The AI Problem That's Costing You Sales

AI platforms like ChatGPT, Gemini, and AI overviews inside Google Search are recommendation engines, not just search engines. They run background searches, they synthesize sources, and they build answers. When you're operating at enterprise scale, simple misconfigurations can make an entire product catalog invisible to every AI crawler simultaneously. And with only about 60% overlap between traditional search and AI visibility, strong organic rankings don't necessarily guarantee strong AI search performance. Well, at Exposure Ninja, we've helped a ton of enterprise businesses with their AI search visibility. So, today we're going to talk about how to build an enterprise AI search strategy and where it can differ from smaller business AI search strategy. Anytime you're improving your AI search visibility, the first step is generally an audit to get a sense of exactly where you're starting from. This is no different whether you're a tiny business or a massive multi-billion-dollar global enterprise. But of course, when you're at the enterprise scale, your audit needs to be that much more detailed. When we're doing audits for enterprise, we're not using manual prompting, i.e., going onto Google Gemini and asking questions and seeing what it says about your brand. Instead, we're using professional audit tools like Semrush's AI visibility toolkit, Profound, or Peak.ai. One of the great things about these tools is they give you metrics so you can track your performance over time. Of course, with AI search, the answer is different every time. So, you don't want to take your visibility just based on one prompt and one answer. You want to measure how your visibility is changing over time and for different types of topics. And overview allow you to do this. You get a visibility score, you can see the number of citations, and you can see your performance for different topics over time. Now, whilst there's not perfect crossover between traditional search and AI search, there is some crossover. So, you will need to continue monitoring things like your keyword visibility cuz if you're ranking better for more keywords, then you're likely to show better in AI search. But these new AI-focused metrics are another layer of things to track and monitor. One example, when we were working with a global skin care brand, we found a whole bunch of technical SEO problems that we were able to fix and significantly improve their visibility in AI search. Even before we started working on the AI search stuff, just fixing those technical issues, things like duplicate titles, duplicate content issues, pages with no links, stuff like that, improved their AI search visibility before we even got to the AI search specific stuff. Generally, when you're doing an AI search audit for enterprise, you don't want to be going in and doing individual prompts. How do I show up for this question? Or what if I ask ChatGPT this thing? Because the AI gives you different answers every time, and if you're working across multiple product lines in multiple territories, that's never going to give you enough information. When we're doing AI search audits for enterprise businesses, we are using professional AI search visibility overview or Profound or Peak.ai. The great thing about these audit tools is that they give you scores so you can track how your performance does over time. For example, overall AI visibility score, and you can see how that's moving against your competitors. You can see the number of citations that you're getting, and you can see where those citations are coming from, what your visibility is like for particular topics. Now, whilst there's not perfect crossover between SEO visibility and AI search visibility, there is about a 60% crossover. So, we still need to track, monitor things like keyword rankings because if you're ranking better on Google, then typically you're going to be showing up better in AI search. But these new AI search visibility metrics are the next layer that you need to track and measure your performance against. Now, often doing this enterprise AI search audit will give us some immediate insights about things that can be improved about the brand's website. What we'll often do with enterprise businesses is we'll recommend a pilot to run in one particular territory or with one particular product line. For example, let's change the structure of these pages or let's tweak something about the website in this particular area. By rolling that out as a pilot with a single product line or a single territory, we can develop a playbook that can then be expanded across other product lines and other territories. That means that when we're going to the other marketing teams in different regions, we're going with here's what works, we've tested it, here's the before and after. That tends to get much better buy-in than if we're trying to experiment with different teams in different countries with different agendas. The second step is to fix the technical foundations. We said there's a bit of crossover between traditional search and AI search visibility, and technical audits are always a part of a good SEO plan. But even without that crossover, these AI search tools also like to crawl websites that load quickly and are free from errors. So, it's a good idea to run a full technical crawl using tools like Screaming Frog, the regular Semrush site audit, or Google Search Console. You're looking for things like crawl errors, redirect chains, broken links, broken pages, duplicate content, missing canonicals, Hreflang errors. You also want to fix any on-page signals like duplicate H1s or missing H1s, meta descriptions, image alt text. We covered this stuff recently in how to do an SEO audit, so go and have a look at that video if you want to dive deeper into this area. There is an AI search-specific layer to all of this as well, something that can really help with these AI tools so they can crawl your site nicely. So, if you use Google Chrome, there's an option in developer tools to disable JavaScript. And this can be interesting cuz this is really how these AI tools are actually seeing your website. So, here I am on dji.com, and if I just reload this page, you'll see what the experience looks like. So, we've got this nice big header image here with the DJI Mavic 4 Pro. We've scrolled down, we can see each product with its own image, a bit of information about it, not a huge amount of content on this page. We've got this sort of a scrolly video thing. Got a bit more info here, and then we've got that stuff towards the bottom of the page. Now, if we go into Chrome DevTools and we disable JavaScript, this more closely approximates what a an AI tool would see when it visits this page. So, we've still got this above the fold section here, but there's no text. If we scroll down, we've got no text about the product that this is talking about. Same here, no text there. So, you can see that these AI tools might struggle to understand the context on a page like this because they're not seeing everything that a web user would do with JavaScript enabled. Some other critical issues that we see enterprise sites making are things like robots.txt blocking these AI search crawlers. Don't block GPTBot. Don't block ClaudeBot if you want your website content to appear in the messages that those tools send to their users. Make sure that you're using schema markup properly. This is the structured data, the hidden structured data on the page that tells these AI tools the type of content that they're looking at in a very standardized way that they understand. Make sure that your web pages load quickly as well. Make sure that you're up to speed with Core Web Vitals. We've found that AI search traffic can convert at up to five times the rate of regular search traffic. But these AI tools tend to favor web pages that load quickly because they're going off doing loads of research to take those answers back to the users. So, they don't want to be waiting for ages if your website takes a long time to load. And users obviously don't want to wait for ages if they click through to your website as well. On the topic of speed, one of the tools that we use most to help us is WP Rocket. And actually, this is a tool that we pay for for the businesses that we acquire. If your website is operating on WordPress, WP Rocket is a really nice way of speeding up the site, and it works pretty much straight out of the box. This is useful for visitors, obviously, because lower load times mean higher conversion rates. But it's also great for traditional search visibility and AI search visibility as well. Now, WP Rocket have kindly sponsored this video, but actually we pay for this tool anyway for the businesses that we own and acquire. So, it's great that they're sponsoring this video, but we'd be recommending this tool anyway. Now, one of the challenges for enterprise businesses is that if they're going through each of the different websites that they own and recommending particular speed upgrades, that can be a huge [snorts] backlog of dev tasks. If you can get WP Rocket installed, it can blaze through a lot of that stuff automatically, meaning you're not having to coordinate with individual developers. Let me just quickly show you how it works. So, I've got it installed on the Elite Renewables website. This is a site that we own. And pretty much straight out of the box, WP Rocket does a really good job of setting you up and improving the load times of your website. It allows you to monitor the performance of each of your pages over time. It optimizes things like CSS. In fact, this is one box that you do want to tick. So, when you load it automatically, you won't see optimized CSS delivery. If you click on that, then you make sure that remove unused CSS is checked, [music] and go and save changes. That will help your pages speed up quite a bit. It can optimize your media, so your images and videos as well, including lazy load. So, that means like the images at the bottom of the page don't load until someone's scrolling. Now, in order to avoid the pages jumping around, you can add missing image dimensions, which which means that there's like like a placeholder there. So, the page doesn't jump up and down, which is something Google doesn't like for Core Web Vitals. It also has caching built in. This serves pre-built HTML to crawlers immediately, so it doesn't need to wait until that loads. Now, for the AI bots, that means that they get complete readable content without waiting for JavaScript to load. It also rebuilds that cache automatically when the content is changed, so you don't have to manually intervene or add developer tickets to your dev team. One of my favorite sections of the WP Rocket plugin is the new Rocket Insight section. This gives you really focused feedback on exactly what is slowing down your website. So, you've got this Rocket Insights tab here, which shows you the performance of each of the different pages that you're tracking, and you can add more pages to your tracking if there are particular pages or particular types of pages that you think are particularly important to you. So, here, for example, we've got the homepage. We can see this is performing well across the four core web vitals metrics. Some of these other pages aren't doing so well, though. The heat pumps statistics 2026 page isn't doing so well, particularly with the time to first byte or the TTFB. So, what do you do about this? Well, if this is a common issue across your website, you'll see it showing up over here on the right-hand side in your Rocket Insights score. So, this is the average of all of your monitored pages. Underneath that, WP Rocket will recommend particular sections of their settings, which might be able to help. Now, in this case, the time to first byte can't be resolved by anything that WP Rocket is doing, but it has made a couple of recommendations, which will impact the largest contentful paint and the cumulative layout score. In some cases, these are just as simple as clicking a button to activate them, but it takes you through to the section in the plugin where you can control this setting. Let me give another example with a really straightforward insight. This is on the Pyramid Eco website. So, on the Pyramid Eco website, uh we're doing pretty well, although time to first byte is also an issue here. If we go down, Imagify is an image compression plugin that's part of the WP Rocket ecosystem. Rocket CDN is their content delivery you can add those on to your account if you want. Um in this case, we've got the self-hosted Google fonts. This is a very straightforward fix. If I just click on activate, it's going to take me down to the font section. I can toggle back to self-host my Google fonts, which means that the fonts are served directly from the server, so when people land on the website, they don't have to then go and load the Google fonts and from Google servers. If I just click save changes, that will now be implemented, and then that recommendation should disappear from the list. So, with some of these changes, it really is as simple as that. And these recommendations on the right-hand side of the WP Rocket plugin just take you straight through to the section in the plugin, which is going to be best for you. Um so, this is great if you don't have a massive amount of technical knowledge, because some of these things can be super quick to do. Others, you might want a developer on hand, things like uh removing unused CSS. Sometimes, if you do that, you can cause some issues on your website, so best to be careful with it. Um by all means, do it, test it, maybe do it in the evening or do it when you've got a developer on hand, so you can undo it if you need to. Um but the uh the Rocket Insight section I find really useful. But regardless of the plugin or system that you're using, website speed is absolutely crucial for AI search visibility. So, a couple of ways of monitoring this, you can use PageSpeed Insights, which will give you feedback and a bit of a breakdown, or you can use a third-party site like GTmetrix, which is pretty industry standard, which shows your performance and gives you this nice waterfall, so you can see how individual components on your site are adding up to delay your load speeds. And by the way, if you like the look of WP Rocket, you can check it out at wp-rocket.me. That's wp-rocket.me. And I've also got to say, it's really good value. So, even with a multi-site subscription, it's $299 per year for up to 50 websites. The third step in your AI search strategy is a big one, because it's normally two steps, but we've rolled it into one to make it simpler, and that is content and your off-site visibility. Now, we've spoken about this in recent videos, so we're not going to go into loads of detail here. The principles for AI search visibility for enterprise and smaller business are actually fairly similar in this regard, albeit the enterprise is going to have more topics to cover, more product lines, more languages, more geographies potentially as well. One of the core principles behind your website content for AI search is going to be the query fanout concept. When you're searching for something or when you're trying to appear for a particular topic in AI search, remember that these AI tools are breaking that query down into multiple subqueries in the background. They're running these searches in parallel using what's called query fanout. Let me give you an example. So, here I am on Perplexity. I love using Perplexity to illustrate this, cuz it actually shows you those underlying queries. I've just searched or asked "Best drones for agricultural spraying," and you can see what it's done here is it's run three different searches in the background. "Best agricultural spraying drones 2026, DJI Agras T50 specifications, XAG agricultural spray drone models 2025." That's interesting. It's then reviewed a bunch of websites and compiled this answer. Now, of course, if we want to get visibility for a phrase like this, we want to make sure that we have content on our site, potentially targeting each of these and any other related topics that are going to be relevant that these AI tools might search when they're compiling an answer like this. Often, that means producing more content about different topics or making the content that we have much deeper, so it covers lots of these subareas. We've talked extensively on the channel about this. There's other videos that we've got about this topic specifically, so I'm not going to go into a ton more detail about that now. Often, this will lead us to recommend a topic cluster approach there, where we have a pillar piece of content, which is about the central topic. This is like the ultimate guide to drones for agricultural use, and then the topic cluster approach, where we have supplementary pieces of content published on the site. "Best drones for agricultural spraying, agricultural drone specification, agricultural drone battery life, agricultural drone range." All of these subtopics can then link into this pillar piece of content, and that means when these AI tools go and search online or they search through our website, they're finding all of the information that they need for these individual searches. That gives us essentially multiple tickets in the raffle, because there's lots of times that these AI tools can refer to our website, because we've got this topic covered from all angles. And don't forget, if you want some help with your AI search visibility and you want the team at Exposure Ninja to carry out an AI search audit for you, including a strategy to dominate in your space, these are very competitively priced, because we use them as loss leaders to find new clients to work with. Cards on the table, being completely transparent, once you see the audit, you will absolutely love it and you'll want to work with the team. That means they're very competitively priced, and they are fantastic. Just last week, I was in a room with a multi-billion dollar collection of brands that team at Exposure Ninja delivered the search strategy to this group, and they were blown away. So, go to exposureninja.com, find the contact page, drop us a line, let us know you'd be interested in an AI search audit, and we'll take care of you. But as well as the content on your website, we also need to think about how you're being seen across the internet, and this means getting featured in high-authority publications. These AI tools absolutely love citing from pages that have high authority and [music] experience. Again, going back to this agricultural drones thing, look at these websites that are being linked to. These are professional sites that have a good level of expertise about the topic. They are detailed articles, and they are clearly written by people who know what they're talking about. Now, that's not to say that this content couldn't be improved, it absolutely could. You could have authorship, you could have a bit of information about who's written this article and their qualifications or their experience level, but these are highly authoritative sites in their niche. So, if you want your brand or your product or your service to be mentioned in the response, it really helps to have visibility and be recommended in this type of content. Often, that means using digital PR. So, for example, inside Exposure Ninja, we have a team that identifies the publications that we need to get our clients referenced in if they're going to be recommended in these AI responses. They will then reach out to these publications or the editorial or writing teams behind those publications to get our clients featured in those articles. So, that means when these AI tools go out to these third-party websites to see which products to recommend to the user, they keep seeing you and your products being recommended. That increases the likelihood that they will recommend your product to the user. But this leads us onto something that enterprise businesses often really struggle with, and that is positioning. When these AI tools are going out and doing their research and finding mentions of your product or brand all around the internet, what do they say about your brand when they take those recommendations back to the user? And that's where your positioning comes in. What are the concepts that you are really closely aligned with? What are your values and the things that online content talks about when it's talking about you? This can be really difficult for enterprises, because often, it's different for different product lines. For example, we work with enterprise clients that want to be known as being innovative, but the problem is nobody's told their marketing team that. So, the head office wants to be seen as innovative, but all of the content that the marketing team has been has been putting out is about being really effective or being really price conscious. Often, these types of conversations can actually go quite far up in the organization, where we have to sit the executive team down and say, "Okay, what do we want this brand to be known for? Let's get a few key concepts that we want to be associated with. For example, when we work with The Ordinary, they want to be known for being good value and scientifically backed. That means when we're producing content for The Ordinary's website or we're getting them coverage in other websites, we need to make sure those concepts of good value and scientifically backed are seen everywhere. That means when the AI tools are going out and doing their research to compile an answer for the user, they're seeing these links concepts all the time. The Ordinary, good value, scientifically backed. The Ordinary, good value, scientifically backed. When they take all of this information back to the user, what do they say? They say, "Oh, The Ordinary, good value, scientifically backed." And that's the type of clarity that you need about your brand or your products or your services if you're going to get a consistent message delivered by these AI tools to their users. Now, often this is a big and potentially messy piece of work. We find ourselves facilitating these conversations with enterprise clients because it involves lots of different teams. We actually built something called the brand and positioning accelerator designed for this purpose. So, we'll go and talk to a brand's customers. We'll find out, "What do you think about this brand? Why do you choose this particular product over a competitor?" We'll take all of that information, talk to the internal team, and eventually over the course of a month or so, we'll come up with a clear brand positioning message that we can then use that can cascade through all the other marketing to make sure we're absolutely nailing how these AI tools are communicating about your products or service. In fact, this is such an important topic, we even built a tool called Mine My Brand, which you're going to go and have a look at at minemybrand.com. You just put your website in there and click start mining. The tool then goes and talks to the AI search platforms to work out, "What do they think about this brand? What are the core concepts that they associate with this brand?" Think of it a bit like, you know when you get the interviewer walking along the street asking people what they think about stuff. Mine My Brand is kind of like that, but for AI tools. It gives you descriptors like innovative, sustainable, reliable, efficient, eco-friendly if those are relevant to your business. It also gives you brand performance scores on how well your positioning is established in the market. Once you've got a clear positioning, it's then time to take that to the market and try to get coverage in lots of publications. You can use tools like SEMrush's AI visibility overview. If you go down to the source opportunity, these will be the websites that are referring to your competitors but not referring to you, and you can identify opportunities to be seen in these articles. In one client example, we helped a brand go from being visible for 45 different topics in AI search visibility to over 110, and it was by doing exactly this, being super clear about what they needed to be known for, by handling all the on-site stuff, and then getting them coverage in lots of different publications so that when these AI search tools went and did their query fan out, they found content on our client's site, but they also found content on third-party sites referring to this client. They understood more about what this client offered, the products and services and queries that they were relevant for, and significantly increased the visibility. Now, of course, if you're in enterprise marketing, you'll know that one of the most difficult things to do is actually getting all of the teams aligned and pulling in the same direction. When we're working with enterprise businesses, we typically find this is one of the areas that their internal marketing team can find the most frustrating. The key that we found is when you're communicating with each of the different teams to frame it in a way that makes sense to them and helps them get to their goals. At the risk of oversimplifying this, part of it is about how you frame the requests that you want made to each different team. For example, with to the website development team, when we need them to make changes, we're not really talking about AI search, we're talking about crawl budget and website health. If we want to get changes past legal, we're not talking about, "Hey, let's talk about AI search." We're talking about brand reputation and risk here. We're already being described by these AI tools. Can we shape what they're saying? When we're talking to the brand team, we're emphasizing that inconsistent AI recommendations are damaging brand perception. And if we get this right and get everyone bought in, we can shape how these AI tools, which are being used by billions of people on a daily basis now, are perceiving our brand. And of course, when we're talking to company leadership, we're talking about the fact that this traffic from AI search is so valuable in revenue terms, and that we want to beat our competitors for the share of search and the share of voice. Typically, how we structure this is in a 90-day initial project where month one, we're diagnosing and fixing the foundations. Month two, we're starting to build out the content and outreach. This is going to take many, many months, but we start to build those foundations of the content and outreach in month two. By month three, we're starting to get some early data, measuring, reporting, and going back to leadership showing, "Hey, this stuff is working." So, as you can see, enterprise AI search strategy is not the same as small business AI search strategy. Yes, there are some of the same components, but there's a lot more to think about and to be conscious of, particularly in the execution. The four steps that we've looked at today, starting off with the diagnosis and that AI search audit, then fixing the foundations, working on the content and getting us coverage on other websites, and then measuring, reporting, and aligning the organization behind this plan. And remember, with enterprise, it's often the technical layer which is where so many sites are losing. Don't forget the impact of speed, not just for AI search, but for usability, for traditional search visibility. This is so important. And don't forget, if you're optimizing speed and your website is on WordPress, WP Rocket, who are sponsoring today's video, are the tool that we use inside our own acquired companies because it just works straight out of the box. Thanks for watching today. If you enjoyed this video, consider subscribe. Talk soon.