LSI keywords
What are LSI keywords?
LSI keywords is a popular SEO name for terms related to a page’s main keyword, such as "crust" and "mozzarella" on a pizza recipe. It comes from latent semantic indexing, a retrieval method from the late 1980s. Google has said it does not use LSI, so the label misleads, though covering related terms is sound.
Latent semantic indexing was designed to find relationships between words in small, fixed sets of documents. It does not scale to the web and predates the methods modern search engines use. In 2019 Google’s John Mueller stated plainly that there is no such thing as LSI keywords. Google instead understands language with its own systems, including machine learning models and its Knowledge Graph of entities.
When people say LSI keywords, they usually mean one of three things: synonyms and variations of the keyword, related subtopics that a thorough page would mention, or entities connected to the topic (people, places, products, concepts). All three are worth covering, not because a formula requires them, but because a page that covers a topic fully tends to mention them naturally.
Why LSI keywords matter for SEO
The useful idea behind LSI keywords is topical completeness. A page about "how to brew pour-over coffee" that never mentions grind size, water temperature or a filter is probably thin. Checking which related concepts the best pages cover is a good way to find gaps in your own.
The risk is treating "LSI keyword lists" as words to sprinkle in. That produces awkward text, can drift into keyword stuffing, and does nothing that writing a genuinely complete page would not do better.
How to find related terms for a topic
- Read the top-ranking pages and note the subtopics and concepts they all cover.
- Check People Also Ask, related searches and autocomplete for the keyword.
- Look at the secondary keywords the top pages also rank for.
- Add the missing subtopics as sections or examples where they genuinely help the reader.
- Skip any term that does not fit; there is no required list.
LSI keywords example
For "how to brew pour-over coffee", related concepts include grind size, water temperature, bloom, coffee-to-water ratio, paper filter and gooseneck kettle. A complete guide explains each because brewers need them, not to hit a list.
LSI keywords FAQ
Does Google use LSI keywords?
No. Google has said it does not use latent semantic indexing. It understands related terms through its own language models and Knowledge Graph.
Are LSI keywords a myth?
The name and the claim that Google uses LSI are myths. The underlying advice, covering related subtopics and concepts, is reasonable.
What should I call LSI keywords instead?
Related keywords, secondary keywords, subtopics or entities, depending on what you mean.
Related SEO terms
Secondary keywords are related terms a page targets alongside its primary keyword: variations, synonyms and subtopics with the same intent.
Semantic search is search that matches the meaning and intent behind a query, not just its exact words.
An entity is a distinct, identifiable thing, such as a person, place, organisation or product, that search engines and AI models recognise.
The Knowledge Graph is Google’s database of entities and the facts and relationships between them, used to power knowledge panels and answers.
Keyword density is how often a keyword appears on a page as a share of total words. It isn’t a ranking target; Google has no ideal density.
Keyword stuffingBlack hat
Keyword stuffing is filling a page with keywords, numbers or place names in an unnatural way to manipulate rankings, against Google’s spam policies.
Content optimizationWhite hat
Content optimization is improving a page’s content so it better matches search intent, covers the topic fully and earns more clicks and rankings.