Hit Enter to Search or X to close
How to Classify Search Intent
Google's search results have changed in recent years. Google has dramatically improved its ability to produce search results that meet user expectations without relying solely on traditional factors like links, meta titles, etc.
It has also been shown that Google only uses meta titles and meta descriptions a third of the time. Search intent then becomes a dominant factor in multiple types of search results.
The big problem with SEO is that current methods used to classify search intentions are often lacking. SEO tools are not always able to give clues about the intentions behind the detected keywords.
The Navigation, Information, Transaction model
There are several search intent classification models.
The first classification that most of us know is "navigational", "informational" or "transactional". This methodology dates back to the article “A Taxonomy of Web Search” by Andrei Broder of Altavista, published in 2002.
The “See, Think, Do” and “Know, Go, Do & Buy” models from Google
Fast forward in the early 2010s and Google began to refer to its own variation in the classification of search intentions by speaking of "micro-moments.” These micro-moments are defined as Know, Go, Do, & Buy. These basically cover the same categories as Andrei Broder's previous model.
It's great for newbies, but it's not best for content creators and marketing strategists.
Unfortunately, I think the systems that currently exist are good at explaining research intent in hypothetical terms. However, when you try to apply them to keyword research and teach creators and writers how to factor them into their content, their usefulness is reduced.
Moreover, this keyword classification framework does not take into account the overlap of intentions. Most of the time, we categorize a keyword without checking the results in Google to validate the true category this keyword should be placed in. An SEO strategist looks at keywords in the hundreds or thousands and fills in a column with Transactional, Informational, or Navigation. In short, the keyword classification process must evolve. That's for sure.
Another Approach to Classifying Search Intent Today
So how do you measure search intent more usefully and reliably? I think it's more helpful to rank search intent in a way that closely aligns with the characteristics of the Search Engine Results Pages (SERP).
Here are the types of intent that we started using:
1. Search Intent
Search intent is one of the most common types of results. These are usually search phrases that generate results like Wikipedia pages, definition boxes, examples of academic work, blog or feature articles, and other SERP features that suggest to users answers to or ideas on a topic that they are looking for.
2. Question / Answer Intent
Slightly different from search intent, there are a number of searches where users generally don't care about clicking on a result. They want a quick response.
Good examples are definition boxes, answer boxes, calculation boxes, sports scores, etc. Some examples to keep in mind:
- Queries about the weather - the user wants a clear answer
- Loan Calculator
- Sports results
- Conversion units
Users looking to buy products or to research those products. These are easy to spot, as Google tends to be aggressive with this type of intent. Multiple prominent ecommerce products or category pages (next to purchase results) and highly rated product snippets are good indicators of transactional intent.
4. Local Intent
Dominated by a combination of local packs and geographic markers, this type of intent is truly geolocated. The location of maps and information panels to the right (knowledge graph) is a good indicator that we are looking at a location-specific query. If a Google Maps map comes out first in the results, it's a very strong local intent indicator.
5. Visual Intent
Many SERPs have image packs in the top 100 results by default. If you see Pinterest in search results or a pack with only images, it has visual intent detected by Google.
6. Video Intent
As we look at more and more search results, it becomes clear that video is a separate type of intent. Between the video carousels, video thumbnails, and even video clips that are now becoming commonplace, video is becoming essential in ranking for certain types of queries.
These video carousels are the most common indicator of video intent that we see. Another very strong indicator of video intent is video thumbnails.
7. Intention to keep abreast of current affairs
Top Stories boxes, recent tweets, and the heavy use of recent dates in the past few days / weeks / months in organic results are all signals which imply that Google sees something as viral.
Featured links in search results are a clear sign that users are looking for news and current affairs. Recent tweets are another type of result typically triggered by keywords classified in this category.
8. Brand Intent
Brand intent queries tend to be dominated by brand home page results with site links (related site links displayed below the main result).
Another option is the grouping of results from the same domain.
Links to important sites are good indicators of brand intent. Multiple blue links for the same term clearly indicate brand intent.
9. Fragmented Intent
When we see signs that there are more than one type of query intent, it means “fragmented intent” for Google.
These ambiguous queries are often difficult to categorize. It is more difficult to position yourself on this type of request, because too many elements are offered to the user by Google.
The goal for the search engine is to offer the minimum to satisfy each potential intention. This doesn't leave much room for performance, as this type of query can have elements of all of the other types listed above.
In conclusion, Google's search is now more effortlessly attuned to users' needs. Search intent becomes a dominant factor in multiple types of search results. Embrace this more human-first approach to create content that ranks well.
Want more? Here are other articles for you