Unlocking Untapped User Growth for OpenAI via SEO
Unlocking Untapped User Growth for OpenAI via SEO
with keyword analysis & action items for OpenAI team
OpenAI and all its own properties get traffic of 1.5B monthly users (source: Similarweb).
Here is a breakdown of its notable properties with traffic and keywords (source: Ahrefs).
For a particular URL, traffic tells the estimated number of monthly visits and keywords tell the number of organic keywords that rank within 100 in SERP.
Despite the significantly high traffic, there are a couple of SEO-based opportunities that OpenAI can implement to increase its traffic further. This post covers those strategies and their implementation in detail.
TLDR:
User-Generated Content from ChatGPT:
Enable public sharing of ChatGPT conversations and use SEO-friendly URLs to tap into Reddit/Quora/WikiHow's keyword markets.
OpenAI Platform:
Optimize example pages into dedicated URLs with meta tags to improve SEO and target high-value keywords.
OpenAI's Cookbook on GitHub:
Rename GitHub notebooks to match high-volume user search queries.
ChatGPT Plugins:
Create dedicated, optimized pages for each ChatGPT plugin to capture searches.
Model Detail & Comparison Page:
Develop individual and comparison pages for AI models using code names to own authoritative results.
ChatGPT Download Search Query:
Build a unified "ChatGPT Download" page with deep links to provide safe download access.
Address unofficial versions and malicious ads.
Other Pages:
Target high-volume keywords like "ai image generator" for DALL-E and create a "ChatGPT wiki" page.
Table of Contents
User Generated Content from ChatGPT
OpenAI Platform
OpenAI's Cookbook on GitHub
ChatGPT Plugins
Model Detail & Comparison Page
ChatGPT Download Search Query
Other Pages
1. User Generated Content from ChatGPT
ChatGPT offers a unique feature that allows users to share their conversations via a URL.
Currently, these URLs look something like this:
https://chat.openai.com/share/2fc08a7f-dc59-431f-b1ff-cb3179cc23f5.While this is functional, it's not optimized for search engine visibility, which is a missed opportunity for OpenAI.
Recommendations for OpenAI
1. Enable Public Conversations for Search Engine Indexing
OpenAI should consider allowing users to make their conversations public, thereby making them indexable by search engines like Google. This would turn user-generated content into a powerful SEO asset.
2. Adopt SEO-Friendly URL Structures
To further enhance SEO, OpenAI should adopt a more human-readable URL structure. For example, a URL could look like this: https://chat.openai.com/share/transformer-model-explained-2fc08a7f. This format is not only easier to read but also more likely to rank well in search engine results.
The Potential Impact: A Quantitative Look
According to SimilarWeb, OpenAI currently receives 2 billion monthly users. By implementing the suggested SEO strategies, OpenAI has the opportunity to tap into a significant portion of the keyword market from other high-traffic platforms. Here's a breakdown:
Reddit: With 2 billion monthly users, Reddit ranks for a total of 75.5 million keywords. Among these, 7.4 million are informational or learning-oriented. By optimizing for SEO, OpenAI could potentially tap into approximately 9.8% (7.4M/75.5M) of Reddit's keyword market in this specific category.
Quora: This platform has 705 million monthly users and ranks for 49 million keywords. A conservative estimate suggests that around 8.2 million of these are informational or learning-oriented. OpenAI could thus tap into about 16.7% (8.2M/49M) of Quora's keyword market in this category.
WikiHow: With 110 million monthly users, WikiHow ranks for 14.2 million keywords. Of these, an estimated 4 million are informational or learning-oriented. OpenAI could potentially capture around 28.2% (4M/14.2M) of WikiHow's keyword market in this category.
These estimates focus solely on informational and learning-oriented keywords, leaving room for even greater reach across other keyword categories.
The Content-Led Product Growth Loop: A Virtuous Cycle
One of the most compelling aspects of implementing these SEO strategies is the creation of a content-led product growth loop. Here's how it works:
Discovery via Search: Users looking for specific information find OpenAI's indexed conversations in their search results. This drives them to ChatGPT.
User Engagement: Some of these users will find the content so valuable or intriguing that they decide to continue the conversation on ChatGPT.
Content Creation: These newly engaged users then generate more conversations, some of which will be shared and made public if they pass OpenAI's moderation checks.
More Discovery: This new user-generated content becomes another asset for SEO, making it discoverable by a new wave of users through search engines.
Loop Continuation: The cycle repeats, each time adding more valuable content to ChatGPT and drawing in more users, who in turn create more content.
Challenges and Solutions
Challenge 1: Risk of Harmful Content
Allowing public conversations could lead to the indexing of harmful or inappropriate content.
Solution: OpenAI already has a robust moderation policy in place. Conversations that violate this policy could be automatically flagged and removed from public view. Additionally, OpenAI could disable the option to make a thread public for conversations that are flagged, ensuring that they never become accessible to search engines in the first place.
Challenge 2: Sensitive Topics like Healthcare
Users might seek medical advice or prescriptions, which could be risky.
Solution: OpenAI's moderation tools can identify such conversations and disable public sharing. A disclaimer could also be displayed, warning users not to act on the advice given.
Implementation Roadmap
Enable Public Sharing: Add an option to make conversations public after they pass a moderation check.
URL Optimization: Update the URL structure to include the conversation title and an ID. Implement permanent redirects for existing URLs.
User Engagement: (Stretch Goal) Develop a feed where users can see shared conversations, follow other users, and engage through likes and comments.
2. OpenAI Platform
According to Ahrefs, OpenAI's Platform receives approximately 1.4 million monthly visitors and ranks for 39.5K keywords. While these numbers are impressive, there's a significant opportunity to further optimize the platform's SEO, particularly the examples page featuring various use cases.
Recommendations for OpenAI
1. Optimize Use-Case Pages for SEO
OpenAI Platform has an examples page showcasing use cases like grammar correction, emoji translation, and parsing unstructured data. However, these examples open in modals, which is not SEO-friendly.
Solution:
Each example should be its own dedicated page with a well-optimized title and meta tags. This would not only improve the page's SEO but also provide a better user experience. The URL for this page could be “https://platform.openai.com/examples/grammar-correction”
2. Target High-Value Keywords for examples
OpenAI could benefit from targeting specific high-value keywords and increasing its catalog of examples or adding a new category of “use cases”. Some of the keywords that can be targetted
3. Broaden the Keyword Scope
Another untapped opportunity lies in targeting a broader category of keywords that start with "AI," such as:
By doing so, OpenAI could expand its reach to users who may not specifically be searching for OpenAI but are interested in AI-powered tools.
Implementation Roadmap
Page Optimization: Convert each use-case modal into a dedicated page with SEO-friendly titles and meta tags.
Keyword Targeting: Incorporate high-value and broad category keywords into these new pages to improve their search ranking.
3. OpenAI's Cookbook on GitHub
According to Ahrefs, OpenAI's Cookbook GitHub repository receives approximately 2,500 monthly visitors and ranks for 2,700 keywords. While this is a good start, there's a significant opportunity to optimize the repository, particularly the notebook names, to align with what users are actively searching for.
Suggestions for OpenAI
Optimizing Notebook Titles for User Intent
Consider OpenAI's notebook on DALL-E: Image_generations_edits_and_variations_with_DALL-E. Users are more commonly searching for terms like "how to use DALL-E." Renaming the notebook to align with this search intent could significantly improve its discoverability. A better name can be “How to use DallE for image generation”
Another example is the Clustering notebook. Users are often searching for "k-means clustering." A more SEO-friendly title could be "K means Clustering in Python using OpenAI"
Implementation Roadmap
Keyword Research: Conduct comprehensive keyword research to identify terms that users are actively searching for, and which are relevant to the notebooks.
Rename Notebooks: Update the titles of the iPython notebooks to match the broader search terms, thereby enhancing their discoverability through Google search.
4. ChatGPT Plugins
As it stands, ChatGPT plugins don't have dedicated detail pages. Instead, all plugins are accessible via modals in the logged-in view of ChatGPT. This approach misses out on valuable SEO opportunities, especially considering that users are actively searching for popular plugin names on Google.
Recommendations for OpenAI
1. Create SEO-Optimized Detail Pages for Plugins
Users are searching for specific plugin names, such as "Visla" with a search volume of 50K, and "ChatPDF" with a volume of 35K. Currently, these searches often lead to OpenAI's forum.
Solution: OpenAI should create dedicated, SEO-optimized detail pages for each plugin. These pages should include well-crafted titles and meta tags, along with comprehensive information about the plugin's features and usage.
An interesting observation here is that these plugins have already figured out a hack to get the distribution from Google. Both Visla and ChatPDF have a page on the OpenAI forum.
By doing so, OpenAI can capture more organic traffic and provide a better user experience.
Implementation Roadmap
Detail Page Creation: Develop dedicated detail pages for each plugin, complete with SEO-friendly titles and meta tags.
Keyword Optimization: Incorporate high-volume and relevant keywords into these new detail pages to improve their search ranking.
5. Model Detail & Comparison Page
Users often search for specific OpenAI models using their code names, such as "text-davinci-003." While ChatGPT does rank for these model names using query parameters, OpenAI is missing the opportunity to provide more detailed, user-friendly information through dedicated pages.
Recommendations for OpenAI
1. Create Dedicated Pages for Individual Models
Users are actively looking for information about specific models. OpenAI should create dedicated, SEO-optimized pages for each model code name.
Solution: These pages should include comprehensive details about the model, its capabilities, and use cases, all wrapped up with SEO-friendly titles and meta tags.
2. Develop Model Comparison Pages
Users are also interested in understanding the differences between various models, such as "GPT-3.5 vs. GPT-4."
Solution: OpenAI should create dedicated comparison pages that outline the features, capabilities, and differences between various models. These pages should also be SEO-optimized to capture high-value comparison keywords.
3. Preempt Competitor Strategies
Competitors could potentially capture traffic by creating their own comparison pages using keywords like "GPT-3.5 vs competitor-model-1" or “gpt-4 vs gpt-3.5 vs competitor-model-2”
Solution: OpenAI should preempt this by creating its own comparison pages that include potential competitors, thereby retaining the traffic and providing authoritative information.
Implementation Roadmap
Model Detail Pages: Create individual pages for each model, complete with SEO-friendly titles and meta tags.
Model Comparison Pages: Develop SEO-optimized pages that compare different OpenAI models, their use cases, and pricing as well as potential competitors.
6. ChatGPT Download Search Query
There are approximately 60,000 monthly searches for the term "ChatGPT download." Currently, the search engine results for this query are inconsistent—sometimes leading to the Play Store, sometimes to iOS, and sometimes to the web. This presents not only a missed SEO opportunity but also a potential security risk.
Recommendations for OpenAI
1. Create a Unified Download Page with Platform-Specific Deep Links
To cater to all user queries effectively, OpenAI should create a dedicated, SEO-optimized "Download ChatGPT" page. This page should feature deep links for each platform, ensuring that users are directed to the appropriate download location based on their device.
2. Address Unofficial Versions and Malicious Ads
There are websites capitalizing on this trend by offering unofficial versions of ChatGPT. Additionally, some ads are being run on the term "ChatGPT Download."
Solution: A dedicated download page from OpenAI would not only improve user experience but also act as a safeguard against unofficial or malicious versions. By ranking at the top for this query, OpenAI can effectively push down unofficial or malicious links.
Implementation Roadmap
Unified Download Page: Create an SEO-optimized download page featuring deep links for Android, iOS, and Web platforms.
Security Measures: Monitor and report unofficial versions and malicious ads to relevant authorities, while also educating users on the official download sources through the new download page.
7. Other Pages
OpenAI’s Dall E page gets a traffic of about 2.3M and ranks for 27.9K keywords. One suggestion here is to focus on the “ai image generator” keyword which has a monthly search volume of 393K. OpenAI could change its URL to “https://openai.com/dall-e-2-ai-image-generator”.
There are 8900 monthly searches for “Chatgpt wiki”. OpenAI already has a developer forum, where they can create a page targetting “chatgpt wiki” intent. This page can have a list of all the popular ChatGPT-related content.
Note:
All the keyword volume and traffic data are taken from Ahrefs.