Inviting you to join me on my journey to build in public a Ranker clone using AI tools like Diffbot, Seomatic, Airtable, and Make

Have you ever wondered how websites like Ranker.com create hundreds of engaging and viral lists on any topic imaginable? How do they come up with so many ideas, write so much content, and rank so well on Google?

If you are curious about the secrets behind their success, you are not alone. I have been fascinated by Ranker.com for a long time and wanted to learn how they do it. That’s why I decided to embark on an ambitious project: to create a copycat of Ranker.com using AI tools.

In this ride along post, I will share with you my journey of creating thousands of webpages programmatically using chatGPT and other AI tools. My ultimate goal is to create a website that can rank anything and everything based on different attributes of facts and entities.

In this post, I will show you how I use Diffbot to source information in the form of facts and entities from various websites and databases. I will also show you how I use chatGPT to fill in the blanks between the facts and write content between entities. I will then show you how I use Seomatic to programmatically publish webpages in the style of Ranker based on different attributes of facts and entities. I will also show you how I use a knowledge graph to organize interlinking on the webpages on Seomatic. Finally, I will share with you how I measure the results of my project using Google Analytics and other tools.

If you are interested in learning how to create engaging and viral content using AI tools, this post is for you. You will discover how AI can help you generate ideas, write content, optimize SEO, and measure performance. You will also see some examples of webpages that I have created using this process.

Are you ready? Let’s dive in!

How to use Diffbot to source information in the form of facts and entities

Facts are pieces of information that can be verified by evidence or sources. Entities are things or concepts that have properties or attributes that can be used to describe them. They are important for creating ranking lists because they can help you generate topics, categories, criteria, examples etc.

For example, if you want to create a list of “The Best Movies Directed by Steven Spielberg”, you need some facts and entities such as:

-Fact: Steven Spielberg is an American filmmaker who has directed many blockbuster movies
-Entity: Steven Spielberg (type: person; occupation: filmmaker; movies directed: Jurassic Park; Schindler’s List; E.T.; etc.)
-Fact: Jurassic Park is a 1993 science fiction adventure film directed by Steven Spielberg
-Entity: Jurassic Park (type: movie; genre: science fiction adventure; director: Steven Spielberg; release date: June 11th 1993 etc.)
-Diffbot is an AI-powered tool that can extract structured data from any website or database. It can crawl web pages or APIs and return facts and entities in JSON format. You can also customize your queries using filters or parameters to get specific data that you need.

To use Diffbot for sourcing information for my project, I followed these steps:

-Create an account on Diffbot.com
-Choose a plan that suits my needs (I chose the Starter plan which gives me 10K API calls per month)
-Go to the Developer Dashboard
-Select “Custom API Toolkit”
-Enter a URL or API endpoint that contains information related to my topic
Choose “Analyze” as the extraction type
-Click “Run”
-Review the JSON output that contains facts and entities extracted from the URL or API
-Save or export the JSON output as a file

Here are some examples of facts and entities that I have sourced using Diffbot:

-Fact: The Godfather is a 1972 American crime film directed by Francis Ford --Coppola
-Entity: The Godfather (type: movie; genre: crime; director: Francis Ford Coppola; release date: March 15th 1972 etc.)
Fact: Tom Hanks is an American actor who has won two Academy Awards
-Entity: Tom Hanks (type: person; occupation: actor; awards won: Academy Award for Best Actor in 1993 and 1994 etc.)
-Fact: Apple Inc. is an American multinational technology company that produces iPhones, iPads, MacBooks etc.
-Entity: Apple Inc. (type: company; industry: technology; products: iPhone; iPad; MacBook; etc.)

How I plan to use airtable and make.com to query chatgpt programmatically to fill in the gaps between facts and entities

One of the challenges of creating a website like Ranker.com is to generate engaging and relevant content for each list item. For example, if I want to rank the best movies of all time based on their IMDb ratings, I need to write a short description for each movie that explains why it deserves its spot on the list.

To do this manually would take a lot of time and effort. That’s why I decided to use chatgpt, an AI tool that can generate natural language text based on a given prompt. Chatgpt can write anything from stories and poems to summaries and reviews.

But how can I use chatgpt to write content for my website? That’s where airtable and make.com come in handy. Airtable is a cloud-based spreadsheet tool that lets you store and organize data in different tables. Make.com is a visual platform that lets you design, build, and automate workflows without coding.

I plan to use airtable as my database for storing facts and entities related to different topics. For example, I can have a table for movies with columns for title, year, genre, director, actors, IMDb rating, etc. Then I can use make.com to create a workflow that queries chatgpt programmatically using the data from airtable.

The workflow would look something like this:

Trigger: When a new row is added or updated in the movies table
Action: Create a prompt for chatgpt using the data from the row
Action: Send the prompt to chatgpt via API and get the response
Action: Update the row with the response as the content
This way, I can automatically generate content for each movie based on its facts and attributes. The content would be unique and engaging because chatgpt can write in different styles and tones depending on the prompt.

Of course, this is just one example of how I plan to use airtable and make.com to query chatgpt programmatically. There are many other possibilities for creating different types of lists based on different criteria. The sky is the limit!

So follow me if you're interested in seeing more about my journey here. I haven't created a product page on IndieHackers yet because it's not far enough advanced yet to do that. I still need a name for my project! Any ideas?