Moving Past Exact Match Key Terms

It seems like yesterday that SEOs began designing and optimizing content for Latent Semantic Indexing (LSI). Copywriters rejoiced at the ability to forego keyword stuffing in favor of focusing on creating compelling content. SEOs insisted that it would take time for LSI to be knowledgeable enough to “read between the lines – and here we are six years later. It looks like Google’s baby computer brain has reached critical mass. A multitude of recently published studies imply that exact match keywords may be losing their gusto in lieu of themed content.

As the evidence mounts, now is the time to move away from only targeting exact match keywords and to start testing themed content. Before we can jump into what that means for us as marketers, we should take the time to review the evidence.

Earlier this year Sam Nemzer posted an article on Moz titled “Are Keywords Really Dead? An Experiment.”  Through his experiment of evaluating transitional and informational keyword groups, Nemzer was hard pressed to find definitive evidence that we should jump the exact match train. A few months later Distilled took Nemzer’s test a step further by A/B testing titles and H1s. The test revolved around changing titles and H1s for user intent when shopping for hotels. Terms like “[city] hotel” became “hotels near [city].” Their results showed a rankings improvement when using terminology that was more in line with user intent. The final piece of evidence we’ll explore is Ahrefs’ report titled “On Page SEO in 2016: A (2M Keyword) Data Driven Analysis.” When this report was issued, the SEO community was furiously debating the meaning of the results, and what the future on on-page SEO would look like. While we don’t have to rehash all of the drama, we can acknowledge that this study, involving a data set of two million keywords, deserved the massive attention that it got. This article will focus on the data itself and analyze what it all means.


Figure 1Credit for image:

These results show a very small to non-existent correlation between exact match key terms in traditionally “important” SEO locations (H1, H2, content, metadata) and rankings.

Between this study and the other two mentioned, it can be confusing to figure out where SEOs should be headed and how they should plan on getting there. We know that Google is constantly improving its algorithms. Google is working towards meeting user intent, a difficult goal considering that some searchers (maybe most) don’t know what they are looking for until they find it. SEOs are caught in an interesting paradox; while SEOs benefit in some was from Google reaching towards this goal, they also are challenged by these changes as they must constantly change course to stay ahead of the curve. As Google’s algorithms appear to be straddling a line between serving content that meets user intent and content that has the right search terms in the right places, where does that leave SEOs?

A little in limbo.

When developing a keyword/theme strategy for a client, in addition to doing the traditional keyword research, checking search volumes and competition, SEOs should be taking things one step further. SEOs must “enter the mind” of their customers and brainstorm common questions that relate to informational as well as transactional queries (which we’ve reviewed in great detail in recent blog posts). Ask both easy questions and hard ones.

Here is an example for a pool supply and installation company. Notice the queries are complete sentences, not exact match terms. They represent questions that potential customers may ask. If an SEO only targeted exact match terms like “in ground pool installation” or “pool costs,” they could miss out on ranking for these types of queries.


Below are the SERPS for details about installation times for pools. The results are quite different depending on the user searches using key terms or a complete question.


In this example, we get a number of answers to our exact question. This is what users will see when they perform an informational search.


Above, the results for a shorter phrase are more convoluted as Google tries to figure out the intent. Does the user want to buy a pool or learn how to install it? It’s an SEO’s job to make sure that the content we produce can serve both purposes, hence the use of a brainstorming and mind mapping to generate possible themed content.

SEOs can take this strategy one step further by producing content that fits both transactional and informative needs. A blog topic on the pool installation process can be crosslinked from the pool installation product/service page, and vice versa. Monitor the rankings of these two pages as well as conversions to find out which style suits consumer needs as well as the search engine’s requirements. The results of these tests can guide your content strategy.

Between tests that are evaluating exact match keywords and the increasing opportunities to provide two types of content, SEOs should be aggressively testing these tactics. LSI is becoming increasingly more complex, paving the way for SERPs targeting search intent rather than keywords. It’s up to SEO companies to start bridging the gap and finding opportunities to provide content.