e ways to optimize sitewide entities using AI
Discover three simple yet powerful strategies to help you maximize your efforts faster and more efficiently.
This article will teach you:
- How to use AI as a source of data for optimizing web pages.
- Create multiple schema types quickly for each blog post on your website using AI.
- Internal linking is made easy with this free solution.
Key Terms
It’s important that we connect these concepts with entities before we start.
Entities
In machine learning and natural language processing in particular, an entity can be defined as a thing or concept that is significant in a certain context.
Apple could represent a technology company, or a fruit type, depending on context.
Links to internal pages
They are links that lead to another page within the same website. They can help establish a hierarchy of information for a website and direct visitors to the most important sites.
These links are important from an SEO perspective. They help search engines discover, index, and understand all of the pages on your website.
Search engines can also be helped by internal links to understand the context of different pages, and how they relate to each other. Search engines can use this to identify relevant entities on the site.
Schema
Schema.org is a community-driven collaborative project that aims to develop, maintain and promote schemas on the Internet for structured data.
It’s basically a way of categorizing and tagging information on your website so that search engine spiders can better understand the content. It could be anything, from a review of a product to an announcement about an upcoming event.
You can help search engines understand and identify the entities on your website by using schema markup.
EAV (Entity-Attribute-Value)
EAV is an entity-oriented data model that describes entities with a large number of attributes.
This is especially useful in SEO, as it allows information about entities to be represented in a flexible way. This can be helpful in situations where attributes used to describe a entity may vary greatly.
Search engines try to understand both the intent of the searcher and the context meaning of the terms that appear in the searchable space.
By combining the concept of entities with schema, EAV and internal links, the search engine can better understand the content of a website and provide the most relevant results.
The following three tactics will help you improve the “semantic comprehension” of your site. Together with other SEO techniques, they can improve your website’s visibility on the SERPs.
1. The hexagramarian prompt sequence
This prompt has been designed specifically around entities. The definitive guides to entities explains that understanding of entities is closely tied to grammar.
Entities are integrally linked to nouns and adjectives. Verbs, adverbs, prepositions, and EAV.
It’s actually much simpler than it seems. ChatGPT explains its importance.
Here are six prompts to get you started:
- Provide a list of all the verbs that are commonly used to describe the topic. “Cluster the list, and nest verbs associated with it.”
- Provide a list of all the common nouns that are associated with this topic. “Cluster the list to nest nouns associated with it.”
- Provide a list of all the adjectives that are commonly associated with this topic. “Cluster the list, and nest associated adjectival terms.”
- Provide a list of all the adverbs that are commonly used with this topic. “Cluster the list, and nest verbs associated with it.”
- Provide a list of all the common predicates that are associated with this topic.
- Provide a list of all attributes related to the topic.
You can save your answers in any way you want. I use a Google Sheet.
This list can be used in several different ways.
Use the hexa list to help you create content briefs. The hexa list could be used to optimize existing content. This list could be used to create clever OpenAI prompts.
This list will help you to deploy AI programmatically.
2. Schema and AI
It is a real pain to create schema for hundreds or thousands of web pages. In the last few years, agencies began to productize schemas because no one wanted to do it.
Many people choose an out-of the-box solution because schema is difficult to understand, and its benefits are not always appreciated (e.g. SEOpress, RankMath or Yoast).
Andrew Ansley, the co-author of this article, has developed a simple system for schema creation. This doesn’t require you to know anything about schema.
You only need an OpenAI Playground and GPT-4.
In this example we will use the webpage schema to create a structured and clear declaration of the entities found in the article.
Set up
Step 1. Grab the URL of your blog post.
Step 2 In the system text, provide the name of the author, the website URL, the organization name, the publisher name and the web page URL.
Step 3 : Paste your blog text into the system text.
The system text can be seen on the left in the picture. The prompting appears in the middle of page and works exactly as ChatGPT, but with an upgrade.
You can change the settings in Playground and use the system text as memory/context.
You can use 8,000 tokens in the playground (.75 words = 6,000 words).
The prompt is:
The prompt is simple. Enter the following into the user box:
- “Create WebPage JSON-LD schema. Use mentions and about. Use entities and Wikipedia pages for the about and mentions.
You’ll then get some professional schema that you can add to your post. To inject header code in a blog post, we usually use a plugin. But that’s just one option.
If you don’t trust the AI, you can go to https://validator.schema.org/ and provide the code to see if it has any errors.
You can see in the image that my pro schema code for entity optimization has no flaws. Huzzah!
You can also add the FAQPage schema to your blog.
FAQPage Schema
This schema type is much simpler than the example before. This schema allows you to replace the article text with anything you want.
The prompt is:
- “Create FAQ schema.”
The system will produce something similar to the example below if you have the text in the system.
If you prefer ChatGPT then the web browser plugin is for you (which is easier to use).
You only need to:
- Create FAQ schema for URL [insert a link here].
You can ask ChatGPT for FAQ Schema per URL if you wish to add multiple URLs.
It’s impossible not to be a nerd when I see how simple it is for AI to perform SEO tasks. You can tell the AI what you want and it will create the code.
You don’t need to ask for the AI to provide code in order to achieve a more programmematic approach to schema. Use this simple prompt:
- “Create FAQ schema in JSON-LD format. Once you have completed one URL, you can initiate another prompt to continue the task.
Answer:
The results generated by this method will stop around the sixth URL.
You can use the “continue” word to continue the list if you have more URLs than the output size.
3. This script is used to create internal links
Internal links play a vital role in passing authority from one page to another, improving Google Bot’s crawlability, and communicating topic relationships.
Here is a script Ansley wrote that only requires three things.
- Google Docs
- Google Sheets
- App Scripts.
The document is the article that you are writing and will be linked to all your other articles.
Google Sheets contains all your keywords and URLs. I use a website map.
Next, ask ChatGPT for the keyword to be identified from the URL.
This is the script.
function addLinksFromSheet() // Your Google Sheets ID var sheet = SpreadsheetApp.openById('1AOvyOL0PCVVjY86aEEx77RqUWthV1m5Vzs2SyXE2f7g').getActiveSheet(); // Get all the rows of data in the Sheet, excluding the headers var data = sheet.getRange("A2:B" + sheet.getLastRow()).getValues(); // Iterate over all the rows for (var i = 0; i < data.length; i++) // The first column in the row contains the text to be hyperlinked var searchPhrase = data[i][0]; // The second column contains the URL var hyperlink = data[i][1]; // Use your addLinks function to add the hyperlink to the doc addLinks(searchPhrase, hyperlink); function addLinks(searchPhrase, hyperlink) { if (!searchPhrase) return; // Your Google Doc ID const document = DocumentApp.openById('1_T5BRHiNi3iPnEL6xtys2qEV9WNzPZAwsQdIqtgiFtI'); const body = document.getBody(); let search = null; let count = 0; while (count < 1) { search = body.findText(searchPhrase, search ? search : null); if (!search) break; const searchElement = search.getElement(); const startIndex = search.getStartOffset(); const endIndex = search.getEndOffsetInclusive(); searchElement.asText().setLinkUrl(startIndex, endIndex, hyperlink); count++; } document.saveAndClose(); }
I highlighted the part of your script that needs a unique document slug.
Hamish’s YoTtube Income Surfers channel was the inspiration for this script. Hamish may be a YouTuber newbie, but he has some great content if you want AI and SEO.
The GIF below demonstrates how to use Google App Scripts.
- Save the project.
- Click Run to accept the permissions.
The screenshot below shows the results.
This setup allows you to give any of your authors an easy way of linking internally.
Most outsourced writers are unsure of what they should link to.
Create a database, and keep it updated so that your writers are able to apply the links you have specified.
You can paste additional keywords into ChatGPT. For illustration, I’ve used five keywords and 5 URLs.
Prompt:
- “I would like to expand this table.” Each URL should be repeated 5 times, and 4 keywords added to each URL. The new keywords must be highly semantically relevant to the original keywords in the list.”
Output:
You can now choose between five keywords. Modifying the script is necessary to avoid linking to the exact same URL. Here’s the updated version:
function addLinksFromSheet() // Your Google Sheets ID var sheet = SpreadsheetApp.openById('1jTwgsEgz5X4BbiNDjQeHJrRVtYdGi3LEKEBvPgIYtg8').getActiveSheet(); // Get all the rows of data in the Sheet, excluding the headers var data = sheet.getRange("A2:B" + sheet.getLastRow()).getValues(); // Keep track of the used URLs var usedUrls = []; // Iterate over all the rows for (var i = 0; i < data.length; i++) // The first column in the row contains the text to be hyperlinked var searchPhrase = data[i][0]; // The second column contains the URL var hyperlink = data[i][1]; // Check if this URL has been used already if (usedUrls.indexOf(hyperlink) === -1) // It hasn't, so add the hyperlink to the doc addLinks(searchPhrase, hyperlink); // And mark this URL as used usedUrls.push(hyperlink); function addLinks(searchPhrase, hyperlink) { if (!searchPhrase) return; // Your Google Doc ID const document = DocumentApp.openById('1rLL9J6Lag6McZ6F22R3ptyb7XFqg9Bc1qmEgobrTa3w'); const body = document.getBody(); let search = null; let count = 0; while (count < 1) { search = body.findText(searchPhrase, search ? search : null); if (!search) break; const searchElement = search.getElement(); const startIndex = search.getStartOffset(); const endIndex = search.getEndOffsetInclusive(); searchElement.asText().setLinkUrl(startIndex, endIndex, hyperlink); count++; } document.saveAndClose(); }
AI is a powerful tool for SEO.
Three specific examples show how AI can dramatically reduce the time required to optimize for entities.
We recommend AI for repetitive or highly-technical tasks and spend more time on strategy or creative SEO functions. Entity SEO follows the same rules.
It is important to note that this list is not exhaustive and there are other ways in which AI can be used when optimizing entities.
Attend my session on Entity SEO at SMX Advanced from June 13-14 to learn more. We will also explain why we think Entity SEO is crucial for top SERP rankings in the future.
Andrew Ansley is the co-author of this article.
This is the third in the series of articles on entity SEO. You can read the first two articles by clicking here.
The article 3 ways AI can be used to optimize sitewide entities first appeared on Search Engine Land.