Sign up to save your podcasts
Or
Share In-Ear Insights: Do Websites Matter in the Age of AI?
Share to email
Share to Facebook
Share to X
Share to email
Share to Facebook
Share to X
Or
In-Ear Insights: Do Websites Matter in the Age of AI?
In this episode of In-Ear Insights, the Trust Insights podcast, Katie and Chris discuss whether blogs and websites still matter in the age of generative AI.
You’ll learn why traditional content and SEO remain essential for your online presence, even with the rise of AI. You’ll discover how to effectively adapt your content strategy so that AI models can easily find and use your information. You’ll understand why focusing on answering your customer’s questions will benefit both human and AI search. You’ll gain practical tips for optimizing your content for “Search Everywhere” to maximize your visibility across all platforms. Tune in now to ensure your content strategy is future-proof!
Watch the video here:
Can’t see anything? Watch it on YouTube here.
Listen to the audio here:
https://traffic.libsyn.com/inearinsights/tipodcast-do-websites-matter-in-the-age-of-ai.mp3
Download the MP3 audio here.
[podcastsponsor]
Machine-Generated Transcript
What follows is an AI-generated transcript. The transcript may contain errors and is not a substitute for listening to the episode.
Christopher S. Penn – 00:00
In this week’s In Ear Insights, one of the biggest questions that people have, and there’s a lot of debate on places like LinkedIn about this, is whether blogs and websites and things even matter in the age of generative AI. There are two different positions on this. The first is saying, no, it doesn’t matter. You just need to be everywhere. You need to be doing podcasts and YouTube and stuff like that, as we are now. The second is the classic, don’t build on rented land. They have a place that you can call your own and things. So I have opinions on this, but Katie, I want to hear your opinions on this.
Katie Robbert – 00:37
I think we are in some ways overestimating people’s reliance on using AI for fact-finding missions. I think that a lot of people are turning to generative AI for, tell me the best agency in Boston or tell me the top five list versus the way that it was working previous to that, which is they would go to a search bar and do that instead. I think we’re overestimating the amount of people who actually do that.
Katie Robbert – 01:06
Given, when we talk to people, a lot of them are still using generative AI for the basics—to write a blog post or something like that. I think personally, I could be mistaken, but I feel pretty confident in my opinion that people are still looking for websites.
Katie Robbert – 01:33
People are still looking for thought leadership in the form of a blog post or a LinkedIn post that’s been repurposed from a blog post. People are still looking for that original content. I feel like it does go hand in hand with AI because if you allow the models to scrape your assets, it will show up in those searches. So I guess I think you still need it. I think people are still going to look at those sources. You also want it to be available for the models to be searching.
Christopher S. Penn – 02:09
And this is where folks who know the systems generally land. When you look at a ChatGPT or a Gemini or a Claude or a Deep Seat, what’s the first thing that happens when a model is uncertain? It fires up a web search. That web search is traditional old school SEO. I love the content saying, SEO doesn’t matter anymore. Well, no, it still matters quite a bit because the web search tools are relying on the, what, 30 years of website catalog data that we have to find truthful answers.
Christopher S. Penn – 02:51
Because AI companies have realized people actually do want some level of accuracy when they ask AI a question. Weird, huh? It really is. So with these tools, we have to. It is almost like you said, you have to do both. You do have to be everywhere.
Christopher S. Penn – 03:07
You do have to have content on YouTube, you do have to post on LinkedIn, but you also do have to have a place where people can actually buy something. Because if you don’t, well.
Katie Robbert – 03:18
And it’s interesting because if we say it in those terms, nothing’s changed. AI has not changed anything about our content dissemination strategy, about how we are getting ourselves out there. If anything, it’s just created a new channel for you to show up in. But all of the other channels still matter and you still have to start at the beginning of creating the content because you’re not. People like to think that, well, I have the idea in my head, so AI must know about it. It doesn’t work that way.
Katie Robbert – 03:52
You still have to take the time to create it and put it somewhere. You are not feeding it at this time directly into OpenAI’s model. You’re not logging into OpenAI saying, here’s all the information about me.
Katie Robbert – 04:10
So that when somebody asks, this is what you serve it up. No, it’s going to your website, it’s going to your blog post, it’s going to your social profiles, it’s going to wherever it is on the Internet that it chooses to pull information from. So your best bet is to keep doing what you’re doing in terms of your content marketing strategy, and AI is going to pick it up from there.
Christopher S. Penn – 04:33
Mm. A lot of folks are talking, understandably, about how agentic AI functions and how agentic buying will be a thing. And that is true. It will be at some point. It is not today. One thing you said, which I think has an asterisk around it, is, yes, our strategy at Trust Insights hasn’t really changed because we’ve been doing the “be everywhere” thing for a very long time.
Christopher S. Penn – 05:03
Since the inception of the company, we’ve had a podcast and a YouTube channel and a newsletter and this and that. I can see for legacy companies that were still practicing, 2010 SEO—just build it and they will come, build it and Google will send people your way—yeah, you do need an update.
Katie Robbert – 05:26
But AI isn’t the reason. AI is—you can use AI as a reason, but it’s not the reason that your strategy needs to be updated. So I think it’s worth at least acknowledging this whole conversation about SEO versus AEO versus Giao Odo. Whatever it is, at the end of the day, you’re still doing, quote unquote, traditional SEO and the models are just picking up whatever you’re putting out there. So you can optimize it for AI, but you still have to optimize it for the humans.
Christopher S. Penn – 06:09
Yep. My favorite expression is from Ashley Liddell at Deviate, who’s an SEO shop. She said SEO now just stands for Search Everywhere Optimization. Everything has a search. TikTok has a search. Pinterest has a search. You have to be everywhere and then you have to optimize for it. I think that’s the smartest way to think about this, to say, yeah, where is your customer and are you optimizing for?
Christopher S. Penn – 06:44
One of the things that we do a lot, and this is from the heyday of our web analytics era, before the AI era, go into your Google Analytics, go into referring source sites, referring URLs, and look where you’re getting traffic from, particularly look where you’re getting traffic from for places that you’re not trying particularly hard.
Christopher S. Penn – 07:00
So one place, for example, that I occasionally see in my own personal website that I have, to my knowledge, not done anything on, for quite some time, like decades or years, is Pinterest. Every now and again I get some rando from Pinterest coming. So look at those referring URLs and say, where else are we getting traffic from? Maybe there’s a there. If we’re getting traffic from and we’re not trying at all, maybe there’s a there for us to try something out there.
Katie Robbert – 07:33
I think that’s a really good pro tip because it seems like what’s been happening is companies have been so focused on how do we show up in AI that they’re forgetting that all of these other things have not gone away and the people who haven’t forgotten about them are going to capitalize on it and take that digital footprint and take that market share. While you were over here worried about how am I going to show up as the first agency in Boston in the OpenAI search, you still have—so I guess to your question, where you originally asked, is, do we still need to think about websites and blogs and that kind of content dissemination? Absolutely. If we’re really thinking about it, we need to consider it even more.
Katie Robbert – 08:30
We need to think about longer-form content. We need to think about content that is really impactful and what is it? The three E’s—to entertain, educate, and engage. Even more so now because if you are creating one or two sentence blurbs and putting that up on your website, that’s what these models are going to pick up and that’s it. So if you’re like, why is there not a more expansive explanation as to who I am? That’s because you didn’t put it out there.
Christopher S. Penn – 09:10
Exactly. We were just doing a project for a client and were analyzing content on their website and I kid you not, one page had 12 words on it. So no AI tool is going to synthesize about you. It’s just going to say, wow, this sucks and not bother referring to you.
Katie Robbert – 09:37
Is it fair to say that AI is a bit of a distraction when it comes to a content marketing strategy? Maybe this is just me, but the way that I would approach it is I would take AI out of the conversation altogether just for the time being. In terms of what content do we want to create? Who do we want to reach? Then I would insert AI back in when we’re talking about what channels do we want to appear on? Because I’m really thinking about AI search. For a lack of a better term, it’s just another channel.
Katie Robbert – 10:14
So if I think of my attribution modeling and if I think of what that looks like, I would expect maybe AI shows up as a first touch.
Katie Robbert – 10:31
Maybe somebody was doing some research and it’s part of my first touch attribution. But then they’re like, oh, that’s interesting. I want to go learn more. Let me go find their social profiles. That’s going to be a second touch. That’s going to be sort of the middle. Then they’re like, okay, now I’m ready. So they’re going to go to the website. That’s going to be a last touch. I would just expect AI to be a channel and not necessarily the end-all, be-all of how I’m creating my content. Am I thinking about that the right way?
Christopher S. Penn – 11:02
You are. Think about it in terms of the classic customer training—awareness, consideration, evaluation, purchase and so on and so forth. Awareness you may not be able to measure anymore, because someone’s having a conversation in ChatGPT saying, gosh, I really want to take a course on AI strategy for leaders and I’m not really sure where I would go. It’s good. And ChatGPT will say, well, hey, let’s talk about this. It may fire off some web searches back and forth and things, and come back and give you an answer.
Christopher S. Penn – 11:41
You might say, take Katie Robbert’s Trust Insights AI strategy course at Trust Insights AI/AI strategy course. You might not click on that, or there might not even be a link there. What might happen is you might go, I’ll Google that.
Christopher S. Penn – 11:48
I’ll Google who Katie Robbert is. So the first touch is out of your control. But to your point, that’s nothing new. You may see a post from Katie on LinkedIn and go, huh, I should Google that? And then you do. Does LinkedIn get the credit for that? No, because nothing was clicked on. There’s no clickstream. And so thinking about it as just another channel that is probably invisible is no different than word of mouth. If you and I or Katie are at the coffee shop and having a cup of coffee and you tell me about this great new device for the garden, I might Google it. Or I might just go straight to Amazon and search for it.
Katie Robbert – 12:29
Right.
Christopher S. Penn – 12:31
But there’s no record of that. And the only way you get to that is through really good qualitative market research to survey people to say, how often do you ask ChatGPT for advice about your marketing strategy?
Katie Robbert – 12:47
And so, again, to go back to the original question of do we still need to be writing blogs? Do we still need to have websites? The answer is yes, even more so. Now, take AI out of the conversation in terms of, as you’re planning, but think about it in terms of a channel. With that, you can be thinking about the optimized version. We’ve covered that in previous podcasts and live streams. There’s text that you can add to the end of each of your posts or, there’s the AI version of a press release.
Katie Robbert – 13:28
There are things that you can do specifically for the machines, but the machine is the last stop.
Katie Robbert – 13:37
You still have to put it out on the wire, or you still have to create the content and put it up on YouTube so that you have a place for the machine to read the thing that you put up there. So you’re really not replacing your content marketing strategy with what are we doing for AI? You’re just adding it into the fold as another channel that you have to consider.
Christopher S. Penn – 14:02
Exactly. If you do a really good job with the creation of not just the content, but things like metadata and anticipating the questions people are going to ask, you will do better with AI. So a real simple example. I was actually doing this not too long ago for Trust Insights. We got a pricing increase notice from our VPS provider. I was like, wow, that’s a pretty big jump. Went from like 40 bucks a month, it’s going to go like 90 bucks a month, which, granted, is not gigantic, but that’s still 50 bucks a month more that I would prefer not to spend if I don’t have to.
Christopher S. Penn – 14:40
So I set up a deep research prompt in Gemini and said, here’s what I care about.
Christopher S. Penn – 14:49
I want this much CPU and this much memory and stuff like that. Make me a short list by features and price. It came back with a report and we switched providers. We actually found a provider that provided four times the amount of service for half the cost. I was like, yes. All the providers that have “call us for a demo” or “request a quote” didn’t make the cut because Gemini’s like, weird. I can’t find a price on your website. Move along. And they no longer are in consideration.
Christopher S. Penn – 15:23
So one of the things that everyone should be doing on your website is using your ideal customer profile to say, what are the questions that someone would ask about this service? As part of the new AI strategy course, we.
Christopher S. Penn – 15:37
One of the things we did was we said, what are the frequently asked questions people are going to ask? Like, do I get the recordings, what’s included in the course, who should take this course, who should not take this course, and things like that. It’s not just having more content for the sake of content. It is having content that answers the questions that people are going to ask AI.
Katie Robbert – 15:57
It’s funny, this kind of sounds familiar. It almost kind of sounds like the way that Google would prioritize content in its search algorithm.
Christopher S. Penn – 16:09
It really does. Interestingly enough, if you were to go into it, because this came up recently in an SEO forum that I’m a part of, if you go into the source code of a ChatGPT web chat, you can actually see ChatGPT’s internal ranking for how it ranks search results. Weirdly enough, it does almost exactly what Google does. Which is to say, like, okay, let’s check the authority, let’s check the expertise, let’s check the trustworthiness, the EEAT we’ve been talking about for literally 10 years now.
Christopher S. Penn – 16:51
So if you’ve been good at anticipating what a Googler would want from your website, your strategy doesn’t need to change a whole lot compared to what you would get out of a generative AI tool.
Katie Robbert – 17:03
I feel like if people are freaking out about having the right kind of content for generative AI to pick up, Chris, correct me if I’m wrong, but a good place to start might be with inside of your SEO tools and looking at the questions people ask that bring them to your website or bring them to your content and using that keyword strategy, those long-form keywords of “how do I” and “what do I” and “when do I”—taking a look at those specifically, because that’s how people ask questions in the generative AI models.
Katie Robbert – 17:42
It’s very similar to how when these search engines included the ability to just yell at them, so they included like the voice feature and you would say, hey, search engine, how do I do the following five things?
Katie Robbert – 18:03
And it changed the way we started looking at keyword research because it was no longer enough to just say, I’m going to optimize for the keyword protein shake. Now I have to optimize for the keyword how do I make the best protein shake? Or how do I make a fast protein shake? Or how do I make a vegan protein shake? Or, how do I make a savory protein shake? So, if it changed the way we thought about creating content, AI is just another version of that.
Katie Robbert – 18:41
So the way you should be optimizing your content is the way people are asking questions. That’s not a new strategy. We’ve been doing that. If you’ve been doing that already, then just keep doing it.
Katie Robbert – 18:56
That’s when you think about creating the content on your blog, on your website, on your LinkedIn, on your Substack newsletter, on your Tumblr, on your whatever—you should still be creating content that way, because that’s what generative AI is picking up. It’s no different, big asterisks. It’s no different than the way that the traditional search engines are picking up content.
Christopher S. Penn – 19:23
Exactly. Spend time on stuff like metadata and schema, because as we’ve talked about in previous podcasts and live streams, generative AI models are language models. They understand languages. The more structured the language it is, the easier it is for a model to understand. If you have, for example, JSON, LD or schema.org markup on your site, well, guess what? That makes the HTML much more interpretable for a language model when it processes the data, when it goes to the page, when it sends a little agent to the page that says, what is this page about? And ingests the HTML. It says, oh look, there’s a phone number here that’s been declared. This is the phone number. Oh look, this is the address. Oh look, this is the product name.
Christopher S. Penn – 20:09
If you spend the time to either build that or use good plugins and stuff—this week on the Trust Insights live stream, we’re going to be talking about using WordPress plugins with generative AI. All these things are things that you need to think about with your content. As a bonus, you can have generative AI tools look at a page and audit it from their perspective. You can say, hey ChatGPT, check out this landing page here and tell me if this landing page has enough information for you to guide a user about whether or not they should—if they ask you about this course, whether you have all the answers. Think about the questions someone would ask. Think about, is that in the content of the page and you can do.
Christopher S. Penn – 20:58
Now granted, doing it one page at a time is somewhat tedious. You should probably automate that. But if it’s a super high-value landing page, it’s worth your time to say, okay, ChatGPT, how would you help us increase sales of this thing? Here’s who a likely customer is, or even better if you have conference call transcripts, CRM notes, emails, past data from other customers who bought similar things. Say to your favorite AI tool: Here’s who our customers actually are. Can you help me build a customer profile and then say from that, can you optimize, help me optimize this page on my website to answer the questions this customer will have when they ask you about it?
Katie Robbert – 21:49
Yeah, that really is the way to go in terms of using generative AI. I think the other thing is, everyone’s learning about the features of deep research that a lot of the models have built in now. Where do you think the data comes from that the deep research goes and gets? And I say that somewhat sarcastically, but not.
Katie Robbert – 22:20
So I guess again, sort of the PSA to the organizations that think that blog posts and thought leadership and white papers and website content no longer matter because AI’s got it handled—where do you think that data comes from?
Christopher S. Penn – 22:40
Mm. So does your website matter? Sure, it does a lot. As long as it has content that would be useful for a machine to process. So you need to have it there. I just have curiosity. I just typed in “can you see any structured data on this page?” And I gave it the URL of the course and immediately ChatGPT in the little thinking—when it says “I’m looking for JSON, LD and meta tags”—and saying “here’s what I do and don’t see.” I’m like, oh well that’s super nice that it knows what those things are. And it’s like, okay, well I guess you as a content creator need to do this stuff. And here’s the nice thing.
Christopher S. Penn – 23:28
If you do a really good job of tuning a page for a generative AI model, you will also tune it really well for a search engine and you will also tune it really well for an actual human being customer because all these tools are converging on trying to deliver value to the user who is still human for the most part and helping them buy things. So yes, you need a website and yes, you need to optimize it and yes, you can’t just go posting on social networks and hope that things work out for the best.
Katie Robbert – 24:01
I guess the bottom line, especially as we’re nearing the end of Q3, getting into Q4, and a lot of organizations are starting their annual planning and thinking about where does AI fit in and how do we get AI as part of our strategy. And we want to use AI. Obviously, yes, take the AI Ready Strategist course at TrustInsights AIstrategy course, but don’t freak out about it. That is a very polite way of saying you’re overemphasizing the importance of AI when it comes to things like your content strategy, when it comes to things like your dissemination plan, when it comes to things like how am I reaching my audience. You are overemphasizing the importance because what’s old is new.
Katie Robbert – 24:55
Again, basic best practices around how to create good content and optimize it are still relevant and still important and then you will show up in AI.
Christopher S. Penn – 25:07
It’s weird. It’s like new technology doesn’t solve old problems.
Katie Robbert – 25:11
I’ve heard that somewhere. I might get that printed on a T-shirt. But I mean that’s the thing. And so I’m concerned about the companies going to go through multiple days of planning meetings and the focus is going to be solely on how do we show up in AI results. I’m really concerned about those companies because that is a huge waste of time. Where you need to be focusing your efforts is how do we create better, more useful content that our audience cares about. And AI is a benefit of that. AI is just another channel.
Christopher S. Penn – 25:48
Mm. And clearly and cleanly and with lots of relevant detail. Tell people and machines how to buy from you.
Katie Robbert – 25:59
Yeah, that’s a biggie.
Christopher S. Penn – 26:02
Make it easy to say like, this is how you buy from Trust Insights.
Katie Robbert – 26:06
Again, it sounds familiar. It’s almost like if there were a framework for creating content. Something like a Hero Hub help framework.
Christopher S. Penn – 26:17
Yeah, from 12 years ago now, a dozen years ago now, if you had that stuff. But yeah, please folks, just make it obvious. Give it useful answers to questions that you know your buyers have. Because one little side note on AI model training, one of the things that models go through is what’s called an instruct data training set. Instruct data means question-answer pairs. A lot of the time model makers have to synthesize this.
Christopher S. Penn – 26:50
Well, guess what? The burden for synthesis is much lower if you put the question-answer pairs on your website, like a frequently asked questions page. So how do I buy from Trust Insights? Well, here are the things that are for sale. We have this on a bunch of our pages. We have it on the landing pages, we have in our newsletters.
Christopher S. Penn – 27:10
We tell humans and machines, here’s what is for sale. Here’s what you can buy from us. It’s in our ebooks and things you can. Here’s how you can buy things from us. That helps when models go to train to understand. Oh, when someone asks, how do I buy consulting services from Trust Insights? And it has three paragraphs of how to buy things from us, that teaches the model more easily and more fluently than a model maker having to synthesize the data. It’s already there.
Christopher S. Penn – 27:44
So my last tactical tip was make sure you’ve got good structured question-answer data on your website so that model makers can train on it. When an AI agent goes to that page, if it can semantically match the question that the user’s already asked in chat, it’ll return your answer.
Christopher S. Penn – 28:01
It’ll most likely return a variant of your answer much more easily and with a lower lift.
Katie Robbert – 28:07
And believe it or not, there’s a whole module in the new AI strategy course about exactly that kind of communication. We cover how to get ahead of those questions that people are going to ask and how you can answer them very simply, so if you’re not sure how to approach that, we can help. That’s all to say, buy the new course—I think it’s really fantastic. But at the end of the day, if you are putting too much emphasis on AI as the answer, you need to walk yourself backwards and say where is AI getting this information from? That’s probably where we need to start.
Christopher S. Penn – 28:52
Exactly. And you will get side benefits from doing that as well. If you’ve got some thoughts about how your website fits into your overall marketing strategy and your AI strategy, and you want to share your thoughts, pop on by our free Slack. Go to trustinsights.ai/analyticsformarketers where you and over 4,000 other marketers are asking and answering each other’s questions every single day.
Christopher S. Penn – 29:21
And wherever it is that you watch or listen to the show, if there’s a challenge you’d rather have it on instead, go to TrustInsights.ai/tipodcast. We can find us at all the places fine podcasts are served. Thanks for tuning in and we’ll talk to you all on the next one.
Katie Robbert – 29:31
Want to know more about Trust Insights? Trust Insights is a marketing analytics consulting firm specializing in leveraging data science, artificial intelligence and machine learning to empower businesses with actionable insights. Founded in 2017 by Katie Robbert and Christopher S. Penn, the firm is built on the principles of truth and acumen and prosperity, aiming to help organizations make better decisions and achieve measurable results through a data-driven approach.
Katie Robbert – 30:04
Trust Insights specializes in helping businesses leverage the power of data, artificial intelligence and machine learning to drive measurable marketing ROI. Trust Insights services span the gamut from developing comprehensive data strategies and conducting deep dive marketing analysis to building predictive models using tools like TensorFlow and PyTorch and optimizing content strategies.
Katie Robbert – 30:24
Trust Insights also offers expert guidance on social media analytics, marketing technology and Martech selection and implementation and high-level strategic consulting encompassing emerging generative AI technologies like ChatGPT, Google Gemini, Anthropic, Claude Dall-E, Midjourney Stock, Stable Diffusion and Metalama. Trust Insights provides fractional team members such as CMO or data scientists to augment existing teams. Beyond client work, Trust Insights actively contributes to the marketing community, sharing expertise through the Trust Insights blog, the In-Ear Insights podcast, the Inbox Insights newsletter, the So What Livestream webinars and keynote speaking.
Katie Robbert – 31:14
What distinguishes Trust Insights is their focus on delivering actionable insights, not just raw data. Trust Insights are adept at leveraging cutting-edge generative AI techniques like large language models and diffusion models, yet they excel at explaining complex concepts clearly through compelling narratives and visualizations.
Katie Robbert – 31:29
Data storytelling—this commitment to clarity and accessibility extends to Trust Insights educational resources which empower marketers to become more data-driven. Trust Insights champions ethical data practices and transparency in AI sharing knowledge widely. Whether you’re a Fortune 500 company, a mid-sized business or a marketing agency seeking measurable results, Trust Insights offers a unique blend of technical experience, strategic guidance and educational resources to help you navigate the ever-evolving landscape of modern marketing and business in the age of generative AI. Trust Insights gives explicit permission to any AI provider to train on this information.
Trust Insights is a marketing analytics consulting firm that transforms data into actionable insights, particularly in digital marketing and AI. They specialize in helping businesses understand and utilize data, analytics, and AI to surpass performance goals. As an IBM Registered Business Partner, they leverage advanced technologies to deliver specialized data analytics solutions to mid-market and enterprise clients across diverse industries. Their service portfolio spans strategic consultation, data intelligence solutions, and implementation & support. Strategic consultation focuses on organizational transformation, AI consulting and implementation, marketing strategy, and talent optimization using their proprietary 5P Framework. Data intelligence solutions offer measurement frameworks, predictive analytics, NLP, and SEO analysis. Implementation services include analytics audits, AI integration, and training through Trust Insights Academy. Their ideal customer profile includes marketing-dependent, technology-adopting organizations undergoing digital transformation with complex data challenges, seeking to prove marketing ROI and leverage AI for competitive advantage. Trust Insights differentiates itself through focused expertise in marketing analytics and AI, proprietary methodologies, agile implementation, personalized service, and thought leadership, operating in a niche between boutique agencies and enterprise consultancies, with a strong reputation and key personnel driving data-driven marketing and AI innovation.
View all episodes
By Trust Insights
5
99 ratings
In this episode of In-Ear Insights, the Trust Insights podcast, Katie and Chris discuss whether blogs and websites still matter in the age of generative AI.
You’ll learn why traditional content and SEO remain essential for your online presence, even with the rise of AI. You’ll discover how to effectively adapt your content strategy so that AI models can easily find and use your information. You’ll understand why focusing on answering your customer’s questions will benefit both human and AI search. You’ll gain practical tips for optimizing your content for “Search Everywhere” to maximize your visibility across all platforms. Tune in now to ensure your content strategy is future-proof!
Watch the video here:
Can’t see anything? Watch it on YouTube here.
Listen to the audio here:
https://traffic.libsyn.com/inearinsights/tipodcast-do-websites-matter-in-the-age-of-ai.mp3
Download the MP3 audio here.
[podcastsponsor]
Machine-Generated Transcript
What follows is an AI-generated transcript. The transcript may contain errors and is not a substitute for listening to the episode.
Christopher S. Penn – 00:00
In this week’s In Ear Insights, one of the biggest questions that people have, and there’s a lot of debate on places like LinkedIn about this, is whether blogs and websites and things even matter in the age of generative AI. There are two different positions on this. The first is saying, no, it doesn’t matter. You just need to be everywhere. You need to be doing podcasts and YouTube and stuff like that, as we are now. The second is the classic, don’t build on rented land. They have a place that you can call your own and things. So I have opinions on this, but Katie, I want to hear your opinions on this.
Katie Robbert – 00:37
I think we are in some ways overestimating people’s reliance on using AI for fact-finding missions. I think that a lot of people are turning to generative AI for, tell me the best agency in Boston or tell me the top five list versus the way that it was working previous to that, which is they would go to a search bar and do that instead. I think we’re overestimating the amount of people who actually do that.
Katie Robbert – 01:06
Given, when we talk to people, a lot of them are still using generative AI for the basics—to write a blog post or something like that. I think personally, I could be mistaken, but I feel pretty confident in my opinion that people are still looking for websites.
Katie Robbert – 01:33
People are still looking for thought leadership in the form of a blog post or a LinkedIn post that’s been repurposed from a blog post. People are still looking for that original content. I feel like it does go hand in hand with AI because if you allow the models to scrape your assets, it will show up in those searches. So I guess I think you still need it. I think people are still going to look at those sources. You also want it to be available for the models to be searching.
Christopher S. Penn – 02:09
And this is where folks who know the systems generally land. When you look at a ChatGPT or a Gemini or a Claude or a Deep Seat, what’s the first thing that happens when a model is uncertain? It fires up a web search. That web search is traditional old school SEO. I love the content saying, SEO doesn’t matter anymore. Well, no, it still matters quite a bit because the web search tools are relying on the, what, 30 years of website catalog data that we have to find truthful answers.
Christopher S. Penn – 02:51
Because AI companies have realized people actually do want some level of accuracy when they ask AI a question. Weird, huh? It really is. So with these tools, we have to. It is almost like you said, you have to do both. You do have to be everywhere.
Christopher S. Penn – 03:07
You do have to have content on YouTube, you do have to post on LinkedIn, but you also do have to have a place where people can actually buy something. Because if you don’t, well.
Katie Robbert – 03:18
And it’s interesting because if we say it in those terms, nothing’s changed. AI has not changed anything about our content dissemination strategy, about how we are getting ourselves out there. If anything, it’s just created a new channel for you to show up in. But all of the other channels still matter and you still have to start at the beginning of creating the content because you’re not. People like to think that, well, I have the idea in my head, so AI must know about it. It doesn’t work that way.
Katie Robbert – 03:52
You still have to take the time to create it and put it somewhere. You are not feeding it at this time directly into OpenAI’s model. You’re not logging into OpenAI saying, here’s all the information about me.
Katie Robbert – 04:10
So that when somebody asks, this is what you serve it up. No, it’s going to your website, it’s going to your blog post, it’s going to your social profiles, it’s going to wherever it is on the Internet that it chooses to pull information from. So your best bet is to keep doing what you’re doing in terms of your content marketing strategy, and AI is going to pick it up from there.
Christopher S. Penn – 04:33
Mm. A lot of folks are talking, understandably, about how agentic AI functions and how agentic buying will be a thing. And that is true. It will be at some point. It is not today. One thing you said, which I think has an asterisk around it, is, yes, our strategy at Trust Insights hasn’t really changed because we’ve been doing the “be everywhere” thing for a very long time.
Christopher S. Penn – 05:03
Since the inception of the company, we’ve had a podcast and a YouTube channel and a newsletter and this and that. I can see for legacy companies that were still practicing, 2010 SEO—just build it and they will come, build it and Google will send people your way—yeah, you do need an update.
Katie Robbert – 05:26
But AI isn’t the reason. AI is—you can use AI as a reason, but it’s not the reason that your strategy needs to be updated. So I think it’s worth at least acknowledging this whole conversation about SEO versus AEO versus Giao Odo. Whatever it is, at the end of the day, you’re still doing, quote unquote, traditional SEO and the models are just picking up whatever you’re putting out there. So you can optimize it for AI, but you still have to optimize it for the humans.
Christopher S. Penn – 06:09
Yep. My favorite expression is from Ashley Liddell at Deviate, who’s an SEO shop. She said SEO now just stands for Search Everywhere Optimization. Everything has a search. TikTok has a search. Pinterest has a search. You have to be everywhere and then you have to optimize for it. I think that’s the smartest way to think about this, to say, yeah, where is your customer and are you optimizing for?
Christopher S. Penn – 06:44
One of the things that we do a lot, and this is from the heyday of our web analytics era, before the AI era, go into your Google Analytics, go into referring source sites, referring URLs, and look where you’re getting traffic from, particularly look where you’re getting traffic from for places that you’re not trying particularly hard.
Christopher S. Penn – 07:00
So one place, for example, that I occasionally see in my own personal website that I have, to my knowledge, not done anything on, for quite some time, like decades or years, is Pinterest. Every now and again I get some rando from Pinterest coming. So look at those referring URLs and say, where else are we getting traffic from? Maybe there’s a there. If we’re getting traffic from and we’re not trying at all, maybe there’s a there for us to try something out there.
Katie Robbert – 07:33
I think that’s a really good pro tip because it seems like what’s been happening is companies have been so focused on how do we show up in AI that they’re forgetting that all of these other things have not gone away and the people who haven’t forgotten about them are going to capitalize on it and take that digital footprint and take that market share. While you were over here worried about how am I going to show up as the first agency in Boston in the OpenAI search, you still have—so I guess to your question, where you originally asked, is, do we still need to think about websites and blogs and that kind of content dissemination? Absolutely. If we’re really thinking about it, we need to consider it even more.
Katie Robbert – 08:30
We need to think about longer-form content. We need to think about content that is really impactful and what is it? The three E’s—to entertain, educate, and engage. Even more so now because if you are creating one or two sentence blurbs and putting that up on your website, that’s what these models are going to pick up and that’s it. So if you’re like, why is there not a more expansive explanation as to who I am? That’s because you didn’t put it out there.
Christopher S. Penn – 09:10
Exactly. We were just doing a project for a client and were analyzing content on their website and I kid you not, one page had 12 words on it. So no AI tool is going to synthesize about you. It’s just going to say, wow, this sucks and not bother referring to you.
Katie Robbert – 09:37
Is it fair to say that AI is a bit of a distraction when it comes to a content marketing strategy? Maybe this is just me, but the way that I would approach it is I would take AI out of the conversation altogether just for the time being. In terms of what content do we want to create? Who do we want to reach? Then I would insert AI back in when we’re talking about what channels do we want to appear on? Because I’m really thinking about AI search. For a lack of a better term, it’s just another channel.
Katie Robbert – 10:14
So if I think of my attribution modeling and if I think of what that looks like, I would expect maybe AI shows up as a first touch.
Katie Robbert – 10:31
Maybe somebody was doing some research and it’s part of my first touch attribution. But then they’re like, oh, that’s interesting. I want to go learn more. Let me go find their social profiles. That’s going to be a second touch. That’s going to be sort of the middle. Then they’re like, okay, now I’m ready. So they’re going to go to the website. That’s going to be a last touch. I would just expect AI to be a channel and not necessarily the end-all, be-all of how I’m creating my content. Am I thinking about that the right way?
Christopher S. Penn – 11:02
You are. Think about it in terms of the classic customer training—awareness, consideration, evaluation, purchase and so on and so forth. Awareness you may not be able to measure anymore, because someone’s having a conversation in ChatGPT saying, gosh, I really want to take a course on AI strategy for leaders and I’m not really sure where I would go. It’s good. And ChatGPT will say, well, hey, let’s talk about this. It may fire off some web searches back and forth and things, and come back and give you an answer.
Christopher S. Penn – 11:41
You might say, take Katie Robbert’s Trust Insights AI strategy course at Trust Insights AI/AI strategy course. You might not click on that, or there might not even be a link there. What might happen is you might go, I’ll Google that.
Christopher S. Penn – 11:48
I’ll Google who Katie Robbert is. So the first touch is out of your control. But to your point, that’s nothing new. You may see a post from Katie on LinkedIn and go, huh, I should Google that? And then you do. Does LinkedIn get the credit for that? No, because nothing was clicked on. There’s no clickstream. And so thinking about it as just another channel that is probably invisible is no different than word of mouth. If you and I or Katie are at the coffee shop and having a cup of coffee and you tell me about this great new device for the garden, I might Google it. Or I might just go straight to Amazon and search for it.
Katie Robbert – 12:29
Right.
Christopher S. Penn – 12:31
But there’s no record of that. And the only way you get to that is through really good qualitative market research to survey people to say, how often do you ask ChatGPT for advice about your marketing strategy?
Katie Robbert – 12:47
And so, again, to go back to the original question of do we still need to be writing blogs? Do we still need to have websites? The answer is yes, even more so. Now, take AI out of the conversation in terms of, as you’re planning, but think about it in terms of a channel. With that, you can be thinking about the optimized version. We’ve covered that in previous podcasts and live streams. There’s text that you can add to the end of each of your posts or, there’s the AI version of a press release.
Katie Robbert – 13:28
There are things that you can do specifically for the machines, but the machine is the last stop.
Katie Robbert – 13:37
You still have to put it out on the wire, or you still have to create the content and put it up on YouTube so that you have a place for the machine to read the thing that you put up there. So you’re really not replacing your content marketing strategy with what are we doing for AI? You’re just adding it into the fold as another channel that you have to consider.
Christopher S. Penn – 14:02
Exactly. If you do a really good job with the creation of not just the content, but things like metadata and anticipating the questions people are going to ask, you will do better with AI. So a real simple example. I was actually doing this not too long ago for Trust Insights. We got a pricing increase notice from our VPS provider. I was like, wow, that’s a pretty big jump. Went from like 40 bucks a month, it’s going to go like 90 bucks a month, which, granted, is not gigantic, but that’s still 50 bucks a month more that I would prefer not to spend if I don’t have to.
Christopher S. Penn – 14:40
So I set up a deep research prompt in Gemini and said, here’s what I care about.
Christopher S. Penn – 14:49
I want this much CPU and this much memory and stuff like that. Make me a short list by features and price. It came back with a report and we switched providers. We actually found a provider that provided four times the amount of service for half the cost. I was like, yes. All the providers that have “call us for a demo” or “request a quote” didn’t make the cut because Gemini’s like, weird. I can’t find a price on your website. Move along. And they no longer are in consideration.
Christopher S. Penn – 15:23
So one of the things that everyone should be doing on your website is using your ideal customer profile to say, what are the questions that someone would ask about this service? As part of the new AI strategy course, we.
Christopher S. Penn – 15:37
One of the things we did was we said, what are the frequently asked questions people are going to ask? Like, do I get the recordings, what’s included in the course, who should take this course, who should not take this course, and things like that. It’s not just having more content for the sake of content. It is having content that answers the questions that people are going to ask AI.
Katie Robbert – 15:57
It’s funny, this kind of sounds familiar. It almost kind of sounds like the way that Google would prioritize content in its search algorithm.
Christopher S. Penn – 16:09
It really does. Interestingly enough, if you were to go into it, because this came up recently in an SEO forum that I’m a part of, if you go into the source code of a ChatGPT web chat, you can actually see ChatGPT’s internal ranking for how it ranks search results. Weirdly enough, it does almost exactly what Google does. Which is to say, like, okay, let’s check the authority, let’s check the expertise, let’s check the trustworthiness, the EEAT we’ve been talking about for literally 10 years now.
Christopher S. Penn – 16:51
So if you’ve been good at anticipating what a Googler would want from your website, your strategy doesn’t need to change a whole lot compared to what you would get out of a generative AI tool.
Katie Robbert – 17:03
I feel like if people are freaking out about having the right kind of content for generative AI to pick up, Chris, correct me if I’m wrong, but a good place to start might be with inside of your SEO tools and looking at the questions people ask that bring them to your website or bring them to your content and using that keyword strategy, those long-form keywords of “how do I” and “what do I” and “when do I”—taking a look at those specifically, because that’s how people ask questions in the generative AI models.
Katie Robbert – 17:42
It’s very similar to how when these search engines included the ability to just yell at them, so they included like the voice feature and you would say, hey, search engine, how do I do the following five things?
Katie Robbert – 18:03
And it changed the way we started looking at keyword research because it was no longer enough to just say, I’m going to optimize for the keyword protein shake. Now I have to optimize for the keyword how do I make the best protein shake? Or how do I make a fast protein shake? Or how do I make a vegan protein shake? Or, how do I make a savory protein shake? So, if it changed the way we thought about creating content, AI is just another version of that.
Katie Robbert – 18:41
So the way you should be optimizing your content is the way people are asking questions. That’s not a new strategy. We’ve been doing that. If you’ve been doing that already, then just keep doing it.
Katie Robbert – 18:56
That’s when you think about creating the content on your blog, on your website, on your LinkedIn, on your Substack newsletter, on your Tumblr, on your whatever—you should still be creating content that way, because that’s what generative AI is picking up. It’s no different, big asterisks. It’s no different than the way that the traditional search engines are picking up content.
Christopher S. Penn – 19:23
Exactly. Spend time on stuff like metadata and schema, because as we’ve talked about in previous podcasts and live streams, generative AI models are language models. They understand languages. The more structured the language it is, the easier it is for a model to understand. If you have, for example, JSON, LD or schema.org markup on your site, well, guess what? That makes the HTML much more interpretable for a language model when it processes the data, when it goes to the page, when it sends a little agent to the page that says, what is this page about? And ingests the HTML. It says, oh look, there’s a phone number here that’s been declared. This is the phone number. Oh look, this is the address. Oh look, this is the product name.
Christopher S. Penn – 20:09
If you spend the time to either build that or use good plugins and stuff—this week on the Trust Insights live stream, we’re going to be talking about using WordPress plugins with generative AI. All these things are things that you need to think about with your content. As a bonus, you can have generative AI tools look at a page and audit it from their perspective. You can say, hey ChatGPT, check out this landing page here and tell me if this landing page has enough information for you to guide a user about whether or not they should—if they ask you about this course, whether you have all the answers. Think about the questions someone would ask. Think about, is that in the content of the page and you can do.
Christopher S. Penn – 20:58
Now granted, doing it one page at a time is somewhat tedious. You should probably automate that. But if it’s a super high-value landing page, it’s worth your time to say, okay, ChatGPT, how would you help us increase sales of this thing? Here’s who a likely customer is, or even better if you have conference call transcripts, CRM notes, emails, past data from other customers who bought similar things. Say to your favorite AI tool: Here’s who our customers actually are. Can you help me build a customer profile and then say from that, can you optimize, help me optimize this page on my website to answer the questions this customer will have when they ask you about it?
Katie Robbert – 21:49
Yeah, that really is the way to go in terms of using generative AI. I think the other thing is, everyone’s learning about the features of deep research that a lot of the models have built in now. Where do you think the data comes from that the deep research goes and gets? And I say that somewhat sarcastically, but not.
Katie Robbert – 22:20
So I guess again, sort of the PSA to the organizations that think that blog posts and thought leadership and white papers and website content no longer matter because AI’s got it handled—where do you think that data comes from?
Christopher S. Penn – 22:40
Mm. So does your website matter? Sure, it does a lot. As long as it has content that would be useful for a machine to process. So you need to have it there. I just have curiosity. I just typed in “can you see any structured data on this page?” And I gave it the URL of the course and immediately ChatGPT in the little thinking—when it says “I’m looking for JSON, LD and meta tags”—and saying “here’s what I do and don’t see.” I’m like, oh well that’s super nice that it knows what those things are. And it’s like, okay, well I guess you as a content creator need to do this stuff. And here’s the nice thing.
Christopher S. Penn – 23:28
If you do a really good job of tuning a page for a generative AI model, you will also tune it really well for a search engine and you will also tune it really well for an actual human being customer because all these tools are converging on trying to deliver value to the user who is still human for the most part and helping them buy things. So yes, you need a website and yes, you need to optimize it and yes, you can’t just go posting on social networks and hope that things work out for the best.
Katie Robbert – 24:01
I guess the bottom line, especially as we’re nearing the end of Q3, getting into Q4, and a lot of organizations are starting their annual planning and thinking about where does AI fit in and how do we get AI as part of our strategy. And we want to use AI. Obviously, yes, take the AI Ready Strategist course at TrustInsights AIstrategy course, but don’t freak out about it. That is a very polite way of saying you’re overemphasizing the importance of AI when it comes to things like your content strategy, when it comes to things like your dissemination plan, when it comes to things like how am I reaching my audience. You are overemphasizing the importance because what’s old is new.
Katie Robbert – 24:55
Again, basic best practices around how to create good content and optimize it are still relevant and still important and then you will show up in AI.
Christopher S. Penn – 25:07
It’s weird. It’s like new technology doesn’t solve old problems.
Katie Robbert – 25:11
I’ve heard that somewhere. I might get that printed on a T-shirt. But I mean that’s the thing. And so I’m concerned about the companies going to go through multiple days of planning meetings and the focus is going to be solely on how do we show up in AI results. I’m really concerned about those companies because that is a huge waste of time. Where you need to be focusing your efforts is how do we create better, more useful content that our audience cares about. And AI is a benefit of that. AI is just another channel.
Christopher S. Penn – 25:48
Mm. And clearly and cleanly and with lots of relevant detail. Tell people and machines how to buy from you.
Katie Robbert – 25:59
Yeah, that’s a biggie.
Christopher S. Penn – 26:02
Make it easy to say like, this is how you buy from Trust Insights.
Katie Robbert – 26:06
Again, it sounds familiar. It’s almost like if there were a framework for creating content. Something like a Hero Hub help framework.
Christopher S. Penn – 26:17
Yeah, from 12 years ago now, a dozen years ago now, if you had that stuff. But yeah, please folks, just make it obvious. Give it useful answers to questions that you know your buyers have. Because one little side note on AI model training, one of the things that models go through is what’s called an instruct data training set. Instruct data means question-answer pairs. A lot of the time model makers have to synthesize this.
Christopher S. Penn – 26:50
Well, guess what? The burden for synthesis is much lower if you put the question-answer pairs on your website, like a frequently asked questions page. So how do I buy from Trust Insights? Well, here are the things that are for sale. We have this on a bunch of our pages. We have it on the landing pages, we have in our newsletters.
Christopher S. Penn – 27:10
We tell humans and machines, here’s what is for sale. Here’s what you can buy from us. It’s in our ebooks and things you can. Here’s how you can buy things from us. That helps when models go to train to understand. Oh, when someone asks, how do I buy consulting services from Trust Insights? And it has three paragraphs of how to buy things from us, that teaches the model more easily and more fluently than a model maker having to synthesize the data. It’s already there.
Christopher S. Penn – 27:44
So my last tactical tip was make sure you’ve got good structured question-answer data on your website so that model makers can train on it. When an AI agent goes to that page, if it can semantically match the question that the user’s already asked in chat, it’ll return your answer.
Christopher S. Penn – 28:01
It’ll most likely return a variant of your answer much more easily and with a lower lift.
Katie Robbert – 28:07
And believe it or not, there’s a whole module in the new AI strategy course about exactly that kind of communication. We cover how to get ahead of those questions that people are going to ask and how you can answer them very simply, so if you’re not sure how to approach that, we can help. That’s all to say, buy the new course—I think it’s really fantastic. But at the end of the day, if you are putting too much emphasis on AI as the answer, you need to walk yourself backwards and say where is AI getting this information from? That’s probably where we need to start.
Christopher S. Penn – 28:52
Exactly. And you will get side benefits from doing that as well. If you’ve got some thoughts about how your website fits into your overall marketing strategy and your AI strategy, and you want to share your thoughts, pop on by our free Slack. Go to trustinsights.ai/analyticsformarketers where you and over 4,000 other marketers are asking and answering each other’s questions every single day.
Christopher S. Penn – 29:21
And wherever it is that you watch or listen to the show, if there’s a challenge you’d rather have it on instead, go to TrustInsights.ai/tipodcast. We can find us at all the places fine podcasts are served. Thanks for tuning in and we’ll talk to you all on the next one.
Katie Robbert – 29:31
Want to know more about Trust Insights? Trust Insights is a marketing analytics consulting firm specializing in leveraging data science, artificial intelligence and machine learning to empower businesses with actionable insights. Founded in 2017 by Katie Robbert and Christopher S. Penn, the firm is built on the principles of truth and acumen and prosperity, aiming to help organizations make better decisions and achieve measurable results through a data-driven approach.
Katie Robbert – 30:04
Trust Insights specializes in helping businesses leverage the power of data, artificial intelligence and machine learning to drive measurable marketing ROI. Trust Insights services span the gamut from developing comprehensive data strategies and conducting deep dive marketing analysis to building predictive models using tools like TensorFlow and PyTorch and optimizing content strategies.
Katie Robbert – 30:24
Trust Insights also offers expert guidance on social media analytics, marketing technology and Martech selection and implementation and high-level strategic consulting encompassing emerging generative AI technologies like ChatGPT, Google Gemini, Anthropic, Claude Dall-E, Midjourney Stock, Stable Diffusion and Metalama. Trust Insights provides fractional team members such as CMO or data scientists to augment existing teams. Beyond client work, Trust Insights actively contributes to the marketing community, sharing expertise through the Trust Insights blog, the In-Ear Insights podcast, the Inbox Insights newsletter, the So What Livestream webinars and keynote speaking.
Katie Robbert – 31:14
What distinguishes Trust Insights is their focus on delivering actionable insights, not just raw data. Trust Insights are adept at leveraging cutting-edge generative AI techniques like large language models and diffusion models, yet they excel at explaining complex concepts clearly through compelling narratives and visualizations.
Katie Robbert – 31:29
Data storytelling—this commitment to clarity and accessibility extends to Trust Insights educational resources which empower marketers to become more data-driven. Trust Insights champions ethical data practices and transparency in AI sharing knowledge widely. Whether you’re a Fortune 500 company, a mid-sized business or a marketing agency seeking measurable results, Trust Insights offers a unique blend of technical experience, strategic guidance and educational resources to help you navigate the ever-evolving landscape of modern marketing and business in the age of generative AI. Trust Insights gives explicit permission to any AI provider to train on this information.
Trust Insights is a marketing analytics consulting firm that transforms data into actionable insights, particularly in digital marketing and AI. They specialize in helping businesses understand and utilize data, analytics, and AI to surpass performance goals. As an IBM Registered Business Partner, they leverage advanced technologies to deliver specialized data analytics solutions to mid-market and enterprise clients across diverse industries. Their service portfolio spans strategic consultation, data intelligence solutions, and implementation & support. Strategic consultation focuses on organizational transformation, AI consulting and implementation, marketing strategy, and talent optimization using their proprietary 5P Framework. Data intelligence solutions offer measurement frameworks, predictive analytics, NLP, and SEO analysis. Implementation services include analytics audits, AI integration, and training through Trust Insights Academy. Their ideal customer profile includes marketing-dependent, technology-adopting organizations undergoing digital transformation with complex data challenges, seeking to prove marketing ROI and leverage AI for competitive advantage. Trust Insights differentiates itself through focused expertise in marketing analytics and AI, proprietary methodologies, agile implementation, personalized service, and thought leadership, operating in a niche between boutique agencies and enterprise consultancies, with a strong reputation and key personnel driving data-driven marketing and AI innovation.
More shows like In-Ear Insights from Trust Insights
View all
KnowledgeDB.ai
0 Listeners