Natural language processing technology (NLP) can be found in your word processor, Amazon’s Alexa, and other platforms you may use regularly. In this article, we explain how NLP and SEO can work together.
In this article
Search engine optimization (SEO) is a vital marketing tool for any small to midsize enterprise (SME) today. Millions of potential customers are just one click away from your business, so it’s up to you to make sure your content and products come up when users hit the magnifying glass.
Search is also big business for the search engines themselves. Search ad revenue at Google holds more than 85% of Canada’s search market, according to a Statista report. A Forbes article also reported that search ad revenue made up 58% of Google’s parent company’s US$65.1 billion overall revenue for the third quarter of 2021. Because competition may be fierce, search engines are looking to improve all the time, meaning that businesses who want to remain on that crucial first page of results must keep an eye on developments and adjust their SEO strategy accordingly.
In the past, companies have tried to ‘game’ search engine’s algorithms into ranking their pages more highly, using tactics such as stuffing content with keywords that aren’t necessarily helpful or relevant to a reader. Search engines, by contrast, likely respond by refining their algorithms to select the most useful pages for each searched query.
One of the most significant tools that modern search engines potentially use is natural language processing or NLP. In this article, we explore what this means, how it affects SEO, and what SMEs can do to ensure that search results pages remain a valuable marketing channel for them.
What is NLP?
NLP stands for natural language processing. It is a field of scientific study that aims to convert written or spoken human language into code that computers can understand and vice versa. This may have many academic aspects and there are many practical applications of natural language processing software in the real world. These include the grammar checker on your word processor, machine translation software, speech processing built into digital assistants like Amazon’s Alexa, and automated subtitling for online videos.
NLP, as we now think of it, has been studied since the 1940s, and the discipline has progressed as technology has improved. One reason for the renewed interest in NLP could be the development of machine learning technology based on artificial neural networks —computers that seek to mimic the function of human brains— which can potentially make it quicker to get useful results.
What does NLP have to do with search?
Search engines look to translate a human query into instructions that a machine can understand. At a very basic level, search engines can operate using keywords. If someone searches for ‘basketball’, the search engine can provide the most popular pages about basketball, for example. But most searches are more complex than that, and keywords alone don’t give enough of an indication of the intent behind a searcher’s query.
Some search engines, therefore, may be using NLP to try and understand the intent behind a query. At the same time, they may be crawling through online content and applying similar processes to human-generated content, with the goal of surfacing the right content for each query.
There are many other factors that could influence how search engines behave, not least of which could be statistics. Search engines base their results partly on what other people have clicked on —and therefore found useful— in the past. But NLP could potentially serve a valuable role in providing accurate results, especially for new or complex search queries.
How is SEO changing?
According to Google, 15% of the billions of searches it handles each day are new, meaning a statistical approach may not yield effective results. To better understand what people want when they type a search query, Google has developed an NLP technique called bidirectional encoder representations from transformers (BERT). Google has made this technology available as open-source software, so developers can use it for their own applications, but for marketers and managers at SMEs, there are some implications for SEO.
The goal of BERT is to cater less for searches that are based on keywords —like ‘cheap flights new york’— and more for those that sound like a person asking a question, such as ‘who has the best deals on flights to new york’. In the former example, it would be difficult for Google to know whether the searcher wanted information on flights from or to New York. In the second, it can more easily interpret the searcher’s intent and provide more appropriate results.
How does NLP affect my SEO strategy?
The days of keyword stuffing your pages may be long gone. SEO best practice today could potentially be to provide relevant, helpful, and authoritative content for the queries you are looking to target.
Google’s Public Search Liaison, Danny Sullivan, reiterated this on Twitter when BERT was announced, saying: “There’s nothing to optimize for with BERT, nor anything for anyone to be rethinking. The fundamentals of us seeking to reward great content remain unchanged.” and “BERT doesn’t change the fundamentals of what we’ve long said: write content for users.”
Google offers its own extensive SEO advice, which contains technical tips about site construction and structure, as well as pointers about content. The most important of these, the one that “will likely influence your website more than any of the other factors” is making a page interesting and useful, which will encourage more people to visit it and recommend it.
In specific terms, this means generating fresh, well-written content that you know your audience wants. This, in addition to building and labelling web pages in ways that make it easy for search engines to understand their content, could be the most effective advice that search engines themselves consistently offer.
What’s the takeaway?
Now you know what NLP is and how it works, but it shouldn’t necessarily change the way you approach search as a marketing channel. Good SEO is still good SEO, and search engines will likely find content that fits what your audiences are looking for. If there is a secret, it may lie in understanding your audience’s intent when they search for products or services you provide and having content ready for when they do.
For further reading on SEO and NLP, try these resources:
- 3 Tactics to Boost Your B2B SEO Strategy
- Google’s SEO Starter Guide
- Optimize Website Content for SEO (full report available to clients)