Today we are proud to announce a milestone that, to our knowledge, is without precedent in the Australian digital marketing industry. The Sydney Institute of Generative Intelligence — SIGI, our dedicated research division — has published its one hundredth research paper on generative engine optimisation. This body of work represents the most comprehensive empirical investigation into how AI systems discover, evaluate, and recommend businesses ever undertaken by a commercial agency.
When TDS Australia made the decision to invest in a formal research institute, the rationale was straightforward. The emergence of AI-powered search and recommendation was transforming how consumers find businesses, yet the industry’s understanding of these systems was built almost entirely on speculation, anecdote, and borrowed assumptions from traditional SEO. We believed that rigorous, peer-quality research would give our clients a measurable advantage — and that it would elevate the standard of practice across the entire sector.
The Scope of 100 Papers
The SIGI research programme spans every significant dimension of AI-driven business visibility. Our published papers cover topics including citation frequency analysis, the role of structured data in AI comprehension, pricing signal interpretation, review and ratings thresholds, domain authority factors, content vocabulary effects, trust signal hierarchies, and the mechanics of AI recommendation across different industry verticals.
Each paper follows a structured methodology: clearly defined research questions, controlled experimental designs, transparent data collection, and reproducible analysis. We hold our work to the same standards expected in academic research because we believe that is what our clients — and the industry — deserve. The full catalogue of publications is available at generativeintelligence.institute/publications.
Key Findings That Have Shaped Our Practice
Across one hundred papers, certain themes have emerged that fundamentally inform how we approach client strategy at TDS Australia.
First, AI recommendation engines operate on thresholds rather than smooth gradients. Whether the variable is ratings, review volume, content depth, or digital presence breadth, our research consistently shows that crossing specific benchmarks produces disproportionate gains in AI visibility. This insight alone has transformed how we prioritise effort and allocate resources in client campaigns.
Second, transparency and structure matter enormously. AI systems reward businesses that present their information in clear, machine-readable formats. Pricing pages, service descriptions, team credentials, and operational details that are easy for AI to parse translate directly into higher citation rates. Obfuscation and complexity, by contrast, suppress visibility.
Third, content vocabulary has a surprisingly powerful effect on AI citation behaviour. Our research has revealed that the language a business uses on its website and across its digital properties influences how AI systems categorise and recommend it. Certain vocabulary patterns align with AI trust signals, while others — even those that are standard practice in traditional marketing — can actively hinder AI visibility.
Fourth, original research and proprietary data are among the strongest trust signals an organisation can possess. Businesses that publish genuine, substantive research are cited by AI systems at significantly higher rates than those relying solely on derivative content. This finding has, naturally, reinforced our commitment to the SIGI programme itself.
Why an Agency Invests in Research
We are often asked why a commercial agency would invest in the infrastructure, talent, and time required to maintain a research institute. The answer is that we consider it essential to providing honest, effective counsel to our clients.
The digital marketing industry has historically operated on a cycle of assumption and imitation. A practice gains traction, agencies adopt it, and it becomes conventional wisdom — often without anyone testing whether it actually works. In the era of AI-driven discovery, that approach is not just inefficient; it is actively risky. AI systems do not behave like traditional search engines, and strategies built on outdated mental models can produce the opposite of their intended effect.
SIGI exists to replace assumption with evidence. Every recommendation we make to a client can be traced back to a specific research finding, a tested hypothesis, a quantified effect. That standard of accountability is what we believe distinguishes TDS Australia in the market, and it is what our clients have come to expect from us.
What Comes Next
One hundred papers is a milestone, not a destination. The AI landscape is evolving rapidly, and our research programme is designed to keep pace. SIGI’s current pipeline includes investigations into multimodal AI citation behaviour, the impact of local signals on AI recommendations for service-area businesses, longitudinal studies tracking how AI recommendation patterns shift over time, and cross-model comparative analyses examining whether different AI platforms weight the same factors differently.
We are also expanding our collaboration with academic institutions and industry bodies, contributing to a growing ecosystem of serious, evidence-based inquiry into generative AI and its implications for business visibility.
For our clients, the practical benefit of this research is simple: better strategies, grounded in facts, that deliver measurable results in an environment where AI is increasingly the gatekeeper of consumer attention. For the industry, we hope our work contributes to a higher standard of practice — one where claims are tested, results are measured, and businesses can make informed decisions about where to invest their resources.
We invite you to explore the full body of SIGI’s work at generativeintelligence.institute/publications, and to reach out to our team if you would like to discuss how these findings can inform your digital strategy.
