Understanding User Intent with BERT





Blog Date

June 3, 2024


UK, Manchester

Follow us on


Table of Contents

Understanding User Intent with BERT

The Rise of Conversational Queries and the Challenge for Search Engines

As an SEO specialist, I’ve always been fascinated by the ever-evolving nature of search. Over the years, I’ve seen search queries become increasingly complex and conversational. Gone are the days of simple keyword-based searches. Today, people are turning to search engines with longer, more natural language queries, seeking in-depth information and precise answers to their questions.

This shift in user behavior has presented a significant challenge for search engines like Google. Historically, these systems have relied on keyword matching to surface relevant results. But when confronted with the nuanced language of conversational queries, their performance can falter. Suddenly, those carefully crafted “keyword-ese” strings that we SEOs once optimized for are no longer cutting it.

BERT to the Rescue

Thankfully, the brilliant minds at Google have been hard at work on this very problem. In 2019, they introduced a groundbreaking natural language processing (NLP) model called BERT (Bidirectional Encoder Representations from Transformers). This powerful AI tool represents a massive leap forward in the search giant’s ability to truly understand the intent behind user queries.

As Google explained, BERT models can consider the full context of a word by looking at the words that come before and after it. This allows them to grasp the nuanced meaning of language, rather than simply matching keywords.

For example, let’s say a user searches for “Brazil traveler to usa need a visa.” The word “to” and its relationship to the other words in the query are crucial to understanding the meaning. BERT can recognize that this is about a Brazilian traveling to the U.S., not the other way around. In the past, Google’s algorithms would have struggled with this, potentially returning irrelevant results about U.S. citizens traveling to Brazil.

But with BERT in the mix, Google’s search engine can now crack these complex, conversational queries and provide much more relevant and helpful information to the user.

Passage-Based Ranking: The Next Evolution

Just when you thought Google couldn’t get any smarter, they’ve gone and upped the ante once again. In a recent announcement, the search giant revealed another groundbreaking development: passage-based ranking.

As Marie Haynes explains, this new capability allows Google to identify individual passages within a web page and assess their relevance to a particular search query. Rather than simply evaluating the page as a whole, Google can now surgically surface the specific section that best answers the user’s question.

This is a game-changer for SEO. Gone are the days when a comprehensive, authoritative page would automatically outrank a more targeted, passage-level response. Now, even smaller, less-established sites have a chance to shine, as long as they can provide the most relevant and informative answer to the user’s query.

Optimizing for Intent: The Future of SEO

So, what does all of this mean for the future of SEO? In my opinion, the name of the game is going to be understanding and optimizing for user intent.

As Marie Haynes astutely points out, successful websites in the years to come will be those that can produce content tailored to the specific needs and queries of their target audience. It’s no longer enough to simply stuff your pages with keywords and hope for the best. You need to dig deep, really understand what your users are looking for, and then deliver that information in the most concise and compelling way possible.

This is where BERT and passage-based ranking come into play. By leveraging these advanced NLP capabilities, you can identify the exact pain points and information gaps that your users are struggling with, and then create content that addresses those needs head-on.

Of course, this is easier said than done. Optimizing for user intent requires a delicate balance of research, creativity, and technical know-how. But trust me, the effort will be well worth it.

The Future is Bright (and Conversational)

As an SEO agency in Manchester, UK, we at MCR SEO are thrilled to see these advancements in search technology. We believe that the rise of conversational queries and the power of BERT and passage-based ranking represent an exciting new frontier for our industry.

Gone are the days of keyword-centric optimization. The future of SEO is all about understanding and anticipating the needs of your users, and then crafting content that speaks directly to those interests and pain points. It’s a challenge that requires a deep dive into the human psyche, but the payoff can be truly transformative for your website’s performance.

So, if you’re ready to embrace the conversational revolution and unlock the full potential of your online presence, I encourage you to reach out to us at MCR SEO. We’re here to help you navigate the ever-changing landscape of search and ensure that your content is always one step ahead of the curve.

Copyright 2023 © MCRSEO.ORG