Building Brand Trust in the Age of AI

Is Your Brand Invisible to AI? Five Things That Actually Matter Now

If you read my last blog, you’re probably already thinking about how you can make your brand more trustworthy. If you missed it, here’s a brief recap: AI is turning search behaviour on its head by presenting answers to prompts directly within the platform — no need to click on links or visit websites. We’ve all been watching this happen over the last couple of years and have been rushing to optimise our sites for this new age.

But things have gone one step further. Transactions are now taking place within AI platforms themselves — this happened in New Zealand for the first time last month. No visit to a beautifully branded, engaging user experience. No clicks at all. Welcome to zero-click commerce.

For an AI platform to serve up reliable, accurate information — and now, to facilitate a transaction on your behalf — it needs to be confident that your brand is trustworthy. So what signals is it actually looking for? Here are the five that matter most.

1. Entity Consistency Across the Web

AI platforms build what are called ‘entity graphs’ — essentially a picture of your brand stitched together from every mention of it online. Your website, your Google Business Profile, your LinkedIn page, industry directories, media mentions — they’re all part of it. When the information across these sources is consistent, the AI gains confidence that it’s dealing with a legitimate, stable business. When details contradict each other, that confidence evaporates.

This is one of those small things that business owners tend to know is a bit messy, but never quite get around to fixing. In the world of zero-click commerce, that messiness has real consequences.

💡 Real-world example: Imagine a Wellington-based accountancy firm that has three slightly different business names across various platforms — ‘McKenzie Accounting’, ‘McKenzie & Associates’, and ‘McKenzie Accounting Ltd’. A client asks an AI assistant to recommend a local accountant. The fragmented entity data makes the firm harder to surface confidently. A competitor with a spotless, consistent presence across every platform wins the recommendation instead.

2. E-E-A-T Signals: Experience, Expertise, Authoritativeness, Trustworthiness

Originally developed by Google to evaluate content quality, E-E-A-T has become a foundational concept for how AI platforms assess whether a source is worth citing. It’s not just about having good content — it’s about demonstrating the credentials behind it. Who wrote this? What qualifies them to say it? Is the information current? Is the website secure? Does the business stand behind what it publishes?

In practical terms, this means author bios with genuine credentials, clearly attributed content, transparent business information, and a regular commitment to keeping things accurate and up to date.

💡 Real-world example: A Christchurch physiotherapy clinic publishes a helpful article on recovering from a rotator cuff injury. Without an author bio, the AI has no way to verify that a qualified physio wrote it — it could have been written by anyone. A competing clinic with the same article, but attributed to a named, registered physiotherapist with a linked profile, is far more likely to be cited as a trustworthy source.

3. Third-Party Citations and Mentions

What others say about you carries considerably more weight than what you say about yourself. This has always been true in marketing, and it’s especially true for AI. Third-party mentions in publications, on industry websites, in forums, and even unlinked references across the web all serve as validation signals. The more reputable the source doing the mentioning, the stronger the signal.

One 2025 study found that referring-domain authority is the strongest predictor of being cited in AI-generated answers. In other words, being talked about by trusted websites matters enormously — even without a direct link back to yours.

💡 Real-world example: An Auckland interior design studio has a wonderful website but very little presence elsewhere online. A rival studio has been featured in Architecture NZ, quoted in a Stuff article on home renovation trends, and mentioned on several popular property blogs. When someone asks an AI assistant for interior design recommendations in Auckland, the second studio is recommended because the AI has found abundant third-party validation. The first studio, despite its beautiful website, is largely invisible.

4. Structured Data and Schema Markup

This one sits firmly in technical territory, but the impact is hard to ignore. Schema markup is essentially a layer of code added to your website that helps AI platforms — and search engines — understand exactly what your content is about. Organisation schema tells the AI who you are. Article schema tells it what your content covers. Review schema helps it understand what customers think of you. Without it, the AI has to guess.

A 2025 analysis of over 2,000 prompts across ChatGPT, Google AI Overviews, and Perplexity found that 81% of the web pages receiving citations included schema markup. That’s a compelling number.

💡 Real-world example: Two Queenstown adventure tourism operators offer similar experiences at similar price points. One has schema markup that clearly identifies their business type, location, services, pricing range, and customer reviews. The other has none. When a tourist asks an AI assistant for recommended guided kayaking tours in Queenstown, the first operator is surfaced clearly and confidently. The second doesn’t get a mention, despite offering an equally good experience.

5. Content Freshness and Clear Structure

AI platforms favour content that is current and easy to parse. Outdated statistics, old pricing, or pages that haven’t been touched in years send a signal that your information may not be reliable. Visible publish and update dates, regularly refreshed content, and timely responses to industry changes all contribute positively.

Equally important is how your content is structured. Clear headings, direct answers near the top of the page, and short readable paragraphs make it far easier for AI to extract and use your information accurately. Content that buries the key points in lengthy preamble tends to be overlooked.

💡 Real-world example: A Bay of Plenty mortgage broker has an excellent blog, but the most recent post is from 2022 — before the significant interest rate changes that followed. A prospective borrower asks an AI assistant about current mortgage advice and local brokers. The AI identifies the 2022 content as potentially outdated and instead recommends a competitor whose website is regularly updated with commentary on current rates and lending conditions. Freshness, in this case, directly influences trust.

The Bigger Picture

None of these five things are particularly new ideas. Consistency, credibility, third-party validation, technical clarity, and fresh content have always been the foundations of good digital marketing. What’s changed is the stakes. AI platforms are applying these signals far more rigorously than traditional search engines ever did, and with zero-click commerce now a reality in New Zealand, the consequences of ignoring the small things have never been more significant.

The businesses that get this right won’t just be easier to find — they’ll be the ones an AI confidently recommends, and eventually, transacts on behalf of. That’s a significant competitive advantage worth paying attention to.