What are the benefits and challenges of using AI and machine learning for SEO market planning?

One of the issues about working with big data has to do with data accuracy. For example, as we have all seen with ChatGPT output - it can sometimes be fantastically wrong. This has to do with an error or a series of errors in the data that GPT has been trained on. This focus on data accuracy is as true for purely numeric tools as it is for an AI chat tool. Where AI scores over traditional data gathering and analysis is the sheer scale of data that can be analysed and the many different ways that the data can be sliced and diced.

For the last two years, we have been using machine learning in our schema building and tracking the latest changes in Google's schema preferences. However, Generative AI has changed the game drastically. One not only has to continue with the SEO activities but also try to figure out Large Language Model Optimization (LLM), or as one of my developers put it - L2O, and apply that also. Fortunately, from what we see in this short time, there are many similarities between L2O and Focused SEO practices (generating lower but more focused traffic).

Content optimization particularly for meta data is a huge opportunity for SEOs to fix content to make it easier to search engines to crawl, index and rank content. AI created meta data benefits from best practices that the AI tool has been able to analyze and distill from thousands of websites. Before we get ahead of ourselves, we need to remember that in using AI generated meta data we are reverting to a cookie cutter output that the AI machine deems workable. What is lost here is not just originality, what is lost here is the SEO's ability to create genuine differentiation for a website. Expressed another way, the use of AI generated meta data eliminates the opportunity to create a signature tone or communication style for the website.

Focused SEO practices require you to optimize the content for the purchasing audience. This form of content optimization has forced us to feed the Generative AI with our data and content from many of our successful campaigns. In these campaigns, we improved the conversion by modeling the content exclusively for our client's buyers' persona. On the other hand, ChatGPT doesn't give us "selling" content because it is trained chiefly on wide casting (traffic maximizing) practices.

The benefits and methodology for creating good UX are well-known and utilized. In this case, AI will assist the many site owners who started their sites without any knowledge of UI/UX to improve the user experience. So, AI intervention should make Web 3.0 more pleasant to browse than Web 2.0 ever was.

There is an old folk tale regarding Rumi: when one of his children was "telling" him about what true love should be, he stopped the young man and said, "What I hear is talk; what I have is the knowledge of it." and then pointing to Shams' heart "and on the subject of love therein is the wisdom. I have long espoused the notion that all algorithms must be transparent to a great degree to ensure that the program will not cause harm. But, as we have done with nuclear energy, we should also create an international body to oversee AI and the code behind it, and how it builds on itself. After all, AI is far more powerful and far more transformative technology than nuclear power. After all, we have shown a lack of wisdom in the past.

A concern that I have with the use of AI tools for SEO is that this is rapidly becoming an expensive "arms race". The number of AI based SEO tools is expanding, and at an expanding rate. Implicit in their success is that the SEO has to use these tools or get left behind. The smart entrepreneurs behind these tools are also jacking up their tool costs by increasingly charging a per website or a per output rate. I fear that like in an arms race, the SEO with smaller resources is going to be left behind. I also see an impact in client costs or SEO profitability (or both) as the cost of doing SEO begins to skyrocket. This is inevitable as the number of AI tools needed and the cost of these tools, to do great SEO increases.

This is why my team built our internal generative AI engine to interact with our data. If anyone is concerned with losing their lead because The AI would share their latest SEO tips and tricks, they should set up a walled garden and only use that engine for work.