Why International SEO Now Depends on GEO and EEAT
Assume a company operates forty websites, one for each country it serves. Every site is written by local writers, reviewed by local experts, and designed for local audiences. To a human reader, those forty sources appear to be credible and trustworthy.
An artificial intelligence system may perceive that a single brand says the same thing forty times. Motoko Hunt, an international search strategist, described this scenario in a recent Search Engine Journal article. Her point is simple yet important: expertise does not become visible to machines simply because it exists.
If brands want AI-powered search to recognize their market authority, they must provide machine-readable evidence (Hunt, 2026). Anyone considering a career in marketing, design, business, or technology should be aware of this shift. It changes how brands build trust online, what skills employers look for, and what it means to be "visible" in an age where answers increasingly come from AI rather than a list of links.
From keywords to credibility. From keywords to credibility.
For many years, search engine optimization was primarily focused on keywords, links, and technical structure. These factors remain important, but the questions that search engines pose have become more sophisticated.
Today, they are attempting to determine not only whether a page is relevant but also whether it is credible. Google defines this as a set of characteristics known as experience, expertise, authoritativeness, and trustworthiness, commonly abbreviated as EEAT in the industry.
Google added the first "experience" to the previous framework in 2022, recognizing that firsthand experience with a topic, in addition to formal expertise, is a valuable signal of quality. Trust is essential: a page can show experience and expertise, but if it is untrustworthy, it will not be rated highly. These concepts are described in Google's guidelines for human raters who assess the quality of search results (Google, 2025).
Google has also stated unequivocally that content should be written for people rather than search engines. Its guidelines for useful content ask whether a page demonstrates genuine knowledge, provides original information, and gives the reader the impression that they have learned enough to achieve their goal (Sullivan, 2022). None of this is entirely new.
Research into how people evaluate websites has long shown that credibility is based on more than just information. In a large survey of over 2,600 people, the visual design and presentation of a website was the most frequently mentioned factor in deciding whether to trust it (Fogg et al., 2003). Later studies defined online credibility as a combination of source, message, and medium, with users relying on quick signals because they rarely have time to independently verify everything (Metzger, 2007).
What's new is that machines are also making these decisions.
How machines choose who to trust. How machines choose who to trust.
To understand Hunt's point, consider how modern search engines "see" the world. In 2012, Google launched its Knowledge Graph with the memorable phrase "things, not strings" (Singhal, 2012). Instead of matching words, search would better understand entities such as people, places, organizations, and concepts, as well as their relationships. This concept arose from an earlier vision of a "semantic web," in which information is organized so that computers can understand meaning rather than just text (Berners Lee et al., 2001).
Knowledge graphs are now a major area of study in their own right, describing how machines can organize and use information about connected entities (Hogan et al., 2021).
Structured data is one of the practical tools for achieving this. Shared vocabularies, such as Schema.org, allow websites to clearly label information, such as identifying a person, the organization they work for, and the credentials they possess, so that search engines can interpret it consistently (Guha et al., 2016). Generative AI adds an extra layer.
Many AI assistants respond to questions by retrieving relevant information and then generating an answer based on it, a process known as retrieval augmented generation (Lewis et al., 2020). Researchers studying "generative engine optimization" (GEO) discovered that the way content is written and supported, such as using citations, quotations, and statistics, has a significant impact on how visible it is in AI-generated answers (Aggarwal et al., 2024). Commercial stakes are real. A Pew Research Center analysis of real-world browsing behavior found that when Google displayed an AI summary, users clicked on a traditional search result in only 8% of visits, compared to 15% when no summary was displayed (Pew Research Center, 2025). If fewer people click through, being recognized and cited in the answer itself becomes far more valuable.
Why does authority in international SEO not travel alone? Why does authority in international SEO not travel alone?
This brings us to the central issue that Hunt has identified. Global brands have frequently assumed that authority gained in one market transfers to others. Hunt claims it does not, at least not for free (Hunter, 2026). She makes a useful distinction between two questions.
The first is "Who is this company?" which means being the most reliable source of information about your own brand.
The second question is, "Does this company, or the author, truly understand this subject?" This is the question EEAT is designed to answer. AI systems seem to treat them separately.
A brand can be completely open about who it is while still not being recognized as an authority in its field. Her professional credentials serve as a vivid illustration of the point.
Architects in Germany, France, and Japan demonstrate genuine expertise. However, a model trained primarily on English terms such as "licensed architect" may fail to recognize that these local credentials have equal weight (Hunt, 2026). This is a valid concern.
Language technology research has revealed that the vast majority of the world's languages are underrepresented in the data used to build modern language systems, with only a few languages, most notably English, dominating (Joshi et al., 2020). Scholars have also warned that large training datasets tend to favor dominant viewpoints while underrepresenting others (Bender et al., 2021). If local expertise is only described in local terms, it may not register. Hunt discusses two related effects. In "market aggregation bias," models gravitate toward markets with high representation in their training data and combine similar regional content into a single overall picture of a brand. Canonical amplification occurs when a brand's dominant content is amplified across markets, drowning out local variations (Hunt, 2026).
What is required to make expertise machine readable? What is required to make expertise machine readable?
Hunt's answer is a practice she refers to as "authority translation." She contends that localization is more than just translating languages. It's about translating evidence (Hunt, 2026). In practice, this requires several steps.
Expose the local context. Make the institutions behind a credential explicit: professional associations, regulators, certifications, universities, and standards bodies. Do not assume a machine knows what a local qualification means.
Connect the entities. Link credentials to the organizations that issue them, link experts to the associations, publications, and universities they are connected with, and link products to the regulations and standards that apply in each market. This is exactly the kind of relationship that knowledge graphs and structured data are designed to capture (Guha et al., 2016; Hogan et al., 2021).
Strengthen author pages. An author page should do more than list qualifications. It should connect each credential to the institution that makes it meaningful.
Add something new in every market. Forty translated product pages do not create forty sources of expertise. Local regulations, customer concerns, case studies, and commentary from local experts do. Hunt encourages brands to audit each regional site and ask whether it genuinely contributes new knowledge.
Ask the diagnostic questions. Are local credentials understandable to AI, or only to people already familiar with that market? Are the relationships that create trust visible, or merely assumed? This is where the future of international SEO is heading: away from simply copying content across borders and towards building clear, connected, original evidence of expertise in every market a brand serves.
Why this matters for the next generation of professionals
All of this may sound like a concern only for large global companies. In reality, it describes the environment that every graduate entering marketing, design, business, or technology will work in. Indonesia is a powerful example. With more than 200 million internet users, it is one of the largest digital markets in the world (Kemp, 2025).
Brands competing for attention in Indonesia, and Indonesian brands reaching out to the region and beyond, need professionals who understand both local audiences and the global systems that decide what those audiences see. Employers are already looking for these capabilities.
The World Economic Forum’s latest research identifies AI and big data as the fastest-growing skills worldwide, alongside creative thinking, technological literacy, and curiosity (World Economic Forum, 2025). McKinsey estimates that generative AI could add trillions of dollars in value to the global economy each year, with marketing and sales among the functions where the impact is greatest (Chui et al., 2023).
The skills that Hunt’s article calls for cut across traditional disciplines:
Strategic thinking: understanding how trust, reputation, and authority are built across different markets and cultures.
Content and storytelling: creating original material that adds genuine value rather than repeating what already exists.
Visual communication: presenting information clearly and credibly, since design strongly shapes how trustworthy a source appears (Fogg et al., 2003).
Technical and AI literacy: understanding how search systems, structured data, and AI models interpret information.
Cultural intelligence: knowing how credentials, institutions, and expectations differ from one market to another.
These are exactly the kinds of skills that are best developed through practice. Research on experiential learning describes effective learning as a cycle of doing, reflecting, and trying again (Kolb & Kolb, 2005), and a large analysis of 225 studies found that students in active learning environments perform significantly better than those in traditional lectures (Freeman et al., 2014).
Building these skills at Raffles Jakarta
At Raffles Jakarta, several programs prepare students for precisely this kind of work, combining creativity, business understanding, and technology.
Business Administration builds the foundations of marketing, strategy, and management that brands need to grow across markets. For working professionals ready to lead digital and international strategy, the MBA develops the leadership perspective to make these decisions at a senior level.
Fashion Marketing & Management explores how brands build identity, trust, and reach in one of the most global and image-driven industries in the world.
Digital Media Design and Visual Communication Design and develop the visual and digital storytelling skills that help brands communicate clearly and credibly, whether to people or to platforms.
For students who want to understand the technology behind modern search and AI, the Artificial Intelligence program builds the technical knowledge that sits behind the systems now shaping how information is found and trusted.
Learning in an international setting is one of the most natural ways to understand how ideas, credentials, and brands cross borders, and when they don't.
See it for yourself at the November Open House.
Learning about the future of digital marketing is one thing. Another option is to see how it is taught. According to research on how students choose where to study, visiting a campus in person is one of the most influential factors in their decision.
The proportion of new students who considered a campus visit to be very important in their decision increased from 37.6 percent in 2003 to 47.3 percent in 2017 (Stolzenberg et al., 2019), and admissions researchers describe the visit as one of the most powerful sources of information a student can have (Secore, 2018).
Visitors can tour the campus, participate in hands-on workshops, view live portfolio showcases, and meet the International Academic Team. The Open House also includes an exclusive scholarship opportunity, as well as a free evaluation from international experts, so you can determine where you stand and which program best suits your goals.
Intake January 2027 is now open.
Search is changing, as are the careers that revolve around it. Professionals who can combine creativity, strategy, and technology, as well as understand how to earn trust across cultures and markets, will thrive.
The future of search belongs to people who can make expertise visible. Your own expertise can start taking shape this January.
Frequently Asked Questions
What does EEAT mean in search?
It stands for experience, expertise, authoritativeness, and trustworthiness, the qualities Google uses to describe high-quality, reliable content.
Why does expertise need to be machine readable?
AI systems need to recognize expertise before they can evaluate it. Clear, connected, and structured evidence helps them understand who an expert is and why they should be trusted.
Does authority automatically carry over from one country to another?
Not necessarily. Local credentials, institutions, and evidence may not be recognized unless they are explained and connected in ways AI systems can interpret.
What skills are needed for a career in digital marketing today?
Strategic thinking, original content creation, visual communication, AI and technical literacy, and an understanding of different cultures and markets.
Which Raffles Jakarta programs relate to digital marketing and AI?
Business Administration, the MBA, Fashion Marketing and Management, Digital Media Design, Visual Communication Design, and Artificial Intelligence all build relevant skills.
When is the next Raffles Jakarta Open House?
Saturday 28 November 2026, from 10.00 AM to 2.00 PM, at Jl. M.H. Thamrin No. 20, Central Jakarta. Applications for the January 2027 intake are now open.
Head of Marketing
References
Aggarwal, P., Murahari, V., Rajpurohit, T., Kalyan, A., Narasimhan, K., & Deshpande, A. (2024). GEO: Generative engine optimization. In Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. https://doi.org/10.1145/3637528.3671900
Bender, E. M., Gebru, T., McMillan-Major, A., & Shmitchell, S. (2021). On the dangers of stochastic parrots: Can language models be too big? In Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency. https://doi.org/10.1145/3442188.3445922
Berners-Lee, T., Hendler, J., & Lassila, O. (2001). The semantic web. Scientific American, 284(5). https://doi.org/10.1038/scientificamerican0501-34
Chui, M., Hazan, E., Roberts, R., Singla, A., Smaje, K., Sukharevsky, A., Yee, L., & Zemmel, R. (2023). The economic potential of generative AI: The next productivity frontier. McKinsey & Company. https://www.mckinsey.com/capabilities/mckinsey-digital/our-insights/the-economic-potential-of-generative-ai-the-next-productivity-frontier
Fogg, B. J., Soohoo, C., Danielson, D. R., Marable, L., Stanford, J., & Tauber, E. R. (2003). How do users evaluate the credibility of Web sites? A study with over 2,500 participants. In Proceedings of the 2003 Conference on Designing for User Experiences. https://doi.org/10.1145/997078.997097
Freeman, S., Eddy, S. L., McDonough, M., Smith, M. K., Okoroafor, N., Jordt, H., & Wenderoth, M. P. (2014). Active learning increases student performance in science, engineering, and mathematics. Proceedings of the National Academy of Sciences, 111(23). https://doi.org/10.1073/pnas.1319030111
Google. (2025). Search quality rater guidelines. https://static.googleusercontent.com/media/guidelines.raterhub.com/en//searchqualityevaluatorguidelines.pdf
Google Search Central. (2022, December 15). Our latest update to the quality rater guidelines: E-A-T gets an extra E for Experience. https://developers.google.com/search/blog/2022/12/google-raters-guidelines-e-e-a-t
Guha, R. V., Brickley, D., & Macbeth, S. (2016). Schema.org: Evolution of structured data on the web. Communications of the ACM, 59(2). https://doi.org/10.1145/2844544
Hogan, A., Blomqvist, E., Cochez, M., d’Amato, C., de Melo, G., Gutierrez, C., Kirrane, S., Labra Gayo, J. E., Navigli, R., Neumaier, S., Ngonga Ngomo, A.-C., Polleres, A., Rashid, S. M., Rula, A., Schmelzeisen, L., Sequeda, J., Staab, S., & Zimmermann, A. (2021). Knowledge graphs. ACM Computing Surveys, 54(4). https://doi.org/10.1145/3447772
Hunt, M. (2026, August 25). Why international SEO needs machine-recognizable E-E-A-T. Search Engine Journal. https://www.searchenginejournal.com/why-international-seo-needs-machine-recognizable-e-e-a-t/583439/
Joshi, P., Santy, S., Budhiraja, A., Bali, K., & Choudhury, M. (2020). The state and fate of linguistic diversity and inclusion in the NLP world. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. https://doi.org/10.18653/v1/2020.acl-main.560
Kemp, S. (2025). Digital 2025: Indonesia. DataReportal. https://datareportal.com/reports/digital-2025-indonesia
Kolb, A. Y., & Kolb, D. A. (2005). Learning styles and learning spaces: Enhancing experiential learning in higher education. Academy of Management Learning & Education, 4(2). https://doi.org/10.5465/amle.2005.17268566
Lewis, P., Perez, E., Piktus, A., Petroni, F., Karpukhin, V., Goyal, N., Küttler, H., Lewis, M., Yih, W., Rocktäschel, T., Riedel, S., & Kiela, D. (2020). Retrieval-augmented generation for knowledge-intensive NLP tasks. Advances in Neural Information Processing Systems, 33. https://arxiv.org/abs/2005.11401
Metzger, M. J. (2007). Making sense of credibility on the Web: Models for evaluating online information and recommendations for future research. Journal of the American Society for Information Science and Technology, 58(13). https://doi.org/10.1002/asi.20672
Pew Research Center. (2025, July 22). Google users are less likely to click on links when an AI summary appears in the results. https://www.pewresearch.org/short-reads/2025/07/22/google-users-are-less-likely-to-click-on-links-when-an-ai-summary-appears-in-the-results/
Secore, S. (2018). The significance of campus visitations to college choice and strategic enrollment management. Strategic Enrollment Management Quarterly, 5(4). https://www.aacrao.org/news/the-significance-of-campus-visitations-to-college-choice-and-strategic-enrollment-management/
Singhal, A. (2012, May 16). Introducing the Knowledge Graph: Things, not strings. Google. https://blog.google/products/search/introducing-knowledge-graph-things-not/
Stolzenberg, E. B., Eagan, M. K., Romo, E., Tamargo, E. J., Aragon, M. C., Luedke, M., & Kang, N. (2019). The American freshman: National norms fall 2017. Higher Education Research Institute, UCLA. https://www.heri.ucla.edu/monographs/TheAmericanFreshman2017.pdf
Sullivan, D. (2022, August 18). What creators should know about Google’s August 2022 helpful content update. Google Search Central Blog. https://developers.google.com/search/blog/2022/08/helpful-content-update
World Economic Forum. (2025). The future of jobs report 2025. https://www.weforum.org/publications/the-future-of-jobs-report-2025/in-full/
