What is User Intent? How to optimize for it like a pro

User Intent has become one of the most important concepts in modern SEO. In this guide, I explain what it is, how to identify it, and how to optimize for it.

User Intent is an old concept in information retrieval and yet one of the most important concepts in modern SEO. The growing use of machine learning allowed Google to take its application to new heights, making it a critical gatekeeper to organic traffic.

Over the recent years, I’ve written and spoken a lot about advanced methods to identify User Intent at scale, but I never took the time to define it properly. With this article, I aim to fill the gap and explain the ins and outs of User Intent, why it matters, and how to optimize for it.

What is User Intent?

User Intent, or Search Intent, is the goal a user aims to achieve when searching on Google or other search engines. Searchers have learned to use specific language to navigate search results, and search engines use machine learning models to understand what exactly users mean. In other words, User Intent is all about turning implicit into explicit meaning.

When I search for “sushi near me”, for example, I expect a list of sushi restaurants in reach. That’s simple. When I search for “DevOps”, though, I might want to learn more about what it means, or maybe what DevOps tools there are. It’s much harder for a search engine to decipher what a user is looking for.

In SEO, User Intent is a ranking enabler: If a page doesn’t meet the intent Google deems right for a keyword, it won’t rank!

Google takes many different signals into account when ranking search results. We can group them into quality, relevance, authority, and experience. User Intent is a relevance signal (think: “how relevant is a piece of content for a keyword?”). Meeting Search Intent means serving content in the format users expect and meeting their needs in the context of the keyword they used.

Simple, but not easy.

Why does Search Intent matter?

If you get Search Intent right, you can rank a single piece of content for thousands of keywords. Users express the same goal in different ways. “What to do in SF” is the same as “things to do in SF”, for example, and Google understands that users want to accomplish the same thing. Both terms share the same intent. In plain terms, by meeting User Intent right on the head, you maximize the traffic potential of your content.

If you don’t satisfy User Intent, your chances of ranking are low. Search Intent has become more powerful than backlinks and content in SEO. In other words, no matter how good your content or backlink profile is, without meeting User Intent, you won’t drive traffic.

How Google determines User Intent

Two milestones significantly improved Google’s understanding of User Intent. First, the launch of Hummingbird, an update of the ranking architecture, in 2013 allowed Google to take the context of words and their relationships into account when determining the relevance of content.

Before Hummingbird, Google matched the words in a search query precisely with how they appeared in meta titles or body content. In other words, Google took everything literally. Post-Hummingbird, Google could better match the meaning (semantics) of search terms with results. The machine was able to get what people really meant.

Hummingbird introduced the concept of entities to Google Search. In 2019, I gave a presentation about entities in Cambridge, England, where I explained:

If I were to try to define what entities are, I would say they are semantic, interconnected objects that help machines to understand explicit and implicit language.

And this is the crucial part right here to understand: it’s not just about the explicit thing you say, it’s also about implicit concepts, ideas, trends and all these kind of things.

Presenting the concept of entities at Cambridge in 2019
Presenting the concept of entities at Cambridge in 2019

So, User Intent and entities are concepts that build on each other. The relationship between entities helps Google understand the intent for related queries.

The second milestone is Rankbrain, the evolution of Google’s understanding of semantics. Rankbrain uses machine learning to map words to vectors, called Word Embeddings, and match them to similar or related words. This way, Google can understand queries it has never seen before better than ever.

Googlers stated that Rankbrain is “the third most important factor in the ranking algorithm” (source). However, don’t confuse the wording with what SEOs call “ranking factors.” Rankbrain is not a ranking factor but a part of Google’s technology responsible for matching what people search for with the most relevant results.

The most common challenges with User Intent

Identifying and optimizing for User Intent poses 6 big challenges for SEOs:

  1. User Intent can change over time
  2. It’s challenging to identify User Intent at scale
  3. Devices make a difference
  4. Location plays a role
  5. Google might want to show a mix of different results
  6. Keywords can have different interpretations

In the rest of the article, I will cover each of those challenges and provide suggestions for solving them.

User Intent isn’t static

User Intent can change over time as searchers’ goals change. One example is the keyword “Wuhan,” which used to be a search term for the city in China. When the Covid Pandemic broke out, people’s intentions when searching for it changed. Instead of learning more about the city, searchers wanted to know more about the virus outbreak. So, Google showed Top Stories and results that cover the virus instead of destination sites.

The results for "Wuhan" changed when the Covid pandemic broke out
The results for "Wuhan" changed when the Covid pandemic broke out

In Google’s Quality Rater Guidelines, a catalog of outcomes and concepts for human quality raters that help them evaluate the algorithmic changes Google makes to search, Google explicitly points out that queries naturally change over time.

Google mentions in the quality rater guidelines that query intent may change over time
Google mentions in the quality rater guidelines that query intent may change over time

Another example is the keyword “independence day,” which can mean both the holiday or the movie. Every year around July 4th, Google shuffles the search results to show more results related to the holiday than the movie.

The results for "independence day" change around July 4th
The results for "independence day" change around July 4th

The difference between the two examples is that the former had a constant intent that changed at a point in time; the latter has seasonal intent changes. As an SEO, it’s essential that we understand the difference, optimize accordingly and measure success in the right way.

Different types of User Intent

There are many different types of User Intent. A common misconception is that every intent fits into three categories: informational, navigational, transactional. Searchers either want to learn, buy, or navigate. That understanding is outdated.

At least, Google now shows an updated definition of User Intent types In the Quality Rater Guidelines:

  • Informational = Know and Know Simple
  • Navigational = Website and Visit-in-person
  • Transactional = Do and Device Action
Google shows finer segmentations for user intent
Google shows finer segmentations for user intent

In reality, searchers have very specific types of goals. For example: “I want to learn what a word means” (definition), “I want to compare prices,” “I want to find an image for my presentation,” “I want to book a trip.”

User journeys are messy, which is why we have to come up with a more refined segmentation of their intentions.

In Maximizing every click with Search Journeys, I show an example of how simple some User Intents seem while user behavior is actually very complex:

All too often, we get lost in the notion that a funnel leads users in a straight line from awareness to decision. However, most search journeys are messier. In a paper from 2016 (!), Think With Google shows how many digital touchpoints users have when going through more complex search journeys. Google used clickstream data to analyze searches, clicks, visits, and video views of several people when planning a trip. In one of the case studies, “Snapshot of a traveler’s decision-making journey,” the participant has over 850 digital touchpoints over the course of 3 months! She performs 166 searches total, 24% of interactions happened on mobile, and 24% of interactions happened on maps (19% on Search).

I suggest coming up with your own definition of User Intent for the keywords you care about. Ask yourself, “what do people really want when searching for a keyword?” and come up with your own classification. In the section about optimizing for User Intent, I will explain how to do that for many keywords.

The intent behind a query can also depend on a user’s location. For example, when searching for “Tumeric”, users might get results for restaurants with that name if they search on their smartphone and a restaurant with that name is nearby and results for the plant if that’s not the case. So, User Intent is location-based.

Queries with multiple meanings

Keywords can have multiple interpretations
Keywords can have multiple interpretations

When a keyword has more than one meaning, we speak of “ambiguous queries” or “the degree of ambiguity.” A keyword like cat, for example, is not as indicative of what searchers want as dry cat food for sensitive stomachs.

I extensively cover Query ambiguity in Solving Fragmented Intent: the shorter a keyword is, the higher is the degree of ambiguity. The cat food example is just one of many. Short keywords carry many intentions. Searchers looking for the query business might want a definition of the word, business ideas, or navigate to the Small business Administration.

For a query with multiple meanings, Google distinguishes between three types of interpretations (page 69 of the Quality Rater Guidelines):

Dominant Interpretation: The dominant interpretation of a query is what most users mean when they type the query. Not all queries have a dominant interpretation. The dominant interpretation should be clear to you, especially after doing a little web research.

Common Interpretation: A common interpretation of a query is what many or some users mean when they type a query. A query can have multiple common interpretations.

Minor Interpretations: Sometimes you will find less common interpretations. These are interpretations that few users have in mind. We will call these minor interpretations.

As a result of query ambiguity, Google shows a mix of several results that might satisfy the dominant, common, and minor interpretation of the keyword. For SEOs, that means we have to be conscious of what results can appear for a keyword in the top 10. It might be that there are only 3 slots in the top 10 for a specific type of site.

Look at the example “video conferencing software”: it shows 3 review sites, 2 brands, and 4 publishers in the top 10.

Search results pages show different types of sites based on what users expect
Search results pages show different types of sites based on what users expect

That’s the power of User Intent right there! Understanding it means understanding that Google might limit the number of results from a particular site.

Long keywords, however, have a low degree of ambiguity. The keyword business ideas for college students in Kenya brings very exact results of the same type: listicles with ideas specifically for people living in Kenya.

How to optimize for User Intent

To optimize for User Intent, we need to do two things: identify user intent and understand user expectations.

Identifying User Intent can be very simple when we talk about a single keyword: google the keyword and look at the sites and types of content that Google ranks in the top positions. You can derive User Intent from the top positions like a self-fulfilling prophecy.

If you see a lot of articles at the top, you probably need to write an article to rank well. If you see tools, you need to build a tool. The best performing content leads the way.

You should pay attention to the title of the results because they might tell you what type of content you need: reviews, a definition, long-form, a product page, etc.

Based on User Intent, different types of content can rank
Based on User Intent, different types of content can rank

Identifying User Intent for many keywords is a challenge because you cannot look at all search results for thousands of keywords. To solve this challenge, I came up with a method to reverse engineer User Intent based on what SERP Features Google shows for a keyword.

See, Google wants to provide searchers the best answer possible, and it does so by augmenting the search results with modules like Map Packs, Shopping Ads, Image Carousels, Videos, etc (=SERP Features). By using a 3rd party tool like SEMrush to get a list of SERP Features that appear for your target keywords, we can let Google do the work and learn from the search results.

I outline an exact step-by-step process in User Intent Mapping on Steroids. The evolution of the model is Solving Fragmented Intent, where I provide an example of User Intent classification for many keywords, show how their intent changes over time, and explain how to track it.

Unfortunately, we cannot look at a number to quantify Search Intent. Google doesn’t tell us “your article satisfies User Intent to 80%.” Instead, we need to apply a mix of common sense and data to reverse engineer how Google identifies User Intent for a keyword or set of keywords.

The second step, understanding user expectations, is tricky. Google expresses user expectations as “Needs Met,” a spectrum of how helpful search results are for users.

Results fall into one of 5 categories:

  • Fully meets needs
  • Highly meets needs
  • Moderately meets needs
  • Slightly meets needs
  • Fails to meet needs

Note that ambiguous queries cannot have “Fully Meets Needs” ratings because they have multiple interpretations. “Fully meets needs” can only be assigned when there is a clear answer to the search. That’s why we see higher fluctuations for search results for shorter keywords, which I describe in Patterns of SERP Volatility.

“Highly meets needs” applies to results that satisfy the common interpretation of a query and have a high degree of authority, accuracy, and credibility. Google calls out that results that highly meet users’ needs “often have some or all of the following characteristics: high quality, authoritative, entertaining, and/or recent (e.g., breaking news on a topic).” Medical and scientific results must represent well-established consensus and be accurate.

In simple terms, you need to think about what expectations users have for a specific intent - and then top that. Unfortunately, there is no way to do this at scale. Writers need to assess user expectations when creating the content if we look at a site that produces content itself.

A method I use to fine-tune User Intent targeting and top user expectations is Content Tuning. I publish content to my best understanding and then monitor what queries Google tries to rank it for and expand the content over time. Google tests new content in the search results, and Content Tuning is an effective way to signal Google that your content is the best.

For marketplaces and other aggregators, user expectations need to be measured on the product level. SEOs need to ask themselves whether a product or category page satisfies and tops expectations and build or optimize features accordingly.


Meeting User Intent is one of the most critical factors for ranking high in Google Search. Intent is not static, so we must understand and monitor it for the keywords we want to rank for.

To determine User Intent, Google looks at the context (location, device, previous searches) users search in, what results are available, and how close the query is to other known queries. SEOs can look at the top results and SERP Features to learn from Google and identify User Intent. Then, they have to translate that into content and aim to top users’ expectations in the keyword context.

As Google’s technology evolves, it better understands the implicit meaning behind users’ searches. It’s not perfect - yet - but we should expect it to reach a human-like level over time. That should also lead to continuous ranking fluctuations when big (political, socio-economic, or environmental) events happen or core algorithm updates launch.