Your next website visitor might not be human

We’ve spent years designing websites for humans. Now we need to make sure machines can understand them too.

Imagine a potential customer looking for a new accountant, a charity researching creative agencies, or a business comparing IT providers. Traditionally, they’d head to Google, browse a handful of websites and make their own comparisons. But now they can ask an AI assistant to do much of that research for them, searching the web, examining several businesses, weighing up their services and presenting a shortlist of recommendations. It’s a fundamental shift in how businesses are discovered online, and it raises an important question…

How well does your website communicate with someone who isn’t human?

Your website still plays a crucial role in that buying process, even if the potential customer never actually visits it. The information you publish could influence whether your business is considered, how your expertise is interpreted, and whether you make it onto that all-important shortlist. In other words, your website could be helping to win or lose opportunities without generating a single visit in your analytics.

Your website has a new audience
Good website design has always been about understanding your audience. Who are they? What matters to them? What questions are they asking? What do they need to make a decision?

Those questions haven’t changed. But we now need to consider how AI systems find, interpret and communicate the answers. AI-powered search tools, including ChatGPT, Google AI Mode and Microsoft Copilot, can use information from websites to help people research businesses, products and services. Rather than simply displaying a list of links, these systems can interpret information from multiple sources and present responses tailored to the user’s questions. That means your website needs to do more than look good and persuade visitors. It needs to make your business easy to understand, accurately represent and confidently recommend.

So how do you give your AI audience a great experience?

1. Make it obvious who you are and what you do
This sounds basic, but you’d be surprised how many websites struggle to explain their own proposition. Vague headlines and generic statements about delivering excellence might look impressive, but they don’t necessarily communicate anything meaningful. AI systems need to establish what your business does, who it serves and what makes it relevant. So be specific, explain your services clearly, identify the sectors you work in, and describe the problems you solve for your clients.

Your website should answer three fundamental questions:

• Who are you?
• What do you do, and for whom?
• Why should someone choose you?

If your website can’t communicate those things clearly, how can you expect an AI assistant to do it on your behalf?

2. Structure your content so it’s easy to understand
Content buried within complex layouts, hidden behind interactive elements or presented entirely through imagery can create barriers to discovery. Clear headings, logical page structures, descriptive links and accessible HTML help search systems understand how information fits together. Structured data, such as Schema.org markup, can also provide additional context about your organisation, services, articles and other relevant information. This doesn’t mean stripping away creativity or designing websites that look like databases, it means ensuring the underlying structure is as carefully considered as the visual experience.

Google’s own guidance on AI search reinforces the importance of established SEO practices, including accessible content, clear structure and useful information. There’s no magic piece of code that guarantees your website will appear in an AI-generated recommendation, but good design, good content, and good information architecture can help you compete.

3. Answer the questions your customers are actually asking
Traditional SEO has often focused on individual keywords and search phrases, but conversations with AI tend to be more detailed and contextual.

Someone might ask:

“Which Essex-based accountancy firms specialise in helping growing businesses?”
Or:
“Can you recommend a branding agency with experience working with healthcare charities?”

These are questions about suitability, experience and credibility, not just keywords. Businesses should think carefully about the questions their audiences might ask an AI assistant and ensure their website contains genuinely useful answers. Dedicated service pages, FAQs, industry-specific content and detailed case studies can all help. The objective isn’t to create hundreds of pages stuffed with variations of the same question, it’s to develop useful, authoritative content that reflects a genuine understanding of your customers.

4. Give AI something worth recommending
Anyone can claim to be experienced, innovative or customer-focused. The challenge is proving it.

Consider two law firms. One describes itself as a leading firm of solicitors, providing exceptional legal services with a client-focused approach. The other clearly identifies its areas of specialism, explains the types of clients it supports, introduces its legal experts and provides examples of relevant cases and successful outcomes.

The second firm hasn’t necessarily done anything particularly clever from a technical perspective. They’ve simply made its expertise easier to understand and verify. Case studies, measurable outcomes, testimonials, specialist knowledge and original insights all help communicate why your business deserves consideration. Rather than simply saying you’re good at something, show what you’ve achieved. And wherever possible, be specific.

5. Make sure AI can actually access your content
Even the best content won’t help if the systems looking for it can’t access it. Website owners should review their technical setup to ensure relevant search crawlers can discover and interpret their pages. This includes checking indexing settings, robots.txt rules, XML sitemaps, internal linking and the accessibility of important content. Different AI services use different crawlers, and some distinguish between crawling for search results and collecting information for AI model training.

For example, OpenAI provides separate controls for its search crawler and its model-training crawler. That distinction gives businesses greater control over how their content is accessed and used. It’s also worth remembering that allowing a crawler access doesn’t guarantee your business will appear in AI-generated recommendations. But unnecessarily blocking relevant systems could limit your opportunities to be discovered. And if AI agents increasingly interact directly with websites, straightforward navigation, accessible interfaces and clearly labelled actions will become even more valuable.

6. Think beyond your own website
Your website remains an important source of information about your business, but it doesn’t exist in isolation. AI systems may draw on business directories, reviews, articles, industry publications and other publicly available sources, which makes consistency across your wider digital presence important. Are your business descriptions accurate? Are your services clearly represented? Do third-party websites reinforce the same positioning and expertise? A strong brand needs a consistent story, wherever it’s being discovered. And genuine mentions, recommendations and independent evidence can help support the claims you’re making about your business.

7. Start measuring a different kind of visibility
For years, website performance has been measured through rankings, visits, engagement and conversions, and those metrics still matter. But AI is asking another question:

Is your business being mentioned or recommended before someone reaches your website?

Businesses should begin monitoring how they appear in relevant AI-generated answers, which competitors are being recommended and which sources are being referenced. Search platforms are beginning to provide more insight into this activity, including reporting on website citations within AI-generated answers. However, AI responses can vary depending on the question, context and system being used. A single test isn’t a reliable measure of visibility. The important thing is to establish a baseline, identify gaps and monitor meaningful changes over time. Understanding the changing discovery process will become an increasingly important part of measuring digital performance.

And don’t forget about the humans!
There’s a danger that businesses become so focused on optimising for AI that they forget who they’re ultimately trying to reach. People.

The purpose of your website hasn’t changed. It still needs to communicate your value, build confidence and encourage meaningful action. AI doesn’t replace those objectives. It introduces another route through which people can discover your business. Many of the things that make a website effective for AI are the same things that make it effective for people.

Clarity. Structure. Relevance. Accessibility. Credibility.

These aren’t new principles. They’re fundamentals of good communication and good website design. The difference is that the consequences of getting them wrong could become much more significant. Your next website visitor might not be human, but the decision they’re helping to make almost certainly will be.

At sixredsquares, we believe effective websites start with understanding your business and the people you want to reach. As AI changes how customers discover and evaluate businesses, that understanding matters more than ever. The challenge isn’t simply to make your website visible to AI, it’s to make sure your business is worth recommending.