What answer is a searcher looking for? For sustainable, valuable search traffic, you’d better provide it.
Satisfying search intent is Google’s fundamental goal. But algorithms haven’t always kept pace. Proxies like backlinks and keywords have long been—and still are—stand-ins for the likelihood that a web page will satisfy user intent.
Optimizing for intent is the long play, for Google and your site. A page that’s well-matched for user intent can outperform those that optimize primarily for search engines—in search and after the click.
It’s an SEO strategy that focuses on making users happy rather than hitting a particular keyword density or winning exact-match anchor text.
Still, to translate the “make users happy” bromide into something executable, you need to know a few things:
- How Google (and others) define search intent;
- How to evaluate your target keywords for intent;
- What to do with search intent data.
Table of contents
- 1. How Google (and others) define search intent
- 2. How to evaluate your target keywords for intent
- 3. What to do with search intent data
1. How Google (and others) define search intent
For Google, understanding search intent is the key to returning useful search results. (And, by extension, the key to maintaining and growing its search market share, thus capturing more eyeballs for ads.)
The classic division of search intent offers three variations of queries:
- Informational. Learn something (e.g. how to train for a marathon).
- Transactional. Buy something (e.g. running shoes order online).
- Navigational. Go to a specific site (e.g. runners world training plans).
Past studies have estimated that as many as 80% of queries are informational, with the remainder split equally between the other two types.
Google’s latest Search Quality Evaluator Guidelines identify four main types of intent:
- Know. “The intent of a Know query is to find information on a topic. Users want to Know more about something.”
- Do. “The intent of a Do query is to accomplish a goal or engage in an activity on a phone. The goal or activity may be to download, to buy, to obtain, to be entertained by, or to interact with a website or app.”
- Website. “The intent of a Website query is to locate a specific website or webpage that users have requested.”
- Visit-in-person. “Some queries clearly ‘ask’ for nearby information or nearby results (e.g., businesses, organizations, other nearby places).”
The guidelines also identify two sub-types:
- Know Simple. “Know Simple queries seek a very specific answer, like a fact, diagram, etc. This answer has to be correct and complete, and can be displayed in a relatively small amount of space: the size of a mobile phone screen. As a rule of thumb, if most people would agree on a correct answer, and it would fit in 1–2 sentences or a short list of items, the query can be called a Know Simple query.”
- Device Action. “Device Action queries are a special kind of Do query. Users are asking their phone to do something for them. Users giving Device Action queries may be using phones in the hands-free mode, for example, while in a car [. . .] A Device Action query usually has a clear action word and intent.”
Many keywords fall clearly into one bucket or another. Some don’t.
What happens when search intent is ambiguous?
Over time, Google has gotten better at parsing search intent, particularly for ambiguous queries. (The 2013 Hummingbird update is often cited as a major improvement in Google’s understanding of search intent.)
Bill Slawski offers a simple example of a query with ambiguous intent:
If someone enters “new york pizza sunnyvale” (without the quotation marks) into a search box at Google or Yahoo or Bing, it’s not quite clear whether they are looking for: (1) pizza in New York, in a neighborhood or area referred to as Sunnyvale, (2) New York style pizza in a place called Sunnyvale, (3) a place called “New York Pizza,” in Sunnyvale, or (4) some other result.
As Kevin Indig notes, longer queries tend to be less ambiguous. (Voice search may also reduce ambiguity because voice queries are usually longer than text queries.)
Queries closer to a sale tend to be longer and less ambiguous, too. The initial consumer research that starts with “coffee grinder” may yield follow-up queries like “conical burr grinder reviews” as the searcher progresses toward a purchase.
Geography (i.e. IP address) can provide clues to search engines, as can search history, time of year, or time of day. For example, an ambiguous query like “flowers” may return different results on February 14 compared to July 14.
Because some queries blend multiple types of intent, intent categories are best understood as “probabilistic.”
For site owners, ambiguity can be an advantage. For example, Justin Briggs suggests that forums and other sites full of user-generated content reveal “when Google is ‘reaching’ for a good result.” The imperative? If you can answer that query clearly, the traffic is up for grabs.
There are other methods of evaluating search intent, too, like active vs. passive intent.
Active vs. passive intent
Active intent, A.J. Kohn notes, is “explicitly described by the query syntax.” It’s not the only intent of the query, however. And, Kohn continues, to satisfy users, you need to meet passive intent, too.
Passive intent is implicit in the query. It’s best identified by asking yourself “what the user would search for next … over and over again.”
In an example shared by Kohn, the query “bike trails in walnut creek” asks explicitly (i.e. active intent) for a list of bike trails. It also implicitly asks (i.e. passive intent) for other information like maps, trail reviews, and photos.
Satisfying passive intent, Kohn argues, is essential for user engagement and conversion. If active intent brings in users at the top of the funnel, passive intent engages and converts them. It’s “the way you build your brand, convert users and ween yourself from being overly dependent on search engine traffic.”
There is a caveat, according to Kohn, in addition to wider challenges:
One of the mistakes I see many making is addressing active and passive intent equally. Or simply not paying attention to query syntax and decoding intent properly. More than ever, your job as an SEO is to extract intents from query syntax.
So, how do you identify intent for the keywords you care about?
2. How to evaluate your target keywords for intent
For plenty of queries, the intent is obvious. For example, “portable phone charger reviews” is pretty straightforward.
Because bottom-of-funnel queries tend to offer more information (and less uncertainty), evaluating intent is more critical at earlier stages, with informational queries. Those informational queries are often the highest volume terms a site targets—key drivers of awareness and acquisition.
For smaller sites, intent evaluation is quick and easy. A manual process works. For larger sites, however, scaling that process is essential. Here’s how to do both.
How to evaluate search intent manually
Look at the search engine results page (SERP). What does it show? Do all results suggest a similar intent? Or do they satisfy a range of potential intents?
In SERPs, Google shows its hand. Top-ranking search results are ample evidence of what users want:
- Which types of sites rank highly? Individual sites? Aggregators? Blogs? Government and university sites?
- What type of content is on those pages? Long-form articles? Short explanations? Images? Videos?
- What is the first question answered? What text is offset or included in headers? What subtopics are (or aren’t) covered?
The SERP for “best restaurants richmond va” tries to satisfy two different intents:
- Local map listings with tons of five-star Google reviews. For searchers in Richmond who want to call or visit a local restaurant.
- Blue links of aggregator sites with “Best Restaurants” lists. For searchers anywhere who want to browse options.
One takeaway: If you run a restaurant, thinking that you can “optimize” your site to get listed among the blue links would be a lost cause.
While this process is simple and intuitive, it doesn’t scale. So what can you do when need to decode intent for thousands of pages?
How to scale intent evaluation
If you’re already tracking keywords in one of those tools, you can export the list and assign intent categories based on the type of search result. For example:
- SERPs that return a featured snippet are more likely to be Know Simple queries.
- SERPs with a high cost-per-click (data those tools also return) suggest a bottom-of-funnel or transactional query.
- SERPs without any ads suggest top-of-funnel informational intent.
- SERPs with map results suggest Visit-in-person intent, etc.
Depending on your industry, different features may hint at different intents. You can sample keywords with various SERP features and code the intent.
So, if you’re trying to assign intent to 10,000 keywords, manually review the intent for 50 keywords for each SERP feature, then programmatically assign intent to the remainder.
Another way to do it is to classify keyword modifiers by intent. (A lengthy list of modifiers is available here.) Research From STAT, now part of Moz, suggests where certain modifiers fall along the intent spectrum:
If you’re starting with a massive list of keywords, you can use an N-gram tool to identify common modifiers within your keyword data. The most common phrases can serve as a basis for classification (and help automate tagging in a spreadsheet).
Categorizing keyword modifiers is especially useful for sites that have hundreds or thousands of similar pages, like review sites with city-specific content or sites with hundreds of similar products.
Keyword modifiers are also an easy way to expand the set of keywords you track. After all, the goal of identifying intent is not just to see where you meet it but where you may need to expand content to do so (more on that later).
A Moz article offers an example of informational modifiers for products:
- [product name]
- what is [product name]
- how does [product name] work
- how do I use [product name]
The end result—for manual or automated tagging—is a spreadsheet that classifies keywords by intent:
Whether you choose Google’s four-intent model or another is up to you. You could, for example, map keywords based on your user journey. It’s one of several high-value things you can do with search intent data.
3. What to do with search intent data
Search intent data can support initial research, improve keyword tracking, or add a sharper business focus to reporting. It can also guide on-page content choices, content strategy, or web design.
Using search intent data for research and evaluation
1. Map content to the buyer journey.
A Think with Google report contends that
People turn to their devices to get immediate answers. And every time they do, they are expressing intent and reshaping the traditional marketing funnel along the way.
Customers use search engines from initial consideration through the purchase—and past it. You can map that intent to your funnel. The result is a framework for evaluating search performance based on larger business goals.
For example, while all of your blog posts may qualify as “Informational” in intent, some may serve users in different stages of awareness:
A journey-based mapping of keyword intent pays dividends for competitor research, as well as keyword tracking and reporting.
2. Identify content gaps with competitor research.
Where in the user journey are you struggling? Which intent gaps are competitors filling? Tools like SEMRush and Ahrefs offer keyword-based domain comparisons.
You can enter your domain and several competitor domains. Then, filter for keyword modifiers that you’ve mapped to intent. For example, Ahrefs and Moz are outperforming SEMRush for several informational “how to” queries:
This analysis scales to compare performance at each stage of the funnel. At the same time, it provides a ready-made list of topics to try to close the gap.
Competitor analysis identifies keywords for which your site might reasonably rank. A blue-sky approach to keyword research often yields queries that you’d like to rank for but for which Google perceives an alternative intent (e.g. displays aggregators when you’re an individual site, or vice versa).
3. Track rankings based on intent.
Instead of reporting on keywords by topic (e.g. “We rank well for Product X but not Product Y”), you can measure performance in the context of your marketing funnel.
For example, you may do well for bottom-of-funnel queries (across several products) but struggle to rank for top-of-funnel informational content.
Tracking based on intent is a smarter way to prioritize content expansion, new page creation, or page design tweaks.
Using search intent data for page design and development
4. Add content to answer active and passive intent.
What else could you answer for users? What questions will they have next?
Google’s Knowledge Cards, Kohn offers, are a perfect example of aggregating intent—answering a query and providing valuable context. A restaurant name query, for example, answers so many more questions:
What type of restaurant is it? Is it expensive? Where is it? How do I get there? What’s their phone number? Can I make a reservation? What’s on the menu? Is the food good? Is it open now? What alternatives are nearby?
You may need to expand content on an existing page. Or you may want to create new pages to address unfulfilled user intent. The “Expand vs. Create” decision often hinges on search volume. If the subtopic has search volume, create a new page; if it doesn’t, expand the current one.
Briggs offers a framework for ongoing page development:
One method we’ve used is to write a broad, robust article first while trying to cover several aspects of the topic. We wait for it to start ranking well, then mine Google Search Console for the keywords where we’re 6 through 15. These are typically good candidates for longtail, specific follow-up posts.
Larger sites, he notes, may succeed by targeting high-volume, highly competitive terms first. Smaller sites, in contrast, benefit by targeting several long-tail queries, then attacking a top-level keyword after they’ve built topical authority.
5. Tailor content to win more clicks in the SERPs.
Google’s definition of a Know Simple query hints at several guidelines for featured snippets:
- 1–2 sentences in length;
- Short lists;
- “Correct and complete” responses;
- Fit neatly within a mobile phone screen.
Featured snippets are a common target of search engine optimizers. They generate in-SERP visibility and tons of clicks—but they can cannibalize them, too.
If the search intent is to get a quick answer and not click any link, optimizing for featured snippets may satisfy users (and Google) but ultimately erode organic traffic for all sites (a Prisoner’s Dilemma, according to Rand Fishkin).
It still makes sense to optimize for featured snippets. But the value may be limited to “URL awareness” as users, especially those on mobile, don’t click through.
Beyond featured snippets, there are other ways to try to improve click-through rates. Fishkin highlights an underused strategy: writing page titles and meta descriptions for intent, even at the expense of keyword targeting.
That strategy has risks, but it’s a potential pathway for “underdog” sites to compete against industry stalwarts. If you can get to the bottom of Page 1, a page title and meta description written for humans (rather than search engines) could help differentiate your site, earn more clicks, and (probably) send positive signals back to search engines.
6. Design pages to satisfy active intent first.
“It’s essential to understand the hierarchy of intent so you can deliver the right experience,” Kohn contends. “This is where content and design collide with ‘traditional’ search.”
For SEO, page design has two imperatives:
- Answer active intent clearly and immediately;
- Provide a logical hierarchy of information to satisfy passive intent.
For Know queries, is the answer clearly visible via header tags, larger font, or an offset block? Are follow-up questions answered with subheads? For Transactional queries, is the next click clear and easy to find?
These are basic principles of UX—but they also have an impact on search performance. Users who don’t find answers immediately are likely to bounce straight back to search results. The “UX is a ranking factor” argument has some truth—and controversy.
We all endure recipe sites that require a lengthy scroll to get to the recipe. That’s because the preceding text (usually a banal essay about the recipe’s origin) provides context for search engines. That context can help sites rank in a vertical like recipes, where search engines can’t differentiate a good chocolate chip cookie from a life-changing one.
Googlers like John Mueller continue to dissuade webmasters from producing content that serves search engines at the expense of the user experience. But the tension remains—those tactics still work.
The lesson, then, is to take the long view. Google would prefer not to value supporting text when it’s superfluous, though it may still reward sites for it now. Slowly, that need will decline. Periodically testing its removal—to see the impact on rankings and user behavior—is worthwhile.
“Target the keyword, optimize the intent.” Kohn’s maxim is the best summary of how search intent fits into an SEO effort. Keywords remain the starting point for a page. But intent should guide decision-making about what those pages should look like.
Creating a list of relevant keywords and categorizing them by intent—whether you target them now or not—can show you where in the user journey you enjoy visibility, and where you don’t.
That intent data can:
- Prioritize content expansion on existing pages;
- Identify the need for new pages;
- Suggest a page design that quickly and clearly solves for active intent first.