Authored by Abhilash Siddarkar. Senior Programmatic Manager, B2B, dentsu
AI agents are completing the research-to-checkout journey without a single click on a brand's website. Here's what marketers in APAC need to do and why the window to act is narrowing fast.
Imagine a consumer in Singapore opens Gemini and types: 'Find me the best noise-cancelling headphones under $600 and buy the top-rated one.' Thirty seconds later, the purchase is done, no product page visited, no ad clicked, no brand name searched. For the brand that didn't appear in Gemini's recommendation set, this wasn't a lost click. It was invisibility.
Agentic commerce - where AI acts as a personal shopping agent is growing at 30-40% annually, and AI-driven shopping traffic is expanding several hundred percent year over year. In APAC, 68% of queries are now zero-click journeys, AI completes product research 12 times faster than tab-based browsing, and trust in AI recommendations outpaces trust in search by 45%. Consumer electronics will feel this shift first: battery life, RAM, ANC (Active Noise Cancellation) performance, exactly the multi-variable comparisons AI agents are built to handle.
From Tab-Hopping to In-Chat Checkout
The traditional purchase journey - discover, research, validate, checkout hasn't disappeared. It has been compressed into a single AI interface.
| Phase | Traditional Era | Agentic Era |
| Discovery | Scroll through links and sponsored ads. | Describe the problem in natural language. |
| Research | Open 5-10 browser tabs across review sites and blogs. | AI cross-references all sources and delivers a synthesised comparison instantly. |
| Validation | Manually check specs on retailer pages. | User challenges the AI; answer refined in real time. |
| Checkout | Redirect to brand site; fill in form manually. | In-chat purchase via Google Pay; zero redirects. |
Table 1: How each stage of the purchase journey has shifted from SEO to GEO era.
Google's Universal Commerce Protocol (UCP), launched in January 2026, is the infrastructure making this possible. An open-source standard, it allows AI agents like Gemini to discover products, compare real-time pricing and inventory, and complete a purchase inside the chat via Google Pay. Brands that integrate via a JSON manifest become visible to the agentic layer. Those that don't are effectively absent from it.
| The cost of inaction is not just lost traffic. It is becoming invisible in the future path to purchase. |
Winning Recommendation Share: B2C And B2B
For consumer brands, success shifts from owning a search ranking to owning a recommendation. AI reads structured product data and third-party reviews not keywords. Google’s Merchant Centre attributes must be conversational ('built for gaming', 'perfect for commuters'). Review volume and quality are now the primary trust signal. And top-of-funnel media CTV, YouTube, OOH must be designed to shift users from generic queries to branded ones: not 'best laptop' but 'best laptop from [Brand X] for gaming.' That branded prompt locks in the agentic journey before it starts.
In B2B, the AI functions as a technical analyst, cross-referencing whitepapers, peer-review platforms (G2, Capterra) and analyst reports before recommending. Winning here means becoming the source those reports cite: structured technical content with Schema.org markup, deep niche use-case articles, and comparison-ready product documentation that AI can use to build decision matrices for procurement teams.
The Metrics That Matter Now
The measurement framework must evolve alongside the channel.
| Old KPI Era | New KPI Era | What It Measures |
| Search Engine Rank | Share of Model (SoM) | How often the AI cites your brand across relevant queries. |
| Click-Through Rate (CTR) | Branded Prompt Volume | Users asking for your brand by name inside the AI. |
| Website Traffic | Agentic Conversion Rate | Purchases completed without leaving the chat. |
Table 2: The KPI shift as agentic commerce scales.
Four Steps to Agentic Readiness
- Foundation audit - Audit Merchant Centre attributes and Schema.org markup for conversational parity.
- Strategic pivot - Move approximately 30% of the budget into placements and content opportunities with well-known and trusted publishers to improve visibility and reach that LLMs (Large Language Models) are more likely to cite.
- Salience Amplification - Run awareness-focused campaigns that consistently expose audiences to the brand, increasing the likelihood that they actively mention or search for the brand when engaging with AI tools. Measure success through growth in brand-matched AI queries and mentions rather than impression volume.
Example:
- A cybersecurity brand consistently promotes its expertise in securing remote workforces across video, display, LinkedIn, podcasts, and industry content.
- Over time, users begin searching for the brand more often and include it in AI prompts such as “Compare Brand X with CrowdStrike” or “Best solutions for remote workforce security.”
- Success is measured through growth in branded search demand, AI-generated brand mentions, AI Share of Voice, comparison query volume, and AI referral traffic, incremental lift in citations tracked using tools such as: Athena, Scrunch AI, Google Search console, GA4, Bombora’s (company surge platform) etc.
- Agentic integration - Integrate with UCP to enable in-chat checkout. Minimum viable step: ensure real-time data accuracy across Merchant Centre.
Brands that act early have an opportunity to influence the information ecosystem that future AI models learn from, creating a cumulative advantage that traditional media channels cannot easily match. That opportunity exists today, but it is unlikely to remain open indefinitely.