𧩠#8-Google Search Generative Experience is here... to stay!!
Nov 28, 2023πTHE news that needs your attention
AAlphabet 2023 - Q2 Earning Call from July 25th, 2023. With Sundar Pichai, Philipp Schmidt, and Ruth Porat.
The short of it: Google reminds its shareholders that they are in their seventh year as an “AI-first company” and emphasizes its continued leadership in AI and excitement about the evolution of Search. This evolution is characterized by Google Search Generative Experience (“SGE”).
During the call, someone at Morgan Stanley asked Sundar Pichaï “How he thinks about the timing for more broadly integrating generative AI into Search?”
Sundar Pichai's answer was crystal clear:
“We've constantly been bringing in AI innovations into Search for the past few years. And this is the next step in the journey. But it is a big change (...) it really gives us a chance to now not always be constrained in the way search was working before, allowing us to think outside the box. And so you will see us continue to bring it to more and more users, and over time, this will just be how search works. While we are taking deliberate steps, we are building the next major evolution in Search”.
A few days later, on August 4th, after several months of beta testing with a select number of SEO experts since their Google I/O event, Google opened its new Search Product, the Search Generative Experience, to anyone in the United States with a Gmail account (π here).
SGE is here to stay.
π A few takeaways to keep in mind:
-
Google as an “AI-First Company” …: On Nov. 30th, 2022, OpenAI essentially stole Google's thunder on the broader "AI topic." Not only did they offer the first serious alternative to Search in the past 20 years of Google's dominance, but they also took a pioneering stance on the topic of "AI open to everyone" (aka AI in Open Source).
These “two legs”- “Chat” from a user's standpoint and “Open Source” from a developer's standpoint—are now shaky, and Google intends to address that ASAP. It’s likely no coincidence that Sergei Brin has returned to Google’s business operations to work on Gemini which is presented as Google’s own groundbreaking LLM (π here)
-
Making “AI helpful to everyone”: Google established 4 strategies to implement this “vision” (e.i., to win the AI + Chat war). In other words, 4 battlefronts.
-
Product Enhancement: They’re integrating AI into all of their products, starting with a focus on Search.
-
Fostering Creativity and Productivity: They're venturing into conversational AI with "Bard" to compete against tools like ChatGPT, Claude 2, Perplexity AI, and Lama2.
-
Facilitating AI Innovation: By leveraging Google Cloud's advanced infrastructure for AI, they aim to empower developers and reclaim the lead in the open-source space, challenging platforms like Hugging Face and GitHub (remember the “we have no moats” story - π here).
-
Ethical AI Development: Google partners with OpenAI, Anthropic, and Microsoft on the Frontier Model Forum to champion responsible AI practices, marketing themselves as the web's most ethically conscientious data scraper π…(π Here).
-
Incorporating AI into all their products … especially in Search: Google will NOT leave the playing field open for Open AI and its “Chat”. Google intends to remain the getaway to the internet: “SGE is conceived as a new experience, another jumping off point for exploring the web”. Period.
Sources: Listen to the YouTube video (π here).
π§ What does that mean for us, humble humans?
Google is navigating a fine balance, striving to both maintain Search as the primary gateway to the internet and introduce a Chat feature that bridges “them” with us.
π Name it: Google Search Generative Experience.
Google calls SGE, “Search supercharged” (supercharge by generative AI). What does that mean? Few important things:
-
AI-generated mashed-up content: SGE results pages are now officially made of amalgamated content.
-
Google’s own answer: When you “Google” something, instead of a single link to a website page that likely answers your query, Google now presents an answer, its own answer. Pretty well summarized by Joe Pulizzi (πhere).
-
Citation or no citation? This answer will be made of mashed-up content, compiled from different web pages and sources Google is scraping (the entire idea of Generative AI) but… those sources no longer need to be cited nor will Google probably cite them. The current testing of SGE shows that sometimes they do cite - something they don’t (πhere).
-
So what’s the BIG change? Instead of using AI to improve their Search Engine to better detect our search intent and provide us with the best link to the page that will answer our query - they are now merging Generative AI into their Search Engine to provide us with their own answer.
-
Top-of-the-funnel content is no longer a brand opportunity: Google will now be taking it for itself leaving SEO to mid-funnel. That means your content should focus mostly on 2 things: your personal brand, even if you’re an employee, AND your process, your best practices, your methodology, your “own stuff” - whatever makes you different in the way you do your job.
Conclusion: Bad and good news.
The bad one: SEOs and content marketers who keep having as a marketing goal to generate traffic through backlinks to a page optimized around keywords will have to rethink their job.
The good one: Those “answers” could contain your brand… And actually, when they do, the visibility offered at the top of SERPs is pretty outstanding. Kind of like online billboard visibility on top of Google Results.
π― THE insight from Linkedin
This (as always) deliciously bright post from Ethan Mollick.
The short of it: “Too many organizations think that having LLMs work with their data is the key to making AI valuable. And so they build chatbots that can answer questions about previous work, like some sort of internal Google Search (…). The real key is finding prompting approaches that let AI leverage their best practices”.
Sources: Check the full post (π here).
π§ What does that mean for us, humble humans?
Connect Ethan Mollick’s post with the rise of SGE and think of the following:
-
Your unique value proposition: If web content associating your brand with your main topic is widespread, Google will prominently feature you in their AI-generated answers and allocate significant space for you in their “AI-generated box". Ask yourself: how can you achieve “widespread” recognition in an AI-relevant way? How do you plan to leverage AI to provide unique value to your customers based on your “only I”?
-
Think processes, methodologies, framework: To make sense of this, meditate on how AI can enhance or scale the methodologies, processes, and unique positioning that genuinely define you, not just the content that claims so. Will you use AI in the same ways that many others are increasingly adopting, like creating a chatbot that acts as a personal Google search? Or will you use AI to enhance your processes, methodologies, and unique value propositions, and then create content that reflects those personal-branded values as Kahn Academy did?
-
Create content about this and repurpose it like crazy. Search, supercharged by AI, pushes SEO to be more social than ever. It’s now a fact. The good news is that AI is making it easier than ever to repurpose content for all social media platforms. Make good use of it.
-
Replace SEO with AIO: Artificial Intelligence Optimization. Get used to it… We are creating content to now be optimized for artificial intelligence mostly: a new kind of Search Engine supercharged by AI. And with Google being more powerful than ever, you probably won’t be able to avoid it (π here).
π‘ Some examples and ideas to do that:
-
AI Personal Assistant: Use AI to create your personal assistant and market it. The hype around this idea of a personal assistant is far from being abandoned. I already talked about it (π here). It’s actually hotter than ever and easier and easier to get your own.
-
Avatar: Create an avatar for “him/her” or him/her and you as a team.
-
“Content about your AI processes”: Use AI to help you set up processes you imagine and create content about it.
Example: Andy Crestodina is doing a fantastic job at exactly this. He wrote one of the most famous books about Content Marketing and he is revisiting each of the tactics he developed in his books by using AI. Doing so, he is creating new processes and methodologies for very traditional digital marketing tactics like this incredible post about “AI-powered web page audit and optimization: SEO and conversion in 8 prompts” (π here).
-
Personal branding: Focus on building your personal brand. Whether you’re an employee or not, shouldn’t prevent you from doing so. It’s actually the opposite.
Example: Amanda Natividad did just that while being an employee at Sparktoro, Rand Fishkin’s new start-up. And guess what? She created great content about it (π here).
𧩠AI Marketing Corner
Topic: Personal branding, AI-personal assistant, Avatars
π€ The 1 tactic (from me):
Don’t mix up the good AI and the bad one π.
Don’t use AI to create “Average Informative” content about your product (BAD AI). Use AI to demonstrate how “Awesomely Innovative” you are in whatever is it you doing for a living and in the processes you imagined to be good at it (GOOD AI).
-
Be sure of your topic. List carefully your subtopics.
-
Pick one, focus like hell on it, and master it like no one else.
-
Create your “speech/process/methods” on it.
-
Market it like crazy everywhere.
-
Find the right AI tools to help you both automatize the process and/or create a personal assistant for it.
Few AI tools to help you test that ππ.
π 4 AI tools for your marketing:
Some ideas of tools to help you test and try what’s described above
-
Doppl- Create your own personal digital Artefact. It helps you create the most accurate AI simulation of you. I haven’t tested it yet. On the waitlist. But definitely a must-try (π here).
-
FineTuner.ai- Platform to help you build AI Agents at Scale with no coding skills. An all-in-one solution to build sophisticated AI agents, fast-tracking your path to market, and scaling your intelligent solutions without any coding (π here)
-
AskGenie- Create your own ChatGPT agent trained on your own data (π here).
-
Avaturn– Turn people into realistic avatars (π here).
π The 1 content that is worth your reading
This article from Venture Beats features an interview with the Analyst and Start-up investor, Jeremiah Owyang: Forget SEO, why AI Engine Optimization might be the future (π here).
“Unlike SEO, AI Engine Optimization is not about waiting for a crawler to come to a website. Now, marketers will likely want two things (…): One is to create an API that feeds in real-time information to foundational models.
Secondly, marketers will want to take the same corporate API with all of its product information and use it to train its own branded AI that would interact with consumers and buyers, whether that’s on a website or an app. (…) That would also interact with the buyer side agents that are starting to emerge”.
π The 1 LinkedIn Live event you don’t want to miss!!
To talk about all the major changes coming with Google SGE, I partnered with Andy Crestodina to launch monthly Live Discussions. Each month, we will be hosting live discussions with SEO and Content Marketers experts.
This coming August 28th we are hosting a fantastic panel with Aleyda Solís and Barry Schwartz.
You can listen to the entire conversation π π here
