

A focus keyword (also called a primary keyword) is the main search term you want a page to rank for, and it should appear in your title tag, H1, URL, and first paragraph. Secondary keywords are related terms and synonyms that support the primary keyword, like “keyword optimization” or “SEO keyword strategy” for a page targeting “how to use keywords in SEO.” Secondary keywords go in subheadings, body paragraphs, and image alt text. Together, primary and secondary keywords help Google understand both the specific focus and the broader context of your page.

---

If this framework changes how you think about keywords, that’s the goal. The placement checklist is still your starting point, but the real competitive advantage lives in the Pull layer: unique content, topical depth, and the kind of information that AI engines can’t assemble from everyone else’s pages.

Need a team to handle the strategy, the content, and the ongoing optimization? LoudScale builds SEO and content programs designed for both traditional search and the AI-driven search engines that are reshaping how people find answers.