Name: Building an AI-Ready Discoverability Strategy
Uploaded: 2026-09-22T18:51:07.123Z
Duration: 31 min 24 s
Description: Building an AI-Ready Discoverability Strategy

Transcript for "Building an AI-Ready Discoverability Strategy": Well, hello, everyone, and welcome to build an AI ready discoverability strategy. This is part two of a three part AI discoverability webinar series. And if you joined us last time, you might remember that we shared how to figure out how AI actually sees your brand today. And now we're shifting gears to the diagnosis, the treatment part. So once you have that baseline understanding, what do you do next? As a reminder, I'm your host. I'm Jody Saratani. I'm the CFO of WordPress VIP, and I'm joined once again today by my friend and my partner, Paxton Gray, who is the CEO of ninety seventh floor. So just a little bit of housekeeping. Quick notes. First, we're recording this session, so you will get a copy of the replay, in just a couple of days. Feel free to share that with a friend or colleague or rewatch it if there is anything that you missed. You can drop questions in the chat at any time. We will try to save some time at the end for live q and a. And there's also a live poll running. So please, drop your answer to, those questions at any point, and we'll discuss it a little later in the session. And then finally, as I mentioned, this is session two of a three part series. So if you missed session one for some reason, you can find the recording on wpvip.com. And we would love for you to join us in, in our next session, which is just in a few weeks, and it's really gonna be all about measurement. So just a quick agenda for today. We've got four parts to this webinar. We're gonna start with really talking about how you go from your diagnosis or your baseline to an actual plan. Then we're gonna talk through some of the technical foundation that needs to be in place and then shift gears all around content optimization and how you really influence AI to not only cite you but recommend you. And then finally, we'll talk about operationalizing your AI discoverability practice and what, we would recommend there. So with that, I will hand it over to Pax who's gonna get us started. Great. Thanks, Jody. Yeah. Looking to dive in. Last time, we talked about establishing a baseline so that you could see how you stacked up compared to competitors, see what AI is saying about you. So today, we're talking about how to actually influence and close that gap so that AI is making you show up in front of the right people saying the right things all in the name of growing your bottom line, which is the real trick these days. So, ninety seven floor and WordPress have been working to optimize, sites for AI search since before it became popular to search for things on AI search. And, we had determined that most of all the things that brands need to do can kinda fall into four categories, which become this AI visibility framework. Brand, obviously, is a big part of it. A lot of that is self explanatory, and we're not gonna get too much into that today. But then there's architecture, the technical aspect, the foundation, which is why WordPress is such a fantastic partner. They're literally what trained AI. And then content on your side is a massive factor and then third party validation. So, let's get into, the starting off with the architecture, the technical, and we'll pass it over to Jody to talk about, what WordPress is doing on that front. Cool. Right. So that was really, like, the big picture and the framework that we like to walk customers through. So we'll get first into the technical and operational foundations that really need to be true before anything else can really work. So there's three core building blocks in terms of a technical setup that we wanna talk through. If you're not a super technical person, I see that a lot of folks are on the marketing side, so some of this might be a little bit less familiar. You can kinda think of these three building blocks as a library, as an analogy. So the the first building block is really your information architecture. That's your floor floor plan. It tells, you know, a visitor, hey. There's fiction over here. There's nonfiction over there. This is the alpha you know, everything is alphabetized. So it's it's that it's that kind of structure that that helps a visitor understand where stuff is. Then you've got block number two. And the the block number two is really your structured content. So that's not so much where a book is on a shelf, so to speak, but more what does the individual label say? Who's the author? What's what's the title of the book, etcetera? And then the final block is your governance. So that is really, almost like a librarian. As as as things progress, you need to make sure that all your books are in the right place and all your labels are accurate. So these three things really work together as a technical foundation to make sure you're set up for success. So the truth is when it comes to your content system, in particular your information architecture and structured content, your current content system or your pre GPT content system, I could say it that way, can really fail you. So there are three kind of best practices that we recommend to our customers, and and I'll talk you through those. The first is you wanna name your your stuff the same way everywhere. So if you've got product names or your industry or people names, you wanna name it the same way every time on every page or every site that where it's relevant. So, for example, for WordPress VIP, if we say WordPress VIP on one page and enterprise WordPress, which arguably is talking about the same thing on another page, AI can see it as two different things. That's not always the case. There's workarounds there, and it's getting better, but you don't wanna risk it. So name your stuff the same, everywhere on your site. The second is that you wanna be building your content once and let your system reuse it. So, instead of using a, a free form text field, you wanna use fields that you can update a source once and it publishes live everywhere. So things like pricing tier names or the name of an executive or or product names, feature names, things like that. You wanna make sure that you do it that way, Model the content once, publish it everywhere, use fields, not free text blobs. And then third, you wanna put a a machine readable tag on every page that's actually baked into the template, not something that you have to rely on a person to remember to do. Believe me, we've seen this a lot on the customer side, where we're we're hoping that that someone on the web team remembers, to to put that in. And and sometimes they do, sometimes they don't. So you don't wanna risk it. And then finally, a lot of this is system level work that kind of begs the question, do you have a website software that can allow you to do this automatically, or are you dependent on your human resources to make sure to do all these things? If the answer is the latter and not the former, you might be in trouble. So really think about how your systems are set up and which systems you're using. Okay. And then governance, just a double tap there. Going back to the library analogy, if you're a librarian, how do you really ensure that your structure does not rot or or lose accuracy over time? So I've got a few recommendations, and they may seem super obvious, but even these basic governance principles are sometimes missing in some larger brands that we work with. So number one, you wanna have a government's document. So a a stamp you have to write your standards down. What are your heading and title rules? How are you treating metadata and schema? What are you calling things? Definitely make sure to write it down so that it's referenced by your current team and then anyone who's working on on the site in the future. The the second is that you wanna put some checks into your tooling. So similar to, like, a a spell check or something, you certainly can have folks rereading things and catching spelling errors, but it's so much easier and more reliable if you just have a a system level tooling that can point out when when things are are incorrectly set up. And then finally, you wanna have an owner, for every domain that you have. It's it's common that you have a lot of different folks that are contributing to AI discoverability. That's normal, and that's not a problem, but you wanna have a single accountable party who ultimately owns success and can coordinate with the rest of the teams. And we'll talk about that in a little bit more detail because it is really important, but that's the final recommendation. So the point of this, section is really around how your infrastructure makes you readable, and, that's really, really important. Right? But being readable isn't the same thing as being recommended, and that really is where content optimization comes in. So that's PAX's world. So I'll hand it back to you, Paxton, to to talk us through some of the best practices there. Yeah. Yeah. That's right. So for twenty years, it was all about ranking in those 10 blue links. And, the playbook for that was very straightforward. It was kind of known, and AI came in and disrupted that and changed a lot of things. But, so what we're gonna talk about today is, not just the mechanics of how AI interprets your brand, but why. And the fact is that there's often a big disconnect between what you want AI to see and say about your brand and what it actually does say about your brand. So, in the name of that, it's important to understand that, unlike SEO, AI is not determined by a single page. In SEO, you often had, like, your main pillar page, and it's about getting that thing to rank. AI is much more holistic, and it should be focused on topics and establishing authority around certain topics. So the way to think about this is, you know, the the big mistake a lot of people in the industry are making is they are taking prompts, and they are equating those with keywords, but they are not the same thing. As you can see here, this is a prompt. Prompts are often much longer. They're much and more in-depth, and, we no longer are training ourselves to condense our need into three keywords, but instead kind of laying it all out. And so, the way AI search works, is for these longer prompts, they will do multiple queries. And so we call this the the query fan out. So for this particular prompt, we see that these are actual, the actual query fan out and how the AI will then search. And so the key is answering these secondary questions through multiple pieces of content on our site, not simply on the single page. And that increases the likelihood for us to show up in AI search. And the big thing is we wanna be the recommended brand. So the more that we establish domain authority around the specific topic, the more likely we are to be recommended as a solution to their problem. So, the fact is that things have changed fundamentally. So, you need to focus not on keywords, but instead on topics your audience cares about and addressing their pain points. So gone are the days of the traditional keyword research as ways to establish and build your, strategy around AI. And now we need to go back to fun fundamentals, which is audience research. So scouring Reddit, looking for what words and phrases, what problems do they express. This is where we start to form some of the topics and prompts that we should be, optimizing for. And then all of this research should be, kinda coalesced into this overall strategy around our different personas, their pain points, what motivates them, the language that you use, and then the questions that they ask. And so you take all of that and that becomes the basis for our content strategy. So we what's kind of different here is, in compared to the traditional SEO worlds, we're talking about a lot more content, but it should be very, very specific to certain topics. So in this case, we have our main pillar page around high volume AI recruiting platform, and then we've got all these secondary and tertiary pages supporting and linking to that main key page. But they're a lot more specific in terms of answering, particular problems that they have. Now one big problem that a lot of, especially enterprise brands have, is they've got this this massive I think of it as like a wake, that they've been surfing for twenty, thirty years, a wake of content. And there's so much content sitting on their site that unknowingly, a lot of these brands have old content sitting there teaching AI something wrong about your brand, either it's talking about products that you don't have or messaging and positioning that you currently don't take, but no one went and cleaned up the old content. So we have this content analysis matrix that we use, when, we bring on a new client, and we go through all their old content, and we're looking at three things particularly. So one, is this piece of content pulling in existing traffic? If yes, then we're then look at, like, okay, what kind of metrics did the does this page have? Looking at engagement, time on page, click through rate, balance rate, and then we're also looking at, is this aligned with the business? So there are cases where you will have pages that have good traffic, they've got good engagement rate, but it is not aligned with your business. That page should not exist, and it is steering AI in the wrong direction. So it's, we've had tremendous success with some clients pairing down content, and then, that sends stronger signals to AI about who we are, who we're for, and when we should show up. So, a little case study about this. So, there's a a brand that we work for. They wanna be known as the Alaskan cruise line. They wanna completely dominate, all the ports in Alaska. Anyone searching, wanting to go on an Alaskan cruise, they wanna show up. So we pared down content, and then we started building content all related to Alaskan cruises, things around times of the year to go, different animals, different ports, different excursions. And then this wasn't just like we didn't throw a bunch of random content out there. It was all mapped to the user journey. So it took the customer in mind and, helped answer their queries when they needed it and and solve their problems. And so, these all had really great engagement rates, and they pushed the user down the funnel. The overall upshot of that is, you know, there were many, a long time when it was bouncing back and forth between who was the number one most recommended brand for AI or Alaska related cruises. And then this client, as soon as we start implementing the strategy, took off and it's just been a breakaway. They're the most cited cruise line in the world for anything related to Alaskan cruises, all because of both the cutting out irrelevant content and then fleshing out and building out topical authority, through content creation. So there's content creation on your site. Your site is one of the biggest factors that AI uses to determine who you are and when they should bring you up. But there are also third party and off-site validation, which is very important. So generating brand mentions and citations, really, it's about understanding two key things. So first is what does AI think about us now? You have to understand what language do they use, where do they see you fit within the market, who do they think you're for, and then you need to understand where are they getting the information because it's not always from your site. So doing something like a a brand sentiment analysis, is a great way to understand where they feel like your domains of authority are and where you wanna be, but you are not showing up. But then within that same analysis, you can see where are they learning. So the whole industry is, you know, talking endlessly about Reddit as the strategy for AI. YouTube is very popular. LinkedIn is very popular, and it's true. All those have big impacts on what AI says. However, the real key thing is for depending on your industry, you may see different sites. So for example, Jazz, HR is very important for this particular client. They're the third most cited, website to come up with answers. And so they should have a campaign to get on JazzHR or establish a better presence on Lever, in order to, have a bigger influence over what AI says. Now for enterprise brands, a really cool unlock is the partner marketplaces. So this isn't really accessible by smaller brands as much, but bigger brands, you can get into and you can have, partner marketplaces where your brand is going to be on, multiple listings of multiple marketplaces where you can have the messaging and it's treated within AI search, at least right now, as a third party validation of what your domain expertise is, who you're for, and how AI should talk, about you. And so this is really a fantastic way that partner marketers can help influence, the outcome of AI search. So going back to what Jody was saying, like, governance becomes key here, but everybody in the organization can have play a part in helping to influence what AI, is showing. So it it becomes really important to see what, AI is saying about you and, assessing that is is the first step in in understanding that. Pass it back to you, Before. I before I get into this, I just wanna say that, like, going back to the idea of normalizing how you talk about yourself and being consistent extends beyond just your own website and different pages or domains. Right? It extends to third parties and making sure that there's no confusion about, you know, who you are, what industry you serve, what your products are, etcetera, etcetera. So I think that's really important. And I also love that you are basically saying more is not more when it comes to content. It's about clarity, and I think about, like, using, you know, LLMs internally. We we are a cloud house at at at WordPress VIP, and we often are talking about Drift and how more context is it can be confusing, versus helpful. So I think that's a really, really valuable thing that you said, Paxton. Speaking of assessments, we've got, a fun offer for all of you lovely folks who joined us today. We're doing and we mentioned this in the last in the last session, but we're doing a a free joint AI discoverability assessment. And this is not there's a lot of these out there. You probably are getting LinkedIn, notes from different folks and emails and stuff. But ours is pretty cool. We're we'll do kind of a because we'll talk about it from two parts, like the infrastructure side and then the content optimization side because that's kind of our combined sweet spot. So if you're interested, we will do an audit of your brand. We'll tell you how AI sees your brand today and then, make some customized recommendations on on what you should implement to increase, your your performance there, and then a handful of things that we should that you should do right away. So the QR code is where you can access that that that link. We'd love to hear from you, and we'd love to help you out. Okay. Cool. So just shifting gears to our final topic, and please, go ahead and keep asking questions. I think we will have time for some q and a at the end. Our final topic is really around prioritization, and organization and accountability. So, I I know this is probably captain obvious, but not everything we talked about today is priority number one. You wanna break it at least into two buckets. The first one being quick wins, things that you can implement in weeks that can make a difference, to how you're performing, and it's kind of that low hanging fruit. And then, bucket number two, you know, really more of that foundational rebuild and owned layer, the things that are maybe a little bit more work but still worth doing to optimize your your position. So bucket number one would be, you know, things like making sure that, you're fixing wrong tags on your most important pages, that you're unblocking AI robots that maybe got locked out, by accident, which has happened, to some of our customers, patching holes in your site map and cleaning up any spots where your story or your narrative might be inconsistent across your domains or pages. And once you've gotten all that taken care of, then you can move on to some more about foundational work. That's where you'll wanna really think about rebuilding your whole organization and tagging system and making bigger decisions about what technology is running on your site, and and does it, provide the level of automation that you that you want and need. And then also investing in becoming a trusted resource on your core topics and the stuff that that is suggesting around third party optimization. And then you'll finally wanna make sure that you're writing real governance rules and managing to to those expectations. So this is kind of another way of seeing the same, a little bit more broken out, but I put it out this way in case you wanna screenshot it as as your road map that you can borrow. It's just broken into kind of, like, near term, medium term, and long term stuff, baseline kind of first, some of those foundational elements that we just talked through, then structure structuring your content, then building authority, and then finally making sure that you've got the right operational cadence that you can test, fix, retest, report, and and do that on an ongoing basis. Okay. Now for organization and accountability. I mentioned this in our last webinar and then touched on it, just a few minutes ago, but, AI discoverability is is a, is a team sport. We've got a lot of different folks that kind of have, that are cooks in the kitchen, so to speak, and they're all really, really important. But a lot of times when we, we are talking to customers or have new customers, they onboard, they don't have a single point of accountability, a directly responsible individual to to borrow an Amazon term. And that can be really problematic because there needs to be somebody who's sort of quarterbacking all these different, components, and and making sure that it happens in the in the right sequence. So oftentimes, we've got marketing contributing who's owning, you know, the overall narrative. We've got the content team who's really understanding and staying in tune with what questions your buyers are are asking and and creating content that answer those questions. We've got the SEO team. Often, SEO teams are the first to sort of raise their hand and take accountability for AI discoverability. Sometimes, not always, but that's the team that obviously needs to really map that content fan out and content clusters and making sure that we understand how our brand is doing relative to the competition, the web team who's really organizing our our site templates and hierarchy, etcetera, engineering team, who really make sure that our her website is performance, security, who's making sure our our crawler policy is really matching what we intend, and then finally, the governance team, who's creating those standards and enforcing those standards. So a lot of different folks. And, what our recommendation is and what we have seen is that it can live sort of anywhere in terms of accountability, knowing full well that there's there's responsible parties in in multiple functions. But it's just important to pick someone, have that person ultimately own performance, and work with the right folks to to make sure that, all these recommendations that Paxton and I are are recommending are implemented and in the right order. Okay. So as our final point, and if you don't remember anything from session number two, just remember that there are two parts of AI disc discoverability, and they're both equally important. If you have infrastructure and it's all really buttoned up but no content strategy, that means that your brand is readable, but it not it's not necessarily gonna be cited or recommended. On the flip side, if you've got killer content strategy but no infrastructure, it's a great story that no machine can read. So it kinda doesn't matter. Right? So both are really necessary, and they work together to get cited and recommended by AI. Alright. So I think some of us have submitted the poll results. Interesting. Yeah. It looks. yeah. It looks like, Go ahead. the majority voted that they're currently improving technical and content foundations Mhmm. with a close second oh, tied actively working to improve AI visibility and recommendations. Mhmm. And we've looks like we've got a handful kind of at the beginning stages of not sure yet, still figuring out where to focus first. Jody, is this consistent with what you're saying with with your clients? Yeah. I think everyone has accepted that this is really important. And then I think everyone has kind of, accepted that this is, different than SEO, because I think that at first, it was like, well, you know, which which parts of SEO still apply, which parts have evolved, and and how do you reflect that back in your practices? So I definitely have seen, those two, you know, people kind of being a work in progress, I would say. But it was interesting that only 5% said that they had a clear owner in place. That that seems to be true. And having an ongoing process is also another thing that folks sometimes forget. They think of it as a project. And once it's done, it's done. It's kinda set and forget versus an ongoing thing that you can continue to, like, increase performance on. What about you? Is that what you Yeah. anything surprising there? Yeah. No. I mean, I think I think it's pretty consistent. There seems to be a bit of a bifurcation between people who are, like, getting started and they're in it and people who are still at the very early stages. And I think the biggest bottleneck that I'm seeing in the industry to moving and like, advancing the strategy in AI search, it really centers around measurement. Because it's one thing to know that you need to do something, and it's another thing to know whether or not what you're doing is having an impact, what kind of an impact, and what kind of investment, more importantly, needs to be made in the organization in order to move, you know, grow the bottom line. And, Yeah. so that's great. That's what we're gonna be talking about, more next time. But I think it is the the big kind of unknown and thing to be figured out in the industry, which kinda makes these, results make sense. In recent, board meetings, I know a lot of my peers and and, before I joined, WordPress WordPress VIP, the board would actually ask, like, what are you doing about AI discoverability? You know, they would kinda see some of the macro trends around, unpaid pipeline, which is oftentimes a a result of organic search, and organic search traffic that has been declining for a lot of brands for a lot of time. They see that, and they're like, okay. Well, you gotta win it back. You gotta win your traffic back. You gotta win that unpaid demand back, or those unpaid customer customer acquisition back. And so they wanna know. And that's that's not something I would say or I don't remember being asked about that for SEO. It's like one of those things that now is like top, top, top, tippy, top priority for a lot of folks. Yeah. Alright. Cool. So we've got about three minutes for some questions. I know one thing that I caught that somebody asked is probably best directed to you, Pax. You say, we've got thousands of old blog posts. How do we know which ones are helping us and which ones are actually hurting us? Yep. So, first, I would start with that matrix that I shared where, I would go through my analytics, and I would pull visits to all of my pages, and I would look at the engagement rate of my pages as well. So, again, click through rate, time on page, bounce rate. And then, I would look for anomalies on that, and then I would also see what citations are showing up when I do this AI sentiment analysis. So I ask questions about who who this who my brand is for, what you say about my brand, all that, and then I see where it's drawing those answers. You'll find some things in there where it will recommend or say things that are incorrect about your brand and it's pulling from your own site. So those are some really quick wins. And then I would so after those quick wins, I would do something a little bit more laborious, which is going through, filtering by traffic, to see relevance, of my content kind of one by one. So it's a it's a slow process. It can be a bit of a slog, but, when you filter by traffic and engagement rate, that allows you to put, like, the highest priority first, and that would knock those out. Love it. Alyssa says Reddit and Quora keep coming up as place as as places, AI sites. Do we really need to show up there? And then Utsav says, I thought Quora is dead. What's your take on that? Really? I mean, I think Quora is dead to me, in terms I don't really use Quora or rely on Quora, but, it's it's just really about where is the AI getting its information. And so, if we can see, again, this is why it's really important that you do this assessment for your brand within your industry, because it may be that for you, it doesn't matter. But it could be pulling a bunch of information, about your brand from Quora, in which case you may need to go and either correct the record, add to the records, amend the record, Right. on Quora if that's where it's pulling from. So it really kinda depends on what's happening for your site in your industry. I love it. Okay. Great. Well, we're at the top of the hour. Just one final plug for session three, and I know a couple of folks were asking where you can get session one, and the link was shared. So if you wanna take a peek at that, go ahead. We really covered, like, how to establish a baseline, for AI search visibility. Session three will be all about measuring. And I know this is top of mind for a lot of people as they try are trying to figure out how to recuperate, you know, unpaid traffic and unpaid demand from their website. What should you measure? How do you know that you're making progress? So please join us on October 20. There's a a a link to register for session three, and we appreciate everybody joining us today. Thank you so much for your time and attention, and we'll see you next time. Take care. you.