George Assimakopoulos

Let’s start with you. Who are you, and what lens do you bring to understanding people online?

I’m a marketing strategist and social intelligence professional focused on understanding how people express intent, perception, and emotion online. As the founder of Metric Centric, my work centers on turning large-scale digital conversations into clear, actionable marketplace intelligence for brands, agencies, and organizations.

I bring a human-centered, data-driven lens to online behavior. Instead of relying on surveys or surface-level metrics, I analyze what people actually say and ask across social platforms, forums, reviews, and search. That means focusing on voice-of-customer insight, sentiment, topic relevance, and narrative formation - so we understand not just what audiences think, but why they think it.

A defining part of my perspective is the intersection of human conversation and machine interpretation. I study how AI systems and answer engines learn from public content, summarize it, and repeat it at scale - and how that process increasingly shapes brand perception. Because of this, I help organizations understand both human discourse and AI-generated answers as part of the same ecosystem.

My goal is always practical: translating conversational data into strategic guidance that improves positioning, content, trust, and competitive advantage. I view online behavior as a continuous stream of intent signals, where meaning emerges from context, repetition, and validation - not isolated clicks or impressions.

Simply put, I help organizations listen better - to people and to machines - so they can act smarter and make better decisions.

‍

What’s a working theory you have right now about how people behave online?

People do not behave online to broadcast opinions as much as they behave to resolve uncertainty, seek validation, and reduce cognitive effort. Most digital expressions - questions, comments, reviews, shares, and even silence - are an attempt to orient oneself within a social, informational, or emotional context.

I believe what people say publicly is shaped by immediacy, perceived audience, platform norms, and the availability of trusted reference points. As a result, digital conversations are less about fixed beliefs and more about moment-in-time intent.

Additionally, online behavior is increasingly co-authored by humans and machines. Search engines, recommendation systems, and AI answer engines do not merely reflect behavior - they shape it by reinforcing certain narratives, suppressing others, and accelerating consensus.

In this ecosystem, understanding people online requires listening not just to what is said, but to what is repeated, referenced, and learned.

‍

What’s an insight you surfaced that you still think about? What one stuck with you?

At Metric Centric, we’re always asking, “What will our clients actually do with our insights?” Data without direction has limited value. That’s why we focus on delivering insights that can be acted on to shape more strategic communication and informed decision making.

Voter sentiment in Virginia is not uniform statewide. It varies sharply by voting district, and winning elections requires district-specific messaging tied to locally prioritized issues - not broad, generic campaign narratives.

In 2021, businessman Glenn Youngkin was a first-time gubernatorial candidate running for governor in the State of Virginia. He did not have an established political presence nor much of a campaign platform. However, he did have a natural ability to communicate well to Virginian constituents on important statewide issues and concerns. Youngkin had to defeat six Republican candidates at the gubernatorial convention, and then confront an election challenge against Democrat and former-Virginia Governor Terry McAuliffe.

Our social intelligence analysis revealed that although key issues were discussed across Virginia, their significance and emotional resonance differed meaningfully by voting district. Treating the state as a single audience would have weakened Glenn Youngkin’s message relevance and voter impact.

By mapping unfiltered constituent conversations across all 11 voting districts, our team helped Glenn Youngkin speak authentically to local priorities, giving his campaign a strategic advantage over both primary rivals and Terry McAuliffe. This precision in message relevance helped convert voter attention into electoral support - and ultimately helped lead Glenn Youngkin towards winning the Governor’s seat.

‍

What’s the weirdest rabbit hole your work has ever sent you down? And what did it teach you?

I recently found myself explaining something uncomfortable to a client. I told them that when you ask an answer engine a seemingly harmless question - something like, “What do experts say about the future of remote work productivity?” - the response you get may not be grounded in primary research at all. There may be no real experts interviewed, no original data collected, and no firsthand analysis behind the answer.

They paused, then asked a fair question:
“Then why would I ever trust answer engines?”

That’s when I realized I had created an uphill battle for myself. In trying to warn them, I had inadvertently cast doubt on the very tools I was also advocating they understand. Explaining the importance of analyzing conversational data from both humans and machines suddenly became more complicated.

The lesson for me was clear: it’s not about choosing one over the other. There needs to be a deliberate balance - respecting human insight and lived experience, while also understanding how machines synthesize, amplify, and sometimes distort those signals. Only by prioritizing both can you see the full picture.

‍

What skills or mindsets do you think the next generation of analysts will need?

The next generation of social intelligence analysts will need to operate at the intersection of human behavior, AI systems, and decision-making. They’ll interpret reality, guide strategy, and help organizations understand how narratives form, spread, and mutate across platforms, cultures, and AI answer engines.

As AI gives more visibility to certain voices and narratives, analysts will need to pause and ask what’s being amplified - and what’s being left out. That means developing a strong awareness of bias and staying grounded in truth, context, and responsibility. Most importantly, future analysts must think critically and remember that visibility alone doesn’t make something true.

‍

What’s a niche community, account, or corner of the internet you’re watching right now? And why?

If I had to point you to one niche corner of the internet to watch closely right now, it would be: The open-source LLM / “local AI” community

This is where AI opinions are being formed across a global group of developers, researchers, and tinkerers who build, modify, and run large language models before they scale. These communities begin the debate on what AI models understand, which sources are trustworthy, and how answers should be framed. Those decisions quietly influence how AI systems interpret brands, facts, and authority long before marketing teams notice.

‍

Last non-work thing you read that shaped your thinking?

1929: Inside the Greatest Crash in Wall Street History - and How It Shattered a Nation

Andrew Ross Sorkin’s book feels very relevant today because it shows 

how modern crises are born from timeless human behaviors - even when the tools and technologies look new. AI, financial engineering, and new asset classes often carry the same promise - that structural change has eliminated risk. However, history says otherwise.

Crises are human before they become financial. Sorkin clearly explains how market crashes are driven by psychology, coordination failure, and fear - not just numbers. Even with better data and smarter machines, decisions are still made by people under pressure.

1929 isn’t just a history of a market collapse - it’s a guide to recognizing early warning signals when optimism fades into uncertainty, and uncertainty turns into fragility.

‍

‍