{"id":17999,"date":"2026-07-29T15:24:16","date_gmt":"2026-07-29T13:24:16","guid":{"rendered":"https:\/\/incubeta.com\/?post_type=video-library&p=17999"},"modified":"2026-07-29T15:24:19","modified_gmt":"2026-07-29T13:24:19","slug":"the-digital-edge-s2-ep-11-solving-ai-invisibility-with-jessica-jacobs","status":"publish","type":"video-library","link":"https:\/\/incubeta.com\/sg\/video-library\/the-digital-edge-s2-ep-11-solving-ai-invisibility-with-jessica-jacobs\/","title":{"rendered":"The Digital Edge S2 Ep.11 | Solving AI Invisibility with Jessica Jacobs"},"content":{"rendered":"\n

In this episode of The Digital Edge, host Mark Reed-Edwards talks with Jessica Jacobs, Managing Director of Northern Europe and DACH<\/a><\/strong> at Incubeta about the growing reality of “AI invisibility” and why traditional marketing metrics fall short in the age of answer engines like ChatGPT and Gemini. Jessica explains how hallucination, siloed measurement, and cross-brand contamination can cause brands to lose customers before they even reach a website. She also outlines actionable strategies to audit and restructure existing content to ensure your brand stays visible, trusted, and cited in AI-generated search results.<\/p>\n\n\n\n

Read the Full Transcript Here<\/h2>\n\n\n\n

Mark Reed-Edwards<\/strong>: This is the Digital Edge from Incubeta. I’m Mark Reed-Edwards. This podcast is about how you can balance technology and humanity. How as AI eats the world, you can integrate efficiency with empathy. We’ll talk with leaders from Incubeta and across the industry as we traverse the digital edge into tomorrow’s world.<\/p>\n\n\n\n

Joining me today is Jessica Jacobs, Incubeta’s Managing Director of Northern Europe and DACH. Jessica and I talk about how most brands are still measuring performance in complete silos, even as AI adds a new layer of invisibility. The upshot? Traditional metrics can’t tell whether an AI trusts a brand enough to cite it.<\/p>\n\n\n\n

Let’s get into the discussion. Welcome, Jessica.<\/p>\n\n\n\n

Jessica Jacobs<\/strong>: Thank you.<\/p>\n\n\n\n

Mark Reed-Edwards: Can you tell me about your role at Incubeta and the markets you cover? And, you know, what does a day look like for a managing director spanning Northern Europe and DACH?<\/p>\n\n\n\n

Jessica Jacobs<\/strong>: Yeah, absolutely. I think at the core, my job is all about growth. That means growth for, uh, Incubeta’s business across the markets that I’m in, growth for the clients we serve, and of course, you know, the results, revenue, market share, not necessarily vanity metrics.<\/p>\n\n\n\n

I’m strategic by nature. And the way that I get there is by staying genuinely close to clients, and that’s– I think that’s just, that’s my focus. It’s an extremely client-focused role. I would say day-to-day over the last couple of years, I’ve essentially been focused on building our creative and strategic offerings across Europe, which actually led me to the restructure that’s created essentially the four markets in which I now operate, and those are referred to as Northern Europe and DACH.<\/p>\n\n\n\n

But within those, it’s the Nordics, it’s France, it’s DACH, and it’s Benelux. So it’s quite different in many ways, but that wasn’t arbitrary in design. It was so that clients in each of those regions actually get sharper, more localized attention rather than just a single simple one strategy that was trying to fit everything that we were doing.<\/p>\n\n\n\n

Right now that means that I take the opportunity to understand what is threatening our clients, what is driving their growth, and then making sure that I can bring the right specialists in to actually put us in front of them.<\/p>\n\n\n\n

Mark Reed-Edwards<\/strong>: It’s a fascinating area because there are different cultures. So how do you deal with that?<\/p>\n\n\n\n

Jessica Jacobs: Yeah. It’s definitely I would say one of my most precious gained skillsets is cultural sensitivity and understanding that, and it is something that I don’t think anybody naturally takes into their day-to-day. If you’re so focused on a single market, you tend to not know how much to take into those strategies.<\/p>\n\n\n\n

And what we’ve done is we’ve made sure that when we’re in market we make sure that we’re close to our clients so that we can really understand them. Expertise, however, deep expertise can come from anywhere. But we need to be able to balance that. So we walk the line by making sure that client servicing is something that is closer to where our clients are at, but then we make available to our clients the best of the minds across the world.<\/p>\n\n\n\n

Mark Reed-Edwards<\/strong>: Well, it wouldn’t be a podcast if I didn’t talk about AI. You use the term “AI invisibility.” Can you give me a real-world example, maybe a brand that thinks it has a strong digital presence but is effectively invisible to an answer engine? And what does that cost them in real terms?<\/p>\n\n\n\n

Jessica Jacobs<\/strong>: Yeah, absolutely. So this is a conversation that is increasingly growing, and we’ve definitely gone in with the term “invisible” so that it’s, you know, it’s a hard truth. So I won’t name a brand, but I’ll definitely refer to a bit of a vertical here to kind of call out the understanding around it. So let’s picture a kind of a well-known home appliance brand.<\/p>\n\n\n\n

They’re extremely strong on page one Google rankings, solid paid media, genuinely a nice website as far as website goes. They tick all the boxes. Now, let’s say we ask ChatGPT or Gemini, “What’s the best energy efficient washing machine for a family of four?” They just don’t come up at all.<\/p>\n\n\n\n

It’s citing a comparison by, for example, a blog, and let’s say two competitors instead are the ones that are coming up. That’s AI invisibility. That’s the example that we’re trying to put in front of every single client we’re speaking to. Dominant in the channel everyone spent probably decades optimizing for and completely absent in the one that customers are actually quietly switching to.<\/p>\n\n\n\n

And I think the cost isn’t some vanity metric ticking down, it’s losing the moment of consideration entirely. Before the brand even knows that the customer wasn’t in market, they’re just not on their site anymore.<\/p>\n\n\n\n

Mark Reed-Edwards<\/strong>: Let’s drill down a little bit further into that. So when a customer asks ChatGPT or Gemini a question that your brand should own, maybe you’ve been out there, you know, you’ve got blogs, you’ve got all kinds of pages out on the open web: but the AI answers with a competitor or a third-party source, is that traffic loss or reputation loss, or is it something bigger?<\/p>\n\n\n\n

Jessica Jacobs<\/strong>: You’ve said it. It’s something bigger. It’s a purchase decision that’s being made without you even being in the room. We were so used to competing for it, but we’re not even there.<\/p>\n\n\n\n

This is a deal that’s taking place, and we’re not even there. So with traffic loss, and this is a critical point, you can see it. And because you can see it, you can recover from it. But this is invisible, so the customer never even lands. It’s not a bounce, they’re just never there.<\/p>\n\n\n\n

You were not even there in the first place. So this is actually not a funnel problem. This is something that brands seriously need to address about how are you in the room invisibly?<\/p>\n\n\n\n

Mark Reed-Edwards<\/strong>: The funnel is a concept that goes back a long time, and it seems like AI and specifically these, you know, ChatGPT and Gemini have kind of upended that.<\/p>\n\n\n\n

Jessica Jacobs<\/strong>: I would say that they’ve brought a new dimension to it.<\/p>\n\n\n\n

So in many ways, the funnel will be there in a way that we still have these different phases in which consumers will live, and it won’t completely change and definitely not immediately, but it will be challenged. And what that new dimension will look like, that’s a consideration we have to take into account.<\/p>\n\n\n\n

So my question always back is, are brands actually thinking about that this is something that needs to come together? And what we’re often seeing right now is brands are almost treating this separately. They’re designing separate teams, or they’re putting it with a team that doesn’t necessarily cover the full existing funnel.<\/p>\n\n\n\n

But this is a change in that funnel and needs to then consider the full team that’s responsible for the funnel we know today.<\/p>\n\n\n\n

Mark Reed-Edwards<\/strong>: I’ve read some of your work on AI, and you said in one piece that I read that AI will hallucinate an answer using whatever data is easiest to parse, which kind of blew my mind.<\/p>\n\n\n\n

What does a hallucinated brand answer actually look like in practice, and have you seen it damage a client?<\/p>\n\n\n\n

Jessica Jacobs<\/strong>: Yes, absolutely, and I will actually give you an example. Everybody’s using the term hallucination, and the crazy part about this entire bit is that AI will look extremely confident, and that’s the dangerous part.<\/p>\n\n\n\n

In every single answer it will give you, there will be no element of it that says that it’s uncertain of what it’s telling you. It will make… It’ll be so self-assured, and it will cite whatever it believes is giving it the right answer. Now, I mentioned that I would actually give an example of where we’ve had to share this information with the client, and this was the hard part for them to understand.<\/p>\n\n\n\n

What they’ve cited essentially was the incorrect return policy. Now, in this case, AI cited a third-party retailer page, which we know large brands today have left a lot of that muscle, that advertising muscle to retailers, and essentially did some top-of-funnel work to retain the brand image. But what that now means is that often retailer pages are much easier for AIs to parse, and therefore it cited this old page.<\/p>\n\n\n\n

It gave the customer incorrect returning information, and who does the customer blame? Definitely not AI. It’s gonna reference AI and say, “But this is what it told me,” and the brand is ultimately going to feel the damage as a result of it.<\/p>\n\n\n\n

Mark Reed-Edwards<\/strong>: That’s far-reaching for a brand, isn’t it?<\/p>\n\n\n\n

Jessica Jacobs<\/strong>: I think it’s out of their comfort zones.<\/p>\n\n\n\n

I think it’s an extremely challenging new chapter for them to understand.<\/p>\n\n\n\n

Mark Reed-Edwards<\/strong>: Speaking of things to understand, most marketing leaders still measure success in clicks, impressions, and conversions, but none of those metrics capture, you know, whether an AI is citing you or ignoring you. So what should brands be measuring instead?<\/p>\n\n\n\n

Jessica Jacobs<\/strong>: And it’s not even that. They’re still measuring that in complete silos, and now there’s this added complexity that none of those metrics will tell them whether an AI actually trusts them enough to cite them. So what brands need to start tracking is “citation frequency” across all of the major answer engines, which also all have different methodologies. So the sentiment and accuracy of what is actually being said about them and the share of voice in AI-generated answers, specifically against their named competitors, they need to know all of that.<\/p>\n\n\n\n

What is the gap that exists within that? So think of it as like the new version of share of search, except now search is a conversation. It’s no longer just a simple query. This was a slightly more complex one, so I’ll go in a little bit more to explain also what we’re expecting to see looking ahead.<\/p>\n\n\n\n

I think we’ll see a couple of new metrics be formalized, and they’ll all start getting names. We won’t know the names, but we can definitely contribute to how we’re having conversations, so things like answer share. How often is a brand being cited across the number of AIs? What is the citation sentiment?<\/p>\n\n\n\n

Is it good? Is it favorable? Is it actually accurate? And are you covered? Do you have the coverage you want for the queries or the conversations that you want to exist in? But measuring honestly is only half of it, and we’ve seen this pop up significantly over the last couple of months, all of these measurement tools which is the right baseline step.<\/p>\n\n\n\n

But this isn’t just about building another dashboard and calling it done. The real win is actually in the strategy, and that is where the measurement component to us becomes a lot more of an advanced conversation. We think about things, and I think if you’ve heard one of my conversations, you’ll think I spoke about one example, like how does YouTube feed into this?<\/p>\n\n\n\n

And I try to explain to customers what we’re shown on YouTube is never random. It’s already shaped by many signals. It’s shaped by your relationships. It’s shaped by what part of the algorithm you’re going down on, and then it actually decides what you see. Now, AI answers are gonna have the same underlying logic.<\/p>\n\n\n\n

It’s gonna have a lot of these different insights and things that are feeding into it, and brands can step up the strategy. They can have all of that context feed in, and honestly, the brand who connects the dots on that, those are the brands that are gonna let the models win for them.<\/p>\n\n\n\n

Mark Reed-Edwards<\/strong>: So you argue that optimization needs to move upstream into scripting and briefing.<\/p>\n\n\n\n

What happens to the content that already exists though? You know, years of video, blog posts, white papers, and all that. Is it salvageable, or do brands need to start over?<\/p>\n\n\n\n

Jessica Jacobs<\/strong>: Absolutely salvageable. But what’s really important, and this is… I believe this is the bit that a lot of companies get wrong right now.<\/p>\n\n\n\n

Most of the back catalog, what you just referred to, all of this amazing content that exists that was built for human scanning a page. It wasn’t built for a model to extract it. So I look at things like white papers. Those white papers with all those killer stats buried in them on page 12 or page four– that’s completely invisible to LLMs doing extraction. No matter how good the insight is, if it’s just not readable, it’s invisible. So the fix is actually restructuring for clarity and for that readability. Clean claims, clear attribution, proper structured data. It’s not about throwing away years of content and starting from zero; it’s about making what you already have legible and then essentially what you’re doing moving forward following that similar structure.<\/p>\n\n\n\n

And one thing I also try to let everyone know is that these models, they haven’t been… they’re not learning on things that are being created now. They’ve learnt on everything that was created. So it’s actually more critical to fix what AIs understand of brands today because they’re using your historical footprint to make that decision.<\/p>\n\n\n\n

And if your brand has fundamentally changed over the years and AIs are citing everything from you in the past, what a horror story for us as people if all of a sudden our past was determining our futures. That’s what’s happening to brands. They can pick the content that you believe are the differentiators, restructure them, get them right, get them clean, get them clear, and then you take it forward.<\/p>\n\n\n\n

But AIs also look at real-time things, so some of these tools are pulling live data from the web in the moment that you ask. They’re not just working off the older, trained data. But either way, the same principle holds. Whether it’s baked into older content, it’s also freshly made content, both ultimately play a role.<\/p>\n\n\n\n

Mark Reed-Edwards: It’s an interesting development, where all of that old content can be repurposed. So it’s not necessarily throwing out the old stuff, it’s actually using that stuff and maybe, you know, this is a good thing. It brings new life to that content.<\/p>\n\n\n\n

Jessica Jacobs<\/strong>: Exactly. You said it so beautifully, and I can’t remember the last time we were able to introduce a new channel strategy and just port over old content.<\/p>\n\n\n\n

It was always designed to make you fail if you did that, but this is actually not the case. It’s designed to help you. You just have to slightly restructure and you’re ahead.<\/p>\n\n\n\n

Mark Reed-Edwards<\/strong>: You mentioned cross-brand contamination as a serious risk. Can you walk me through what it looks like when it actually happens?<\/p>\n\n\n\n

And is there a real or hypothetical scenario where an AI gets a brand badly wrong?<\/p>\n\n\n\n

Jessica Jacobs<\/strong>: This is a real risk and I did mention that, and I say this because we’ve run into a couple of these scenarios and if I now take a bit of a nod out of the portfolio I manage across multiple markets, the cross-brand contamination also happens because of the multitude of languages in which brands actually operate.<\/p>\n\n\n\n

And so let’s say a company owns a few different brands and they ultimately sit in the same category, and we know that there are many of these global organizations that ultimately have multiple competing brands. Maybe they’re sold under different names even in different countries. Now, AI is working off of this thin, messy data that can get it mixed up.<\/p>\n\n\n\n

Think about how us as humans, we’ve gotten this wrong, and machines are trying to learn off of everything we’ve tried to piece together. So the simple example, it could misquote. We’ve seen an example now where a premium brand’s price under another budget brand’s name is being misquoted because there’s references to how these brands actually connect in a single company.<\/p>\n\n\n\n

AIs don’t know how to differentiate out of all of the sources it’s citing from. It’s just pulling from data that it believes it’s structured to read from, and it synthesizes some of those responses. And in a way, like I mentioned hallucination, it’ll confidently get it wrong. Now, what happens even now is that here’s where the complication gets even bigger.<\/p>\n\n\n\n

If you operate with multiple languages in the mix of it, AI doesn’t even know how to handle the difference between a brand having a different reply in Dutch or in French or in German, and because these are often different teams that often can also decide what are their priority products, some products stay outdated.<\/p>\n\n\n\n

But if they’ve got a bigger claim to the citation, it actually pulls from all the information and just pivots it across the language that you want.<\/p>\n\n\n\n

So the short version of this fix is: it needs to be cleaned across the board. The consistency is a bigger conversation. A conversation must be held everywhere.<\/p>\n\n\n\n

AI needs to understand exactly what does a brand want from it in every market and in every language, and then you need to have the single way in which you can operate. Finally, define that single language, ’cause that consistency, if it’s not there, that is what will stop the model from actually accurately filling the gaps and either citing competitors or incorrectly citing your own brand, and I guess that’s the risk.<\/p>\n\n\n\n

Mark Reed-Edwards<\/strong>: So if you were advising a brand that wants to act on this right now, what would you tell them to do first thing tomorrow morning?<\/p>\n\n\n\n

Jessica Jacobs<\/strong>: It is as simple as saying run an AI invisibility audit. That’s genuinely it. It’s about essentially asking all the major engines the top 10 questions your best customers would ask you and where do you wanna then essentially show up.<\/p>\n\n\n\n

Now, yes, I would say it can be that easy, but there is a slight nuance to this which makes it more complicated. You need to understand in these top 10 questions right now also, if you have a gap, you’re not there, why does your competitor beat you? And this is why we’re saying, you know, get in touch.<\/p>\n\n\n\n

Get in touch with the partners that have the IP that’s currently really scouting across all of these areas, giving you a view of your gap, what you need to do to beat it, what are your competitors doing, but genuinely start there. Start with that audit.<\/p>\n\n\n\n

Mark Reed-Edwards<\/strong>: So let’s look ahead two or three years from now. What will a brand that gets AEO right look like versus one that ignores it?<\/p>\n\n\n\n

Jessica Jacobs<\/strong>: I honestly… I laugh lately only because the fear of actually predicting two to three years from now, it feels impossible to me. I feel like three months from now we’re gonna sit here and we’re gonna look at the conversation we have, and we’re gonna see that it’s probably moved past what we’ve said.<\/p>\n\n\n\n

But let’s be bold. Brands that get this right become the default answer. It is as simple as that. Cited, trusted, quoted directly by AIs and customers don’t even need to click anymore. Likely your website’s gonna act as something very different to what we know today.<\/p>\n\n\n\n

Only polishing one part of the strategy will eventually start quietly disappearing. They will see it, maybe not now, maybe not today, but they will see it quietly in the revenue numbers of their business. And honestly, if there’s a line I could leave people with, it’s for the first time in years, and it’s going to continue in the next two to three years, any smaller, sharper brand can now sit alongside the big household names.<\/p>\n\n\n\n

Because that is exactly what this opportunity is giving everyone, that the most known brands can get leapfrogged completely, and they will stay behind.<\/p>\n\n\n\n

Mark Reed-Edwards<\/strong>: So would you agree that if you thought the last two or three years things moved fast in marketing, wait until the next two or three years, that it will be faster than ever?<\/p>\n\n\n\n

Jessica Jacobs<\/strong>: Absolutely. It will be tremendously faster than what we even have right now. It’s exciting, but in a way that we have preparation right now. We are being introduced to the tools and the opportunities that will allow us to get there. And honestly, we started the year saying by the end of 2026, brands would lose 25 percent of their search results to AI, and we’re midway, and we’ve already seen it.<\/p>\n\n\n\n

So it’s moving faster than anybody has predicted.<\/p>\n\n\n\n

Mark Reed-Edwards<\/strong>: Well, Jessica, thanks for joining me on “The Digital Edge.”<\/p>\n\n\n\n

Jessica Jacobs<\/strong>: Mark, it’s been fantastic. Thank you for having me.<\/p>\n\n\n\n

Mark Reed-Edwards: Really enjoyed this discussion with Jessica, and I come away with this: brands need to measure their share of voice in AI answers against named competitors. Sure, they have to do that, but they also need to identify gaps. As Jessica says, this is the new version of share of search. Thanks for being here today.<\/p>\n\n\n\n

I’m Mark Reed-Edwards. Join me on The Digital Edge next time.<\/p>\n\n\n\n

Guests<\/strong>: Mark Reed-Edwards; Jessica Jacobs<\/p>\n\n\n\n

Discover More Episodes of The Digital Edge<\/h2>\n\n\n\n