Skip to content

How to Use Google's NLP API Demo to Write Title Tags

Sara Taher
3 min read
How to Use Google's NLP API Demo to Write Title Tags
How to Google NLP API Demo for SEO

Title tags are one of the very few things we have control over in SEO and they still matter. Even when Google re-writes the title tag and chooses to display the H1 tag instead for example (or a random alt tag 😸) it still takes the title tag into account for ranking purposes.

There are many tips and tricks out there to optimize the title tag. Some say make them shorter to avoid getting cut-off and, some say longer is better. I'm with team longer title tags.

Here's an interesting comment from Steve Toth on my post on LinkedIn about Title tags:

As you can see, SEOs have been testing and playing around with title tags to see what works better. And Steve mentioned 2 documented case studies where longer titles were found to be actually better. I agree!

If you're up to one more thing to play around with? I recommend you try, the Google NLP API demo:

You can basically input any title tag and see how google understands it in terms of entities. In this example, I provided the text "sara taher | SEO consultant" and Google identified both as a "person" entity.

Before we jump in (I was about to say dive in but remembered this will sound very chatgpt-y)

What is NLP (Natural language processing) in SEO

Natural Language Processing (NLP) in SEO involves using AI to analyze and understand text on web pages. NLP helps search engines identify the context, intent, and meaning of content. This improves how web pages rank for specific queries by aligning with user intent and enhancing the relevance of content.

I think it's also safe to assume that taking NLP into account, will help LLMs understand your content better.

How to Use NLP to Write Title Tags for SEO

Here's what I do and it's very simple! Let's say I'm preparing recommendations for a blog about "How to plan for a vacation"

Do your usual keyword research process, whatever that looks like. To keep it simple let's say you plug in the main keyword in an SEO tool and get something like this:

Keyword research

I take note of the different words used:

  • vacation
  • trip
  • road trip

Next I pull the top 10 results for the main keyword again "how to plan a vacation":

Now run those different titles in the NLP API Demo, and note down the different entities. For the first title tag "how to plan a vacation in 8 steps", the analysis looks like this:

Entities identified by google NLP demo
Entities identified by google NLP demo

Take note of those entities, and do the same exercise for more of the title tags. You may want to run this analysis for some of the title tags on the second page.

Once you have an understanding of what type of entities appear in those titles (e.g: number, organization, event, etc..), and potential keyword variations (e.g: trip, vacation, road trip) you want to create your title to show the most important entities as well.

This exercise can show you how ineffective some of the title tags are to avoid. Here's an example, let's say you want to make the title tag as "Guide to Vacation Planning"... here's how the analysis would look like:

No related entities identified! So that's something I'd definitely avoid. You can keep tweaking the title, changing the words you use and so on. Here's another failed example:

Obviously a work of art is not the best thing to describe your content 😼

My recommendation is to keep testing till you find the perfect combination that gets you optimal related entities.

Want to learn more about entity SEO? check this guide I wrote a while ago on how to do entity SEO.

That's that for today folks! See you next week!

SEO

Related Posts

Members Public

The Zero/Low Search Volume Dilemma

If we understand why search is changing, we can come up with tactics to optimize for that. In other words, if we know what's Google looking for and why they need AI Overviews and reddit, we can create content to cater for that, right? Last week we discussed

The Zero/Low Search Volume Dilemma
Members Public

Blogging for AI: Creating Content for LLMs and Modern Search

What was once considered helpful content by Google has changed. Take "Retro Dodo" for example, a website that was doing very good with all unique helpful first-hand experience [add whatever positive adjective here] content. In May 2023, the website hit 1 Million organic results - cool stuff right?

How to Blog for the LLMs and the New Era of Search
Members Public

Cutting Through the Noise

SEO is changing, SEO has changed. I have said those things myself few times and we've heard many others say them as well. But I also came to the realization that, we are becoming volatile, flip-flopping between new tactics and latest trends, and this may not sum up

Cutting Through the Noise