GeoVector co-founder and CEO Razvan Iordache took the mic for the Chop Chop Talk Shop — the rapid-fire sharing segment of Lorong AI's Coworking Mondays — this time at Lorong AI's new home at One-North. It's a relaxed afternoon of working, conversations, and light networking, and the Talk Shop is where members get a few minutes each to share what they're building. Razvan used his slot to walk through where GeoVector is today and to share notes from the Block71 x Microsoft AI Accelerate program we'd just been through.
AI Entered the Buyer Funnel
Razvan started with the shift everyone in the room had already felt: buyers now ask ChatGPT, Gemini, and Perplexity before they ever visit a website. The buyer has a question, opens an AI assistant, gets a shortlist, visits only the shortlisted sites, and buys. The brands AI recommends become the shortlist — everyone else is invisible.
- 900M+ weekly active users on ChatGPT alone (OpenAI, Feb 2026)
- 58.5% of Google searches now end zero-click (SparkToro, 2025)
- 94% of B2B buyers are using LLMs to decide (6sense research, 2025)
AI Is Recommending Your Competitors
The problem, he argued, is that AI is already recommending your competitors — you just don't know which ones, you don't know why, and you don't know how to fix it. Walking a Singapore banking example across the funnel made it concrete: ask "How can my debit card help me earn rewards in Singapore?" and the assistant names DBS and MyBank; ask which banks offer the best conversion rates to KrisFlyer miles and it returns Citi, DBS, OCBC; ask which bank can issue a rewards card the fastest and it answers OCBC. In each case a given brand is simply absent.
That leaves most brands with three gaps:
- No visibility — most brands have zero insight into whether AI assistants recommend them or their competitors
- No insights — no read on which AI conversations they're winning, losing, or missing entirely
- No fix — knowing where you're invisible is only half the problem; no tool exists to systematically close the gap
Track. Analyze. Win.
GeoVector is built to close that loop end to end, and most of it is live today across six AI platforms — ChatGPT, Claude, Gemini, Perplexity, Google AI Overview, and Google AI Mode:
- Connect — zero integrations to start; GeoVector generates your competitor landscape and custom prompts for you
- Prioritize — pick the high-intent prompts you want to win, with Ads and Search Console signals surfacing the ones costing you revenue
- Diagnose — weekly LLM response capture across all six platforms, with Share of Voice scored by a research-based, position-adjusted methodology across the Discovery, Research, and Decision stages of the funnel
- Fix — AI-optimized content targeting each citation gap: structured, sourced, and tuned to each platform's patterns
- Measure — track which conversations you're now winning, and which formats and sources each AI actually responds to
Where GeoVector Is Heading: The GEO | CMO Agent
The part Razvan was most excited about is what comes next. GeoVector isn't meant to stay a dashboard you check — it's becoming an AI agent that monitors your marketing 24/7 with GEO at its core: it watches every channel, finds the cross-channel patterns humans miss, and takes action with your approval (or autonomously). The compounding idea is that every channel feeds every other — a search term converting on Ads becomes a prompt to track and a content brief to write; a zero-click query with high impressions becomes a likely AI search to validate; a new post cited by Perplexity within 48 hours informs where ad budget can shift away from keywords AI now covers for free. The more channels connected, the smarter every decision gets — a loop no single-channel tool can build.
Notes from the Block71 x Microsoft AI Accelerate Program
Razvan also shared a quick founder update from the Block71 x Microsoft AI Accelerate cohort — a 10-week program run by Block71, NUS Enterprise, and Microsoft, anchored to a three-day kickoff in early March, for AI-native founders building toward enterprise revenue in Asia. The kickoff ran from foundations (the Microsoft AI and agentic landscape, why Singapore as a regional HQ) through building the engine (FinOps, AI product strategy and enterprise GTM) to going global (production readiness on Microsoft Foundry, foreign incorporation, legal structuring, and fundraising).
One framing from Summer Yang of Microsoft ASEAN stuck with us: "Building fast is no longer the hard part. Scaling right is." In the AI era, building is the easy part — scaling means designing GTM, trust, and channel into the product early, not as an afterthought.
Three honest takeaways we shared with the room:
- You get what you put in. The structure is there, but the depth — and most of the return — sits on the founder.
- Networking was the biggest win. Some of the cohort are now GeoVector users; the other founders, more than the curriculum, turned out to be the asset.
- Programs accelerate what's already moving. They're a multiplier, not a starting point — worth knowing before you go in.
Thank You
Thanks to the Lorong AI community for the new One-North space and for keeping Coworking Mondays one of the best places in Singapore for AI founders to share work in progress and pressure-test ideas.
Want to see what AI assistants are saying about your business — and where your competitors are winning the shortlist? Get in touch with our team.