While Google has made strides in understanding user intent, Director & Product Manager Elizabeth Tucker says specific queries remain challenging.
In a recent episode of Google’s Search Off The Record podcast, Tucker discussed some lingering pain points in the company’s efforts to match users with the information they seek.
Among the top offenders were searches containing the word “not” and queries involving prepositions, Tucker reveals:
“Prepositions, in general, are another hard one. And one of the really big, exciting breakthroughs was the BERT paper and transformer-based machine learning models when we started to be able to get some of these complicated linguistic issues right in searches.”
BERT, or Bidirectional Encoder Representations from Transformers, is a neural network-based technique for natural language processing that Google began leveraging in search in 2019.
The technology is designed to understand the nuances and context of words in searches rather than treating queries as a bag of individual terms.
‘Not’ There Yet
Despite the promise of BERT and similar advancements, Tucker acknowledged that Google’s ability to parse complex queries is still a work in progress.
Searches with the word “not” remain a thorn in the search engine’s side, Tucker explains:
“It’s really hard to know when ‘not’ means that you don’t want the word there or when it has a different kind of semantic meaning.”
For example, Google’s algorithms could interpret a search like “shoes not made in China” in multiple ways.
Does the user want shoes made in countries other than China, or are they looking for information on why some shoe brands have moved their manufacturing out of China?
This ambiguity poses a challenge for websites trying to rank for such queries. If Google can’t match the searcher’s intent with the content on a page, it may struggle to surface the most relevant results.
The Preposition Problem
Another area where Google’s algorithms can stumble is prepositions, which show the relationship between words in a sentence.
Queries like “restaurants with outdoor seating” or “hotels near the beach” rely on prepositions to convey key information about the user’s needs.
For SEO professionals, this means that optimizing for queries with prepositions may require some extra finesse.
It’s not enough to include the right keywords on a page; the content needs to be structured to communicate the relationships between those keywords.
The Long Tail Challenge
The difficulties Google faces with complex queries are particularly relevant to long-tail searches—those highly specific, often multi-word phrases that make up a significant portion of all search traffic.
Long-tail keywords are often seen as a golden opportunity for SEO, as they tend to have lower competition and can signal a high level of user intent.
However, if Google can’t understand these complex queries, it may be harder for websites to rank for them, even with well-optimized content.
The Road Ahead
Tucker noted that Google is actively improving its handling of these linguistically challenging queries, but a complete solution may still be a way off.
Tucker said:
“I would not say this is a solved problem. We’re still working on it.”
In the meantime, users may need to rephrase their searches or try different query formulations to find the information they’re looking for – a frustrating reality in an age when many have come to expect Google to understand their needs intuitively.
Why SEJ Cares
While BERT and similar advancements have helped Google understand user intent, the search giant’s struggles with “not” queries and prepositions remind us that there’s still plenty of room for improvement.
As Google continues to invest in natural language processing and other AI-driven technologies, it remains to be seen how long these stumbling blocks will hold back the search experience.
What It Means For SEO
So, what can SEO professionals and website owners do in light of this information? Here are a few things to keep in mind:
- Focus on clarity and specificity in your content. The more you can communicate the relationships between key concepts and phrases, the easier it will be for Google to understand and rank your pages.
- Use structured data and other technical SEO best practices to help search engines parse your content more effectively.
- Monitor your search traffic and rankings for complex queries, and be prepared to adjust your strategy if you see drops or inconsistencies.
- Monitor Google’s efforts to improve its natural language understanding and be ready to adapt as new algorithms and technologies emerge.
Listen to the full podcast episode below: