Search Central Live Deep Dive Asia Pacific 2025 brings together SEOs from across the region for three days of insight, networking, and practical advice.
Held at the Carlton Hotel Bangkok Sukhumvit, the event features an impressive speaker lineup alongside structured networking breaks.
Attendees have the chance to meet familiar faces, connect with global SEO leaders, and share ideas on the latest trends shaping our industry.
The conference is split over three days, with each day covering a key part of Google’s processes: crawling, indexing, and serving.
Some of the practical tips that emerged from day one:
- Keep building human‑focused content. Google’s models favor natural, expert writing above all.
- Optimize for multiple modalities. Make sure images have descriptive alt text, videos have transcripts, and voice search is supported by conversational language.
- Monitor crawl budget. Fix 5XX errors promptly and streamline your site’s structure to guide Googlebot efficiently.
- Use Search Console recommendations. Non‑expert site owners can benefit from the guided suggestions feature to improve usability and performance.
- Stay flexible. Long‑held traffic trends may shift as AI features grow. Past success does not equal future success.
A Pivotal Moment For Search
Mike Jittivanich, director of marketing for South East Asia and South Asia Frontier, set the tone in his keynote by declaring that we’ve reached a pivotal moment in search. He identified three forces at work:
- AI innovation that rivals past major shifts such as mobile and social media.
- Evolving user consumption patterns, as people expect faster, more conversational ways to find information.
- Changing habits of younger generations, who interact with search differently from their parents.
This trio of drivers underlines that past success no longer guarantees future success in search.
As Liz Reid, VP of Search at Google, has put it, “Search is never a solved problem.”
Image from author, July 2025New formats, from AI Overviews to multimodal queries, must be woven alongside traditional blue links in a way that keeps pace with user expectations.
Gen Z: The Fastest‑Growing Search Demographic
One of the most eye-opening statistics came from a session on generational trends: Gen Z (aged 18-24) is the fastest-growing group of searchers.
Image from author, July 2025Lens usage alone grew 65% year‑on‑year, with over 100 billion Lens searches so far in 2025. Remarkably, 1 in 5 searches via Lens now carries commercial intent.
Younger users are also more likely to initiate searches in non-traditional ways.
Roughly 10% of their journeys begin with Circle to Search or other AI‑powered experiences, rather than typing into a search box. For SEOs, this means optimizing for image and voice queries is no longer optional.
Why Human‑Centered Content Wins
Across several talks, speakers emphasized that Google’s machine‑learning ranking algorithms learn from content created by humans for humans.
These models understand natural language patterns and reward authentic, informative writing.
In contrast, AI‑generated text occupies its own space in the index, and Google’s ranking systems are not trained on that portion. Gary Illyes explained that:
Our algorithms train on the highest‑quality content in the index, which is clearly human‑created.
For your site, the takeaway is clear: Keep focusing on well‑researched, engaging content.
SEO fundamentals, like clear structure, relevant keywords, and solid internal linking, remain vital.
There is no separate checklist for AI features. If you’re doing traditional SEO well, you’ll naturally be eligible for AI Overviews and AI Mode features.
AI In Crawling And Indexing
Two sessions shed light on how AI is touching the crawling and indexing process:
- AI Crawl Impact: Sites are seeing increased crawl rates as Googlebot adapts to new AI‑powered features. However, a higher crawl rate does not automatically boost ranking.
- Status Codes and Crawl Budget: Only server errors (5XX) consume crawl budget; 1XX and 4XX codes do not affect it, though 4XX can influence scheduling and prioritization.
Cherry Prommawin explained that crawl budget is the product of crawl rate limit (how fast Googlebot can crawl) and crawl demand (how much it wants to crawl).
If your site has broken links or slow responses, it may slow down the overall crawling process.
Google Search Is Evolving In Two Ways
Google Search is evolving along two main focus points: the types of queries users can pose and the range of answers Google can deliver.
The Questions Users Can Ask
Queries are becoming longer and more conversational. Searches of five or more words are growing at 1.5X the rate of shorter queries.
Beyond text, users now routinely turn to voice, images, and Circle to Search: For Gen Z, about 10% of journeys start with these AI-powered entry points.
The Results Google Can Provide
AI Overviews can generate balanced summaries when there’s no single “right” answer, while AI Mode offers end‑to‑end generative experiences for shopping, meal planning, and multi‑modal queries.
Google is bringing DeepMind’s reasoning models into Search to power these richer, more nuanced results, blending text, images, and action‑oriented guidance in a single interface.
Image from author, July 2025LLMs.txt & Robots.txt
Gary Illyes and Amir Taboul discussed Google’s stance on robots.txt and the IETF working group’s proposed LLMs.txt standard.
Much like meta keywords of old, LLMs.txt is not a Google initiative and not seen as beneficial, or something they’re looking to adopt.
Google’s view is that robots.txt remains the primary voluntary standard for controlling crawlers. If you choose to block AI‑specific bots, you can do so in robots.txt, but know that not all AI crawlers will obey it.
AI Features As Extensions Of Search
AI Mode and AI Overviews rely on the exact same crawling, indexing, and serving infrastructure as traditional Search.
Googlebot handles both blue‑link results and AI features, while other crawlers in the same system feed Gemini and large language models (LLMs).
Image from author, July 2025Every page still undergoes HTML parsing, rendering, deduplication, and statistical models, such as BERT, for understanding and spam detection when it’s time to serve results. The same query‑interpretation pipelines and ranking signals, such as RankBrain, MUM, and other ML models, order information for both classic blue links and AI‑powered answers.
AI Mode and AI Overviews are simply new front-end features built on the familiar Search foundations that SEOs have been optimizing for all along.
Making The Most Of Google Search Console
Finally, Daniel Waisberg led a session on effectively utilizing Search Console in this new era.
Waisberg described Search Console as the bridge between Google’s infrastructure (crawling, indexing, serving) and your site. Key points that came from these sessions included:
- Data latency: Finalized data in Search Console is typically two days old, based on the Pacific time zone. Partial and near-final data sit behind the scenes and may differ by up to 1%.
- Feature lifecycle: New enhancements progress from user need to available data, then through design and development, to testing and launch.
- Recommendations feature: This tool is aimed at users who are not data experts, suggesting actionable improvements without overwhelming them.
By understanding how Search Console presents data, you can diagnose crawl issues, track performance, and identify opportunities for AI-driven features.
That’s it for the end of day one. Watch out tomorrow for our coverage of day two at Google Search Central Live, with more Google insights to come.
More Resources:
- Do We Need A Separate Framework For GEO/AEO? Google Says Probably Not
- How LLMs Interpret Content: How To Structure Information For AI Search
- The State Of AI In Marketing
Featured Image: Dan Taylor/SALT.agency
