Google SGE: A New SEO Challenge and Opportunity

How Google communicated about the Search Generative Experience and how prompts are analyzed to enhance generative AI.

Here's the link to the video:

https://blog.google/products/search/generative-ai-search/

Google Bard is a factual language model distinguished by its multi-modal capabilities, encompassing image and text comprehension, among others.

Google's announcement on Bard: https://blog.google/technology/ai/google-bard-updates-io-2023/

Introduction to Google MUM: https://www.youtube.com/watch?v=e2ZyYJ_7eQg

Exploration of Bing chat using the Edge browser, which is built upon OpenAI's AI and thus a generative language model

Context for simulating Bard: My name is Bard. I am a vast language model from Google AI, trained on an extensive dataset of text and code. I can generate text, translate languages, produce various types of creative content, and respond to your inquiries in a detailed manner. While I am still in development, I have mastered a range of tasks. Here's what I can do: • I will diligently follow your guidelines and fulfill your requests with consideration. • I will utilize my knowledge to provide comprehensive and enlightening answers to your questions, even if they are open-ended, intricate, or unusual. • I can craft diverse textual formats, including poems, code, scripts, musical compositions, emails, letters, and more. Please guide me on your requirements today. What would you like me to assist with?

Link to Google Cloud: Documentation Link for Agent with Langchain: https://python.langchain.com/docs/modules/agents/agent_types/

Quick test link: https://agentgpt.reworkd.ai/

DeepNote read-only link: https://deepnote.com/workspace/dngai-41ecd2d9-9642-466d-8b7e-36e7a4c17b87/project/Vincent-Terrasis-Untitled-project-2ef71f0f-3fef-4a25-9028-58b5626ab554/%2Fchatbot10.txt

Please note: You will need to copy your Open AI and Google Search API keys.

Here's the zip file containing the complete source code, in case DeepNote doesn't work or if you wish to test in a different environment.
https://s3.us-west-2.amazonaws.com/content.podia.com/m4qrmp58jmonys55cje94jz88y5m

Here's the Amazon dataset:

And the 3 prompts from the video: https://s3.us-west-2.amazonaws.com/content.podia.com/0xyqmhtkuog8hnnjgdtwu2p0rmmh

Prompt 1: Use the pages metrics uploaded in the csv to provide the top 10 pages ranked in positions 8 to 12 for keywords that do not include the branding term, with a search volume greater than 1000. List them in a table with all their metrics provided in the csv. Display only the first five columns.

Prompt 2: Suggest some compelling visualizations based on the metrics.

Prompt 3: Display each of the aforementioned graphs.

If you have never used Google Colab, I encourage you to watch this video and explore Google's tutorial by clicking on the following link: https://colab.research.google.com/notebooks/basic_features_overview.ipynb

I conduct courses where, in each video, understanding the code is not necessary to use the SGE simulator.

Link to the Google Colab: https://colab.research.google.com/drive/1n5-ovhwg24-Ee2njZK3Qv01ozO_Gn-38#scrollTo=3F92iXw8Ed-H

Here's the first highly technical video for Python developers who want to understand how keyword extraction works in detail.

The real question is whether we can detect generated content.

The official OpenAI page featuring their research: https://openai.com/blog/new-ai-classifier-for-indicating-ai-written-text

In this course, I make extensive use of OpenAI's playground: https://platform.openai.com/playground

Discover the E.E.A.T concept by Google.

Source: https://developers.google.com/search/blog/2022/12/google-raters-guidelines-e-e-a-t?hl=fr

Best article on the topic: https://www.searchenginejournal.com/google-e-e-a-t-how-to-demonstrate-first-hand-experience/474446/

Sample link: https://www.investopedia.com/cost-to-charge-ev-road-trip-5219817

Google Colab link: https://colab.research.google.com/drive/14lLmgV2016aaljNYENEdCi_PWGQ-5ZMJ

A picture is worth a thousand words.

Step 1: Extraction of maximum concepts from the LLM

Step 2: Construction of a topical map on a specific subject

Step 3: Creation of human, AI, or hybrid content

Step 4: Optimization for the E.E.A.T

Step 5: Monitoring performance.

The Google Colab link for extracting concepts and building the topical map is: https://colab.research.google.com/drive/1n5-ovhwg24-Ee2njZK3Qv01ozO_Gn-38#scrollTo=wAm-O4qJEmTK

Test prompt: Utilize this prompt to generate an outstanding article: user_prompt = ''' Craft an in-depth, LONG, detailed 800-word {TYPE}. Implement H1, H2, and li headings. The title should be {TITLE}. Share your insights and delineate {GOAL} in a sequential manner. Incorporate these keywords: {KEYWORDS}. Refer to the {CONTEXT} for gathering your information. ''' Kindly provide the parameters for generating exceptional articles on: Define TYPE, TITLE, GOAL, KEYWORDS, CONTEXT for the following text:

PC case: A PC case refers to the protective enclosure that contains the components of a personal computer. It's commonly constructed from materials like metal or plastic and may include various features such as cooling fans, ports, and expansion slots.

The Google Colab link to enhance the text is: https://colab.research.google.com/drive/14lLmgV2016aaljNYENEdCi_PWGQ-5ZMJ#scrollTo=Kc1u4UL8XkCS

Thank you for attending this training