“LLMs tend to place more trust where they detect topical depth, consistency and clear signals of authority and specialization,” says Laura
Welcome back to SEO Confidential, the series where we interview some of the most authoritative voices in international SEO for an open conversation about digital marketing, growth strategies and the evolution of search in the age of artificial intelligence.
The goal is one and one only: understanding what really matters for business today, beyond the hype.
In this episode we chose to sit down with Laura I. Abreu, a professional with over twelve years of experience across Europe and the Asia-Pacific region, built between Hong Kong and Australia. An international career that has allowed her to work on very different markets, adapting SEO, advertising and lead generation strategies to complex, ever-changing contexts.
Today Laura lives in Barcelona and was recognized as SEOFOMOer of the Week by SEOFOMO, the newsletter curated by Aleyda Solis and followed by over 40,000 professionals worldwide. In 2026 she will also be part of the international judging panel of the European Search Awards.
She is also the founder of Global and Search, a platform dedicated to global search marketing for professionals and companies, SEO and PPC, built on community, knowledge sharing and international collaboration.
In this interview we will talk about AI Overviews, LLMs, AI answer engines, GEO, metrics that actually matter, authority, web reputation and business decisions. A down-to-earth conversation for anyone who wants to understand where search is heading and how to stay relevant, beyond short-lived trends.
That said, let’s get started. Enjoy the read!
“The most important metric is always the same: are we bringing customers to the company? We’re talking about qualified leads, sales, margin, real impact on revenue…”
Many studies, such as the Datos report, show that organic traffic has not collapsed and that Google remains the main entry point to purchase journeys. Do you share the growing hype around GEO and LLM optimization, or do you consider it, at least in part, overestimated compared to real business results?
I believe a good strategy should account for both Google and LLM optimization. I don’t really understand the fuss that has built up around having to choose one or the other. I’m very curious about what gets shown in LLMs, because it’s new for all of us and it can still help drive more traffic to the website.
For example, I’ve found that FAQs can play an important role on your site if you answer questions in a conversational way, adding context for readers and also helping LLMs understand the content better.
But yes, as Patrick Cox recently stated in his article on Ahrefs, Google currently sends about 190 times more traffic to websites than ChatGPT does. On top of that, ChatGPT’s CTR is 96% lower than Google’s, so…
Today there is a proliferation of tools promising to measure brand visibility in AI answers and conversational systems. From your operational standpoint, are they genuinely useful for guiding business decisions, or do they remain, at least at this stage, rough estimates that risk creating more confusion than value?
I think it’s very important to start measuring brand visibility in AI, but without losing focus on real business impact, given that, as we saw in the previous question, traffic and revenue coming from LLMs are still very small compared to traditional search.
That said, I consider these platforms useful as directional indicators, but I wouldn’t yet take them as definitive metrics for making business decisions.
Personally, I always like to combine this kind of tool with the data provided directly by Google Search Console, particularly in the Search Appearance > AI Overviews section, to assess the real impact in terms of clicks and traffic to the site.
AI Overviews and LLM-based systems are intercepting a growing share of informational searches. For a conversion-driven company, where is the most delicate battle being fought today: in the loss of clicks, or in the ability to be present at the decision-making stages that lead to purchase?
I wouldn’t frame it so much as a loss of clicks, but rather as a matter of controlling the narrative and being able to influence the purchase decision. In a context where the audience is increasingly fragmented, I wouldn’t focus exclusively on Google, but would also work on other channels in parallel.
For example, in certain industries an SEO strategy adapted to platforms like TikTok, or collaborating with local influencers, can generate a greater business impact, depending on the type of product or service you’re selling.
Someone arriving from an AI answer is often an already informed user, close to a decision. For an entrepreneur, what is the most costly mistake you see today when trying to capitalize on this visibility without a solid technical and business structure behind it?
If I may, I don’t entirely agree that people arriving from an AI-generated answer are often already informed and close to a decision. Many people also use LLMs at the exploration stage, when they’re still trying to understand the context or weighing different options.
As for the second part of the question, for me the same principle as always applies: if you launch a lead generation campaign but the business doesn’t have an adequate structure to support it, the brand experience ends up being negative. The risk is not only losing that customer forever, but also compromising the reputation we’re building as a brand.
Laura, we agree that today it’s no longer about who “shows up” but who gets chosen as the answer by AI. So, from your point of view, which signals actually trigger the models’ trust in a brand, and which ones are instead systematically overrated by companies?
In my view, the key is offering real context and depth about what we do: working on FAQs in a conversational, natural way, having well-implemented schema markup, a complete and clear About page and, in general, a site that covers all the relevant information about the product or service in a solid, consistent way.
LLMs tend to place more trust where they detect topical depth, consistency and clear signals of authority and specialization.
What many companies overrate, in my opinion, is the idea that there’s a specific “trick” to optimize for AI. But we’ve already seen this with SEO: there has always been someone looking for a shortcut to appear at the top of Google!
With AI agents comparing, deciding and buying on the user’s behalf, how much does webutation weigh today compared to owned content, and what mistakes do you most often see from brands that keep telling their own story without being confirmed by third-party sources?
I believe that today a site’s external reputation (so-called webutation) matters more than ever. The most common mistake I see is continuing to focus only on self-promotion, without building real credibility outside your own channels. Strategically using what we know as digital PR is essential.
Very often these mentions on other sites are earned in person, by creating authentic connections, attending events and generating real trust. Of course, what a brand says about itself matters, but AI cross-references sources and trusts far more when third-party validation exists.
With the arrival of advertising inside ChatGPT, the risk is confusing paid visibility with real authority. Many brands are already thinking about shifting budget toward ads in conversational systems. Where do you see the greatest danger? In depending on a channel that stops working the moment you stop paying, or in not investing enough to become the source AI chooses even without sponsorships?
The truth is I don’t see it as very different from what happened with Google: at the beginning it was a purely organic search engine, and then it integrated advertising. It was a logical evolution.
And as we’ve already learned with Google, advertising can accelerate results, but it’s authority that sustains them over time. That’s why it’s essential to work on your brand, your reputation and your authority in your industry.
When a client sees traffic dropping and concludes that SEO no longer works, which data point or result would you use to prove that the business impact still exists?
The first thing I would do is analyze why that traffic dropped and, with humility, accept that maybe we could have done something better. Being transparent and direct with the client is essential. From there, I would dig deep into the data and, if needed, I would also compare notes with other experts from the SEO community on X or LinkedIn, to get a second or third opinion on our hypothesis.
As for the real business impact, I would assess together with the client whether the volume of qualified leads or conversions has decreased, since those are what truly sustain the business. So I would check the conversion rate, branded search and the weight of organic traffic compared to other channels.
And if we see that recovery work is needed, I would consider opening or strengthening other channels, depending on where the audience is, while keeping an SEO approach (for example Instagram or TikTok), or I would increase investment in Google Ads to support the business.
In the end, it’s not about defending one channel against another, but about protecting the results.
If rankings and traffic volumes are no longer sufficient indicators, which metrics should entrepreneurs and SEOs watch today to understand whether their search presence is influencing purchase decisions? And which signals, on the other hand, are still overrated even though they say little about the real value generated?
For me, SEO doesn’t work without a clear business strategy behind it. You can have traffic and great rankings, but if they’re not aligned with concrete goals, they don’t generate value.
The most important metric is always the same: are we bringing customers to the company? We’re talking about qualified leads, sales, margin, real impact on revenue. That’s why it’s essential to have a direct dialogue with the sales department, with the people handling purchasing and with CEOs, to understand what is actually driving results inside the company.
A principle I like to apply is “under promise” and “over deliver”. The only way to keep a client is to make sure they see that, thanks to you, their business is growing. Period.
What is still overrated are vanity metrics: traffic without intent, rankings taken in isolation, or visibility metrics that only serve for a LinkedIn post and collecting lots of likes.
With AI Mode and query fan-out, AI doesn’t “pick a page” but assembles answers by taking pieces from different sources. In practice, which investment really puts a company in a position to enter those reasoning processes: architecture and structured data, knowledge graph, block-style content like FAQs and comparisons, or external proof of authority?
For me it’s not about choosing a single investment, but about consistency. Site architecture and schema markup help, but on their own they’re not enough.
What really makes the difference is having well-structured content on your own site and, as much as possible, concentrated in one or a few pages, rather than fragmented across multiple URLs. This makes it easier for AI to understand the full context and allows it to extract coherent snippets.
And of course, all of this must be supported by external authority. Without real credibility outside your own site, it’s hard for AI to consider you a relevant source.
Today the traffic generated by LLMs is still limited, comes with margins of error and doesn’t always show clear conversion performance. In this context, is investing in answer engine optimization already an operational priority for companies, or does it remain, for now, an immature area compared to revenue goals?
I believe LLMs are still a “green” channel, in full evolution, especially when it comes to measuring real revenue. It’s still early for them to be used massively by the general public. It’s important to understand them, monitor them and get prepared, but without shifting focus away from the channels that actually generate business today.
For me it’s a matter of balance: companies should start optimizing with an eye on how they appear in AI systems too, but without dedicating more time to it than necessary, considering that Google still remains the main engine of the funnel.
In a phase of such rapid transformation of search, which professional and strategic choices do you consider essential to stay relevant over time, without chasing every short-term change?
In a constantly changing context, I believe the essential thing is to never lose focus on generating real business. Because without that there’s nothing, and as professionals we must make sure we at least triple the investment the client makes in our services.
To do that, the key is to deeply understand the business model, its target audience, the consumer’s needs and concerns, and to figure out how to work on visibility, topical relevance and true online authority.
You need to know how to distinguish what’s worth listening to and what’s superfluous for the client. When in doubt, test when it makes sense, measure the impact and make decisions based on results, never on trends.
Beyond self-promotion: build real credibility outside your usual channels
The interview with Laura I. Abreu leaves us with a clear message: in the new search ecosystem, the distinction between “traditional SEO” and “AI optimization” is a false dilemma.
What really matters is substance. The ultimate metric has never changed: are we bringing customers to the company?
On this I’m certain: we need to shift our gaze away from vanity metrics (traffic without intent or rankings for their own sake) toward the data that carries weight: qualified leads, conversions and sales, to put it plainly.
To get there, Laura’s recipe is pragmatic: constant dialogue with sales departments and CEOs, an “under promise and over deliver” approach, and a technical structure that makes LLMs’ work easier.
There’s no need to fragment content across a thousand URLs; what’s needed is depth in pillar pages backed by real external authority.
And yes, because if the world outside your site isn’t talking about you, AI will have no reason to consider you a trustworthy source.
I think that in a market obsessed with the latest “trick”, the obvious often gets lost:
artificial intelligence isn’t changing the rules of business, it’s just raising the quality bar.
SEO today is no longer a matter of “presence”, but of certified relevance. The data tells us that users arriving from an LLM are looking for answers, not slogans. The real opportunity isn’t climbing a ranking, but becoming the logical, unassailable choice.
My advice? Less self-promotion, more solid data and a strategy that speaks the language of the company’s bottom line. Trust me: only those who generate real value survive the hype.
A heartfelt thank you to Laura for this valuable conversation and for sharing her “business-first” vision, exactly what the industry needs today, in my view.
We’ll see you next week with another exclusive interview on SEO Confidential: we’ll have a truly special guest with us. An appointment not to be missed!
#avantitutta
