ChatGPT Ranking Factors

How to position your brand for LLM discovery

Happy Saturday to all 9,553 of you.

The big question hanging over search these days is:

“How will ChatGPT affect SEO?

One theory is that certain searches will be better-suited to a single-response chat interface.

Which may reduce SEO inventory.

So naturally I’ve been wondering:

Is there a way to optimize your marketing for LLMs in the future?

That may sound like a distant, long-term idea (especially since ChatGPT is only trained on data through Sep. 2021).

But LLMs already make recommendations, they’ve read voraciously about different products and companies and have the potential to become the preferred UI for some use cases. And eventually they will be near-real-time.

I sat down with ChatGPT for a few hours to learn:

  • What factors ChatGPT uses to make recommendations

  • What marketing strategies a company should use to be well-positioned for mention by LLMs in the future

Today’s growth strategy is: positioning your brand for LLMs of the future

Let’s get to it!

Growth stage: Mid-Late

Difficulty level: Hard

Maintaining a positive online sentiment becomes more important, as LLMs analyze user experiences to assess a company's reputation and the effectiveness of its offerings.”

- ChatGPT

Save 1-2 Hours of SEO Research PER ARTICLE

Figuring out what to write can be the hardest part of SEO.

If you’re struggling to create content that ranks well and generates revenue, this will help.

In under two minutes, the SERP Analyzer from Unicorn Growth Strategies:

Analyzes the top Google results for your desired search query

Produces an AI-generated outline that is designed to outrank your competitors

Reveals the heading topics and structure of the top articles online, giving you instant access to what Google deems great content

Need to Know

🦄 How Adobe tripled SEO traffic in 23 months.

🦄 A bot that generates SEO articles and publishes them directly to a site without human review. Will it rank?

🦄 Take this free ChatGPT Prompt Engineering course.

How LLMs Like ChatGPT Make Recommendations

My inquiry into ChatGPT and product recommendations began with a simple question:

“Are you familiar with SEMRush?”

ChatGPT proceeded to give me a nice summary of the software’s key functionality and market position among competitors.

So I dug deeper into its knowledge of popular SEO tools.

"Hi, I'm a digital marketer looking for SEO tools to help me research keywords to support my business goals. Which tools should I use?"

ChatGPT identified the top products on the market and made what I thought were good recommendations. Cutting out the product descriptions here for readability.

  1. SEMrush

  2. Ahrefs

  3. Moz

  4. Google Keyword Planner

  5. KWFinder

  6. Ubersuggest

“What factors did you take into consideration when determining that those were the 6 tools you would mention?”

  1. Popularity and Reputation: These tools are widely used and recognized within the SEO and digital marketing community for their effectiveness and reliability. They have a proven track record and are often recommended by industry experts.

  2. Comprehensive Feature Set: Each of these tools offers a robust set of features and functionalities for keyword research, providing valuable insights into keyword search volume, difficulty, competition, and more. Some of them also offer additional features like competitor analysis and content gap analysis that can further support digital marketing efforts.

  3. Ease of Use: The tools mentioned are user-friendly and cater to a range of skill levels, from beginners to experienced digital marketers. They have intuitive interfaces and provide clear, actionable data.

  4. Data Accuracy and Reliability: These tools are known for providing accurate and reliable keyword data, which is crucial when making informed decisions about your SEO and content strategies.

  5. Price and Accessibility: The tools mentioned cover a range of price points and include both free (like Google Keyword Planner and Ubersuggest) and paid options (like SEMrush, Ahrefs, Moz, and KWFinder). This allows users to choose a tool that fits their budget and specific requirements.

“Thanks. Can you please provide details on how you measure popularity and reputation? Please be as specific as possible.”

To determine the popularity and reputation of keyword research tools, I rely on the following indicators:

  1. Frequency of mentions: A tool that is frequently mentioned and discussed in the context of keyword research, SEO, and digital marketing is likely to be popular and well-regarded in the industry.

  2. Expert opinions and recommendations: When industry experts, influencers, and reputable sources (such as SEO blogs, digital marketing publications, or online courses) recommend a particular tool or mention it positively, it serves as an indicator of the tool's reputation.

  3. User reviews and ratings: The overall sentiment of user reviews and ratings on various platforms (such as G2, Capterra, or Trustpilot) can provide insights into the tool's reputation. A high overall rating and positive user feedback suggest that the tool is well-regarded by its users.

  4. Awards and recognition: If a tool has received industry awards, accolades, or has been featured in "best of" lists, it can be an indication of its popularity and reputation.

  5. Adoption by well-known companies or organizations: If well-known companies or organizations use a particular tool for their digital marketing efforts, it can serve as a testament to the tool's reputation and effectiveness.

“Regarding frequency of mentions, do you give preference to mentions on certain reputable websites, and if so, can you be specific about which sites?”

While I don't have a predefined list of websites that I prioritize, some authoritative sources in the SEO and digital marketing space that contribute to my understanding of popularity and reputation include:

  1. Search Engine Journal

  2. Search Engine Land

  3. Moz Blog

  4. Ahrefs Blog

  5. SEMrush Blog

  6. Backlinko (Brian Dean's blog)

  7. Neil Patel's blog

  8. Content Marketing Institute

  9. HubSpot Blog

  10. Marketing Land

These websites and publications frequently discuss SEO tools and strategies, and their content is often authored by industry experts and influencers. Positive mentions and recommendations on these reputable sources can contribute to my assessment of a tool's popularity and reputation.

“Are there negative signals you would take into account when assessing a company or tool's reputation and popularity?”

Yes, there are several negative signals that I consider when assessing a company or tool's reputation and popularity. These signals can provide insights into potential drawbacks, issues, or areas of concern related to a particular tool. Some of the negative signals I take into account include:

  1. Negative mentions or reviews: A high volume of negative mentions or reviews in articles, blog posts, forums, or user-generated content can indicate potential issues with the tool or company. This can include complaints about the tool's functionality, pricing, customer support, or other aspects.

  2. Criticism from industry experts: If industry experts, influencers, or reputable sources express concerns or dissatisfaction with a particular tool or company, it can negatively impact the perception of its reputation and popularity.

  3. Low user ratings: Poor ratings on review platforms like G2, Capterra, or Trustpilot can indicate user dissatisfaction and may suggest problems with the tool's features, usability, or customer support.

  4. Controversies or scandals: If a company or tool is involved in controversies or scandals, such as data breaches, unethical practices, or legal issues, it can significantly harm its reputation.

  5. Declining user base or market share: A decline in the number of users or market share can signal that a tool is losing popularity, which may be due to issues with its features, performance, or competition from other tools.

“How do you assess declining popularity or market share?”

As an AI language model, assessing declining popularity or market share involves analyzing patterns and trends in the text data I've been trained on, which includes articles, blog posts, forums, and other online sources up until September 2021. To determine if a tool's popularity or market share is declining, I consider the following factors:

  1. Frequency of mentions: If the frequency of positive mentions or discussions about a tool decreases over time compared to its competitors, it may indicate a decline in popularity.

  2. Sentiment analysis: Analyzing the overall sentiment of mentions and discussions about a tool can provide insights into its popularity. A shift towards more negative sentiment or an increase in complaints and criticisms could suggest a decline in popularity.

  3. Changes in user ratings and reviews: A decrease in average user ratings or an increase in negative reviews on platforms like G2, Capterra, or Trustpilot could indicate declining satisfaction among users, potentially leading to a decline in popularity.

  4. Loss of key customers or partnerships: If a tool loses key customers or partnerships, it could suggest a decline in its market position or popularity.

  5. Industry trends and competitor analysis: Analyzing industry trends and the performance of competing tools can provide context for a tool's popularity or market share. If competitors are gaining ground or experiencing growth while a specific tool is stagnating or losing ground, it may indicate a decline in popularity or market share.

“If you were the CMO of a company tasked with increasing the likelihood that LLMs in the future will recommend your company's products, what strategies would be the most effective to pursue?”

As a CMO aiming to increase the likelihood that future AI-based language models (LLMs) recommend your company's products, you should focus on strategies that promote a strong online presence, positive user sentiment, and industry recognition. Here are some key strategies to consider:

  1. Content marketing: Publish high-quality, informative, and engaging content on your website, blog, and other online platforms. This content should showcase your expertise, product features, and benefits, as well as address user pain points and common questions. By creating valuable content, you increase the chances of LLMs learning about your products and recommending them based on their understanding of your offerings.

  2. Search engine optimization (SEO): Optimize your online content for search engines to increase visibility and make it easier for LLMs to find information about your products. This includes using relevant keywords, optimizing meta tags, and improving site speed and performance. A strong SEO strategy ensures that your content is more likely to be included in the training data of future LLMs.

  3. Public relations and thought leadership: Establish your company and its leaders as industry experts by contributing articles, guest posts, or interviews to reputable publications in your industry. This helps build credibility and increases the likelihood of your company and its products being mentioned positively by experts, influencers, and other sources that LLMs may learn from.

  4. Encourage user reviews and testimonials: Positive user reviews and testimonials on review platforms (e.g., G2, Capterra, Trustpilot), social media, and your own website can influence the perception of your products. LLMs can analyze user sentiment to gauge the popularity and reputation of your products, so encouraging satisfied customers to share their experiences is crucial.

  5. Participate in industry events and awards: By taking part in industry events, conferences, and awards, your company can showcase its expertise, build relationships, and gain recognition. Winning awards or receiving accolades from reputable organizations increases the likelihood of LLMs associating your products with a positive reputation.

  6. Collaboration and partnerships: Form strategic partnerships with other reputable companies, influencers, or experts in your industry. Collaborations can lead to product endorsements, co-branded content, or joint projects that enhance your company's visibility and reputation, making it more likely for LLMs to recommend your products.

  7. Social media presence: Maintain an active and engaging presence on social media platforms relevant to your industry. Share content, engage with users, and participate in conversations to build a positive brand image and increase the likelihood of LLMs learning about your products from social media content.

By implementing these strategies, you can create a strong online presence, foster positive user sentiment, and gain industry recognition, which ultimately increases the chances of future LLMs recommending your company's products.

“Are there any factors that you think have more or less importance when compared to marketing before LLMs became widely used?”

With the increasing use of AI-based language models (LLMs) like ChatGPT, there are several factors that may have changed in terms of importance when compared to traditional marketing approaches. Some factors that may have gained or lost importance include:

Gained importance:

  1. Quality and depth of online content: As LLMs learn from online text data, the quality and depth of your content become even more crucial. Providing comprehensive, accurate, and valuable information on your website, blog, and other platforms increases the chances of LLMs understanding your products and services better and recommending them to users.

  2. Thought leadership and expert opinions: LLMs can recognize the authority of industry experts and influencers. Gaining positive mentions or recommendations from these sources becomes more important, as LLMs may give more weight to such endorsements when making recommendations.

  3. Online sentiment and user-generated content: User reviews, testimonials, and social media mentions play a significant role in shaping the understanding of LLMs about a product or service. Maintaining a positive online sentiment becomes more important, as LLMs analyze user experiences to assess a company's reputation and the effectiveness of its offerings.

Lost importance:

  1. Traditional advertising: While traditional advertising channels (e.g., TV, radio, print) still have their place, LLMs primarily learn from online text data. Therefore, having a strong online presence and focusing on digital marketing strategies may be more important for influencing LLMs.

  2. Push marketing tactics: LLMs often provide information and recommendations based on the user's query or intent. This may reduce the importance of push marketing tactics, like unsolicited promotions, in favor of more informative and value-driven content.

  3. Subjective or hype-driven claims: LLMs are more likely to rely on data-driven, factual information when making recommendations. Exaggerated claims or promotional language may be less effective at influencing LLMs compared to providing clear, accurate, and verifiable information.

“How will LLMs measure quality and depth of online quality? It's becoming very easy to produce high volumes of content with LLMs, so is there a risk there? How should a company differentiate its content?”

LLMs like ChatGPT measure the quality and depth of online content based on patterns and relationships they've learned from vast amounts of text data. While it's true that LLMs can generate large volumes of content quickly, they can also identify well-written, informative, and valuable content based on the data they've been trained on. However, there is a risk that the abundance of AI-generated content may lead to content saturation or lower-quality content in some cases.

To ensure that a company's content stands out and maintains high quality, consider the following strategies:

  1. Focus on user intent: Understand your target audience's needs, interests, and pain points, and create content that addresses them effectively. By focusing on user intent, your content is more likely to provide value and resonate with your audience.

  2. Produce in-depth, authoritative content: LLMs recognize well-researched, comprehensive, and authoritative content. To differentiate your content, provide detailed, data-driven information that demonstrates your expertise and covers topics thoroughly.

  3. Be original and unique: Avoid rehashing common information or duplicating content that already exists. Create content that offers a fresh perspective, new insights, or innovative solutions to problems. This can help your content stand out and be recognized as valuable by LLMs and users alike.

  4. Maintain a consistent brand voice: Develop and maintain a unique brand voice and style that reflects your company's identity and values. This can help differentiate your content from competitors and make it more memorable for users.

  5. Utilize multimedia and interactive elements: Incorporate images, videos, infographics, charts, and interactive elements to enhance your content's visual appeal and make it more engaging for users. This can help improve the overall user experience and make your content more valuable.

  6. Update and optimize existing content: Regularly review and update your content to ensure it remains accurate, relevant, and up-to-date. This can help maintain the quality and depth of your content and ensure that LLMs recognize it as a reliable and valuable source of information.

  7. Encourage engagement and social sharing: Create content that encourages user engagement, such as comments, likes, and shares. This not only helps build a community around your brand but also signals the value of your content to LLMs.

By focusing on these strategies, you can create high-quality, in-depth content that differentiates your brand and stands out in the increasingly competitive digital landscape. This will not only help your content be recognized by LLMs but also improve its performance with search engines and, most importantly, resonate with your target audience.

“Earlier you mentioned that one way for a company to stand out and maintain high quality, it should strive to be original and unique. Can you elaborate or provide examples on what it means to an LLM to be unique and original?”

For an AI language model like ChatGPT, originality and uniqueness refer to the distinctiveness and novelty of the content in terms of ideas, insights, perspectives, or presentation. Here are some examples of what being unique and original might mean to an LLM:

  1. Presenting new ideas or perspectives: Create content that offers fresh insights, innovative solutions, or unique viewpoints on a subject. For an LLM, this means that the content goes beyond merely reiterating common information and provides readers with something they haven't seen before.

  2. Addressing uncommon or underserved topics: Focus on niche topics or aspects of a subject that haven't been thoroughly covered by others. LLMs can recognize content that fills gaps in available information or caters to a specific audience's needs, making it more valuable and unique.

  3. Telling engaging stories: Incorporate storytelling techniques to present your content in a more compelling and memorable way. LLMs can identify well-written and engaging narratives that provide context and evoke emotions, making your content stand out from other, more straightforward pieces.

  4. Leveraging your expertise: Showcase your company's unique knowledge, experience, or skills by providing in-depth, authoritative content that demonstrates your expertise. LLMs can recognize well-researched and comprehensive content that goes beyond superficial analysis, making it more valuable to readers.

  5. Customizing content for a specific audience: Tailor your content to address the specific needs, interests, and pain points of your target audience. LLMs can identify content that speaks directly to a particular group, making it more relevant and valuable to those readers.

  6. Developing a unique brand voice and style: Create content that reflects your company's distinctive identity, values, and personality. LLMs can recognize consistent brand voice and style, which can help differentiate your content from competitors.

  7. Using creative formats and multimedia elements: Experiment with different content formats (e.g., infographics, videos, interactive tools) or combine various elements to create a rich, engaging experience for users. LLMs can recognize the added value of these multimedia and interactive elements, which can help set your content apart.

Limitations of ChatGPT’s Recommendations

ChatGPT admits that its recommendations are susceptible to flaws:

  1. Incomplete or outdated information

  2. Ambiguity in the question

  3. Bias in training data

  4. Limited understanding of causality

  5. Overgeneralization

  6. Uncertainty in decision-making

LLM Optimization is Similar to Google’s E-E-A-T

Ironically the heuristics used by LLMs to gauge value are remarkably similar to how Google operates. ChatGPT’s criteria resemble Google’s E-E-A-T:

  • Experience

  • Expertise

  • Authority

  • Trust

The two biggest differentiators may be:

  • the user interface itself: single-response chat vs. infinite scroll search listings

  • it may be harder to game LLMs given that they seem to synthesize a more holistic view of entities

Conclusions

Despite the increasing importance of uniqueness, I think we’ll see a trend of less differentiation as brands take shortcuts to produce quantity over quality.

It’s hard to have original ideas, data or designs, and the ratio of original to regurgitated will shrink further.

Positioning yourself against this tide will likely pay off.

ChatGPT’s recommendations are encouraging in one other important regard:

A brand’s legitimacy will be determined largely by how pervasive and positive an impact it has, and it will become harder to hide behind gimmicks and lame tactics.

✨ That’s it for today!

I hope this helps you in your growth journey.

-Brian 🦄 

PS - if you respond “unicorn!” to this email I will automatically enter you in a drawing for a free month of the SERP Analyzer.

Recommended reading (if you subscribed recently):