The way people discover brands is changing. Instead of scrolling through pages of Google results, more consumers are turning to AI tools like ChatGPT, Claude, and Perplexity to get recommendations, compare products, and make purchasing decisions. This shift represents what we’re calling “the new battleground for brands.”
On 26 November, Growth Studio hosted a webinar with Liam Reynolds, founder of Zebora, who shared practical strategies for improving your brand’s visibility and sentiment in large language models (LLMs). The session drew significant interest from our community, highlighting just how crucial this topic has become for founders and marketers alike.
Why GEO Matters Now
Generative Engine Optimisation (GEO) is about how your brand appears in AI models. It’s not just about visibility anymore – it’s about sentiment and context. When someone asks ChatGPT for product recommendations, does your brand appear? Where do you rank compared to competitors? And crucially, what does the AI actually say about you?
As Liam explained, people’s search behaviour is fundamentally shifting. Google’s first results are typically paid ads, followed by brands that have mastered SEO. The best results often sit buried further down the page, where most people never bother to look. LLMs bypass this frustration by understanding context and delivering relevant answers directly.
10 Practical Steps to Improve Your AI Visibility
Rather than dwelling on theory, Liam focused on actionable tactics that any founder can implement, prioritising the quickest wins:
1. Check for crawlability (one for the IT/web team). Ensure your robots.txt, Cloudflare/Akamai settings, WAF rules, and bot-blocking aren’t preventing LLMs from accessing your site. Whitelist GPTBot, Anthropic-AI, Google-Extended, and Common Crawl. Test by asking ChatGPT what it can see on your website and checking the URL citations.
2. Define your canonical message. Establish a single, precise sentence that explains what you do, who you serve, and the outcome you deliver. This becomes your textual “source of truth” used consistently across all content.
3. Align your website around that message. Update your homepage, About page, product pages, and key metadata with simple, literal language. Avoid vague marketing claims—LLMs need clear, factual statements.
4. Seed consistent signals across the public web. Update LinkedIn pages, directories, press releases, partner sites, and documentation with your canonical message. Standardise descriptions, facts, company size, and category language to reduce ambiguity.
5. Add structured product/service descriptions. Create clear descriptions with factual, declarative wording and add schema markup (Product, Organisation, How-To) to help LLMs receive unambiguous signals.
6. Create clear FAQs. Answer top problem-aware, solution-aware, and product-aware questions with short, literal Q&A that has high information density. Drop 5-8 key questions at the bottom of your homepage.
7. Audit your brand inside ChatGPT. Ask it to summarise your brand, describe your products, and explore your category. Capture inaccuracies, missing information, or hallucinations for your gap analysis.
8. Create a simple prompt taxonomy. Map problem-based, solution-based, and use case-based prompts. Test how ChatGPT responds to each and identify missing associations and positioning gaps.
9. Analyse your competitors in ChatGPT. Review how LLMs describe your category and key competitors. Identify what URLs, sources, and signals they rely on to understand what you must strengthen.
10. Engage in high-signal communities like Reddit. Provide useful, authoritative answers (not sales pitches) in relevant subreddits, Quora, and industry forums. But tread carefully—lurk for a week first to understand community norms.
Quality Over Quantity in the AI Era
One particularly relevant question from attendees addressed the circular nature of AI-generated content: if LLMs create content that then gets fed back into LLMs, are we heading toward a self-referential loop?
Liam’s perspective offers reassurance. While some companies are mass-generating content using AI, this approach likely won’t succeed long-term. The models prioritise quality and uniqueness. A well-researched white paper will outperform a hundred generic pages every time.
The key is creating genuinely useful content—the kind of how-tos, comparisons, and FAQs that provide real value to readers. These are the formats that both humans and AI models appreciate. Tools like Zebora’s ChatGPT homepage audit tool can help you monitor how effectively your content is being surfaced by AI models.
The David vs. Goliath Opportunity
What makes this moment particularly exciting for startups is that GEO is still relatively new territory. Large corporations haven’t yet dominated this space the way they have with traditional SEO. Smaller brands with authentic voices and genuine expertise can compete effectively.
As Liam noted in closing, this is fundamentally about the David versus Goliath story that resonates so strongly in the startup world. The playing field is more level than it’s been in years.
Watch the Full Session
Want to dig deeper into these strategies? The complete webinar recording includes demonstrations of Zebora’s tools, detailed prompt examples for testing your brand’s AI visibility, and extensive Q&A covering everything from schema markup implementation to Reddit engagement tactics.
