The Future of Search Engines and Their Impact

Introduction

There was a time when getting your website found was the hard part.

Invest in SEO, create useful content, earn a few decent backlinks and, assuming you did things properly, customers could find you when they needed you. This became the norm, we grew to understand how this worked, and though there still appeared to be an aspect of smoke and mirrors to it, the concept made sense.

Now we’re entering a world where your next customer might never visit a search results page at all. Instead, they ask ChatGPT for recommendations. They use Google’s AI Mode to research suppliers. They ask Gemini to compare products. They get an AI assistant to shortlist agencies, compare services or find the best solution to a problem before they’ve clicked a single link.

This fundamentally changes the old paradigms, because suddenly the challenge isn’t just visibility. It’s understanding.. Can AI systems understand who you are? Can they understand what you do? Can they trust your information enough to recommend you? And increasingly, can they interact with your business on a user’s behalf?

Those questions sit right at the heart of two topics we’ve been spending a lot of time researching recently: Generative Engine Optimisation (GEO) and WebMCP.

Neither is replacing SEO, and SEO still provides the foundation needed to be visible. But both are becoming increasingly difficult to ignore.

Search is changing – whether we like it or not

Google’s recent AI announcements have made one thing very clear: search is evolving faster than most businesses realise. Its I/O 2026 Search update described a new AI-powered search box as the biggest upgrade to Search in more than 25 years, and TIME’s coverage framed the shift as a major change in how people use the internet.

For more than two decades, the web has been built around a familiar journey. Someone searches for something, Google provides a list of results and the user clicks through to a website. Job done.

Increasingly, that’s no longer what happens.

AI-powered search experiences are becoming the starting point for research, recommendations and decision-making. Instead of returning a list of links, they’re returning answers. Detailed answers. Contextual answers. Answers that often summarise information from multiple sources without requiring the user to visit every website individually.

The journey is shifting from ranked links towards synthesised answers and recommendations. This creates a real dilemma for businesses because historically, rankings were the ultimate metric for success. Now, though, a completely different question is taking center stage:

Are we being included in the answer?

Because if your competitors are consistently being referenced by AI systems and you’re not, rankings alone won’t tell the full story.

Welcome to the world of GEO

One phrase you’ll hear a lot more over the next few years is Generative Engine Optimisation, usually shortened to GEO.

It helps to look at GEO as the natural next step built right on top of traditional SEO, where the focus shifts from just helping a search engine index your website to actually helping AI understand your underlying expertise.

The distinction matters.

Traditional SEO asks:

How do we rank higher?

GEO asks:

How do we become a trusted source?

When someone asks ChatGPT, Gemini, Claude, Perplexity or Google’s AI Mode a question, those platforms aren’t simply choosing which website deserves the number one position. They’re deciding which information deserves to be included in the answer itself.

That’s a very different challenge.

And we’re already seeing it affect businesses across a range of sectors. We’ve spoken to organisations that rank exceptionally well in Google but rarely appear in AI-generated answers. We’ve also seen smaller businesses gain visibility because they’ve produced genuinely useful content around a niche subject.

The rules aren’t identical. The underlying principles are similar, but the way AI systems evaluate information introduces a new set of opportunities.

A question we're hearing more often

Over the past six months, we’ve had several conversations with clients that all boil down to the same thing:

We’re ranking well in Google, so why are our competitors showing up in ChatGPT answers and we’re not?

While that wasn’t a question anyone was asking a few years ago, it’s rapidly becoming part of the everyday conversation.

The answer isn’t usually one single thing. It’s often a combination of content quality, topical authority, structured data, brand signals and how clearly a business communicates what it actually does.

That’s one of the reasons GEO is attracting so much attention. Businesses are starting to realise that visibility in traditional search results and visibility in AI-generated answers aren’t always the same thing.

We’re ranking well in Google, so why are our competitors showing up in ChatGPT answers and we’re not?

What we're seeing work

The websites that appear most consistently across AI-generated responses tend to have a few things in common. They’re clear, authoritative and they answer questions directly. They demonstrate expertise rather than simply claiming it while making it easy for machines to understand how information relates together.

In practical terms, that usually means:

Strong topical authority
Clear service and product information
Well-implemented structured data
Consistent entity signals
Logical information architecture
Original insights and expertise

None of this is particularly glamorous, but to be fair, the things that actually work in digital marketing almost never are.

Strong visibility starts with clear planning and useful content.

AI systems reward information that is easy to interpret.

So where does WebMCP fit into all this?

If GEO is about helping AI systems understand your business, WebMCP is about helping them interact with it.

This is where things start to get interesting.

At the moment, AI systems interact with most websites in much the same way a new starter on their first day would. They read pages, click buttons, fill in forms and try to work out what everything does. Occasionally they get confused, but then they try again.

It’s impressive when it works but the problem is that websites were designed for people, not AI agents.

WebMCP aims to make that relationship a little less awkward.

What is WebMCP?

At its core, WebMCP is an emerging standard designed to provide a way for websites to expose actions and functionality directly to AI systems. Chrome’s WebMCP documentation describes it as a proposed web standard for exposing structured tools to AI agents.

Rather than forcing an AI assistant to interpret a user interface visually, a website can provide structured information about what actions are available and how they should be used.

Think about a booking form. A human instantly understands what it’s for. An AI assistant has to analyse the page, identify the form fields, understand what information is required and attempt to complete the process correctly. WebMCP aims to remove much of that guesswork.

Instead of asking an AI system to work everything out for itself, the website can effectively say:

This action books a consultation
These details are required
This is the expected outcome
This is the information returned once the action is completed

It’s a much cleaner way for AI systems and websites to communicate.

It’s worth stressing that WebMCP is still an emerging standard. Browser support and tooling are evolving quickly, and most businesses won’t be implementing it tomorrow. Chrome has already opened an early preview programme, but what makes WebMCP interesting isn’t widespread adoption today. It’s the fact that it offers a glimpse of how websites and AI systems may interact in the future.

WebMCP gives websites a structured way to tell AI systems what actions are available and how to use them.

Why we're paying attention

New technologies appear all the time. Some stick around. Some quietly disappear after generating a few months of excitement and a handful of conference presentations.

Will WebMCP become the dominant standard for AI-agent interactions? Maybe. Maybe not. The web has a long history of producing three competing standards before everyone eventually agrees on one.

What matters isn’t necessarily the acronym itself, it’s the broader shift taking place around it.

Every major technology company is investing heavily in AI assistants, AI agents and AI-powered search experiences. Those systems need reliable ways to interact with websites and services. The current approach, essentially teaching AI to mimic human behaviour, is clever, but it’s unlikely to be the long-term answer.

That’s why developments like WebMCP deserve attention. They’re attempting to solve a real problem.

What this could look like in practice

Imagine a potential customer asks an AI assistant:

Find me a web design agency in Cumbria and book an introductory consultation next week.

Today, the assistant might provide a list of agencies and ask the user to take it from there. In the future, the goal is that AI assistants could complete processes like this more reliably through structured interactions rather than simulated clicks and form-filling.

The same principle could apply to:

Booking appointments online
Requesting quotations
Checking product availability
Scheduling demonstrations
Creating support tickets
Managing customer accounts

For businesses, that’s potentially a significant shift in how customers engage online.

A simplified example

The WebMCP specification is still evolving, and implementations will vary depending on the platform and use case. However, the underlying idea is relatively straightforward.

Instead of expecting an AI assistant to figure out a booking form on its own, a website can explicitly describe the action that’s available and the information required to complete it.

Conceptually, that might look something like this:

document.modelContext.registerTool({
  name: "book-consultation",
  title: "Book consultation",
  description: "Book an introductory consultation",
  inputSchema: {
    type: "object",
    properties: {
      name: { type: "string" },
      email: { type: "string" },
      preferredDate: { type: "string" }
    },
    required: ["name", "email"]
  },
  async execute({ name, email, preferredDate }) {
    return createBooking({
      name,
      email,
      preferredDate
    });
  }
});

From a user’s perspective, very little changes. Behind the scenes, however, it gives AI systems a much clearer understanding of what a website can actually do and how to interact with it.

WebMCP points towards a future where assistants can understand and trigger website actions more reliably.

What should businesses be doing now?

The good news is that most businesses don’t need to rush out and implement WebMCP tomorrow morning.

The less good news is that many organisations still haven’t addressed the foundations that make AI visibility possible in the first place.

We’ve already seen businesses invest heavily in content, SEO and digital marketing, only to discover they’re barely visible in AI-generated answers. In most cases, the issue isn’t a lack of content. It’s a lack of structure, a lack of clarity and a lack of understanding around how AI systems interpret information.

For most organisations, the immediate question is simple: can AI systems clearly understand what the business does, who it helps and why it should be trusted?

That means focusing on the fundamentals.

Your website still needs strong technical SEO. It needs to be fast, accessible and easy to crawl. It needs clear information architecture, because people and machines both benefit from websites that make sense. It needs structured data, because schema markup remains one of the most effective ways to help systems understand your content.

It also needs topical authority. Publishing twenty mediocre articles is not a strategy. Becoming genuinely useful within your area of expertise is.

practical takeaway

Most businesses do not need WebMCP today.

They need clearer content, stronger technical foundations and better structured data.

Consistent business information matters too. The clearer your services, products and expertise are, the easier they become to understand, reference and recommend.

In many cases, improving these foundations will have a far greater impact than chasing the latest technology trend. Technical foundations, structured data and clear architecture still matter. Better source material gives AI systems more reliable context to work with.

Technical foundations, structured data and clear architecture still matter.

Better source material gives AI systems more reliable context to work with.

The bigger picture

Google’s AI announcements have prompted the usual flood of “SEO is dead” headlines.

For the record, we’ve seen those headlines every year since roughly 2004.

SEO survived social media. It survived mobile. It survived voice search. It even survived the metaverse. It’s probably going to survive AI too.

What’s changing isn’t the need for optimisation. What’s changing is the way people discover information.

For years, we’ve built websites for two audiences: people and search engines. It now looks increasingly likely we’ll need to add a third audience to that list: AI systems and agents.

That’s why we’re paying attention to GEO. That’s why we’re watching developments around WebMCP closely. And that’s why we’re helping clients think beyond rankings and start considering how their businesses will be discovered, understood and interacted with in an increasingly AI-driven web.

The businesses that do well over the next few years probably won’t be the ones chasing every new AI buzzword doing the rounds on LinkedIn. They’ll be the ones doing what successful businesses have always done: communicating clearly, building trust and making it easy for customers to find what they need.

The difference is that those customers won’t always be people. Increasingly, they’ll be AI systems helping people make decisions.

That’s ultimately why we’re interested in technologies like WebMCP. Not because it’s the latest shiny object, but because it offers a glimpse of where the web appears to be heading next.

And if there’s one thing we’ve learned from twenty years of search, it’s that paying attention to those shifts early is usually a lot easier than catching up later.

Want to know how visible your business is in AI search?

Understand how your website appears across AI-powered search platforms and identify opportunities to improve visibility.

get in touch

written by

Pelling

We help businesses strengthen their online presence through smart strategy, thoughtful content and high-performing websites.

Why being found is no longer enough: GEO, WebMCP and the future of search