The clearest examples of keyword stuffing come from Google itself. Its spam policies on Google Search Central define the practice as “filling a web page with keywords or numbers in an attempt to manipulate rankings in Google Search results.” Note the two words that matter: keywords or numbers, and attempt. Google is judging intent and effect, not counting how many times a phrase appears.
That distinction is where most advice on the subject goes wrong. It treats stuffing as a threshold you can measure and stay under. It is better understood as a writing problem that a reader would notice before any algorithm did. This piece explains how to recognise it and remove it. It does not explain how to get away with it.
Examples of keyword stuffing, in Google’s words
Google lists three examples in its keyword stuffing policy. They are short, and each one describes a pattern you will still find on live sites:
- “Lists of phone numbers without substantial added value”
- “Blocks of text that list cities and regions that a web page is trying to rank for”
- “Repeating the same words or phrases so often that it sounds unnatural”
The second example is the one service businesses still trip over. A footer or sidebar reading “We serve” followed by forty town names is a classic stuffing block. It adds nothing a visitor needs, and it exists only because someone hoped each town name would pull in a local query. If the business genuinely serves those places, a map, a short list of service areas or a proper page per real location does the job without the padding.
The first example, phone numbers, sounds odd until you see the pages it describes: long lists of numbers published with no context, built to catch people searching a number they do not recognise. The third example is the one most writers picture, and it is also the one that sounds worst when read aloud.
Where it hides on a page
Semrush, in a guide by Rachel Handley, describes keyword stuffing as “a spammy SEO tactic that involves unnaturally forcing keywords into webpages”. Its list of where that forcing shows up is useful because most of the places are not the body copy.
| Location | What stuffing looks like | What to do instead |
|---|---|---|
| Body text | The same phrase repeated until sentences read unnaturally | Write for the reader; use synonyms and related terms |
| Title tags and meta descriptions | Several variants of one keyword strung together | One clear title that states what the page is |
| URLs | Slugs that repeat the keyword or chain variants | A short, readable slug |
| Anchor text | Keyword-rich anchors repeated across many links | Varied, descriptive anchors |
| Image alt text | Keyword lists instead of a description of the image | Describe what the image actually shows |
The five locations come from Semrush; the third column is our summary of its avoidance advice and Google’s examples, not a quote. Alt text deserves the most attention. It exists for people using screen readers, and a string like alt="cheap shoes buy cheap shoes best cheap shoes" on https://example.com/shoes fails them completely while helping nobody else.
Anchor text is the subtler case. Internal links that all use the same exact-match phrase are a mild version of the same habit, and on inbound links the pattern can look like a link scheme. Our guide to anchor text covers how to keep it varied without making it vague.
Does keyword stuffing work?
Not in any way worth the risk. The received wisdom is that there is a safe keyword density, often quoted as a percentage, and that stuffing simply means going over it. Google has never published such a number, and we have taken that idea apart in our piece on keyword density. Search systems that understand synonyms and topics do not need a phrase repeated to work out what a page is about.
It is also on our list of SEO myths that more mentions mean more relevance. Past the point where the topic is clear, extra repetition adds nothing to relevance and makes the page worse to read. That is a cost with no matching benefit.
There is a second cost that gets less attention. A stuffed page converts badly. Visitors who land on copy that reads like a list of search terms tend to assume the site is low quality, and they leave before reading far enough to trust it. For an affiliate or lead-generation page, that lost trust costs more than any ranking the repetition might have bought. So even in the unlikely case that stuffing moved a page up a position or two, the traffic it brought would be worth less than the traffic a clean page earns.
The Google keyword stuffing penalty
Keyword stuffing is one of the named practices in Google’s spam policies, and a page that breaks them can lose rankings or drop out of results. In practice there are two routes. Algorithmic systems can quietly discount or demote the page, with no notice. A human reviewer can apply a manual action, which appears in Search Console and can be fixed and submitted for reconsideration.
Our guide to a Google manual action walks through that reconsideration process. The algorithmic route is harder to diagnose, because the only symptom is a page that ranks worse than its content seems to deserve.
How to avoid keyword stuffing
Semrush’s avoidance advice is short, and it is right. Target a few primary and secondary keywords per page rather than every variant you can find. Write naturally. Vary your anchor text. Everything else follows from those three habits.
A practical audit takes four passes. Read the page aloud and mark any sentence that exists to carry a phrase rather than a point. Check the title, meta description and URL for repeated variants. Review alt text on every image and rewrite any that list keywords instead of describing the picture. Finally, look at footers, sidebars and templates, because a stuffing block in a template appears on every page that uses it.
That last pass catches the most. Site-wide blocks of town names or product terms are often left over from an earlier owner or an old theme, and they multiply quietly. Removing one template block can clean up hundreds of URLs at once, which makes it the highest-value fix on most sites.
One limitation is worth stating plainly. Google does not say how much repetition trips its systems, and it never will, so nobody can tell you a safe number. The only reliable test is the reader’s: if a sentence sounds unnatural, it is a liability whether or not anything detects it.
Frequently asked questions
What are examples of keyword stuffing?
Google names lists of phone numbers without added value, blocks of city and region names a page is trying to rank for, and repeating words or phrases so often that it sounds unnatural.
Is there a safe keyword density?
Google has never published one. Mention the topic as naturally as a good writer would and stop when the meaning is clear.
Can stuffed alt text cause problems?
Yes. Alt text is one of the places keyword stuffing commonly shows up, and stuffed alt text also fails people who rely on screen readers.
How do I recover from keyword stuffing?
Rewrite the affected pages and templates. If a manual action is shown in Search Console, fix the issue and request reconsideration there.
The takeaway Keyword stuffing is a writing problem before it is a ranking problem. Edit for the reader, clean up templates and alt text, and stop chasing a density number Google has never published.
