A Q&A with Nemo Nemeth, Chief Data and Analytics Officer at True Technologies For leaders who grew up in recruiting—not data science—the pressure to become “data-driven” can feel daunting. To cut through the noise, I sat down with Nemo Nemeth, Chief Data and Analytics Officer at True Technologies, to talk about the reality of AI in recruiting, how search firms should actually be thinking about their analytics, and why skipping the data fundamentals is the biggest mistake you can make right now. The New Moat: Proprietary Data > AI Q: Everyone is talking about AI capabilities right now. As a base layer, why should executive search leaders focus on implementing a tool like Thrive TRM over other competitors? Nemo: In the age of AI, the real differentiator will be who has more proprietary knowledge. Large language models (LLMs) are going to be commoditized. Everyone will have access to the same baseline models. What matters is how you architect your entire system. Bar none, the biggest advantage in executive search will be who has the most proprietary data and how they use it to nurture relationships. It’s easy to map executives using externally observable data. The real question is: Which executive is actually going to pick up your phone call? For example, if you launch a high-profile search for a Chief Executive Officer at a world-renowned consumer brand, there are only going to be about 10 names big enough for that role. The differentiator isn’t knowing those 10 names; it’s using your proprietary intelligence and having built a relationship using the intelligence to get those 10 people to answer the phone.Thrive TRM acts as the collection point to systematically capture your team’s institutional knowledge so you benefit from the relationships you’ve worked so hard to build. Thrive’s vision is to surface the insights you didn’t know you already had. The Truth About LLMs and Analytics Q: Can a talent leader without a data team just plug an LLM into their systems to run their analytics autonomously? Nemo: Short answer: No. There’s a lot of noise around this. While text-to-SQL (asking an AI to write a database query) has been around for a while, analytics requires an inherent layer of interpretation. If you ask an AI, “How many customer orders did we have?” there is a deterministic answer. But human context knows you likely mean completed orders, not canceled ones. One of the bread and butter elements of an analytics team is trust that was built over time, often via mistakes that we learned about via collaboration. A good analytics team hears the unsaid. The second an LLM spits out data that looks slightly wrong, people will lose trust in it and you’re back to manually validating everything. Right now, our team uses AI to speed up our existing work—like helping write code for data pipelines—rather than creating a completely autonomous self-serving analytics layer. It’s possible to build, and it will spit out numbers, but the margin for error in our world is often very low. Getting Started: Advice for the Non-Technical Search Leader Q: If you work in executive search but don’t have a data background, what are the very first steps to becoming more data-driven? Nemo: You have to focus on the fundamentals. Full-on AI and the exciting use cases people talk about, many don’t realize that there is a path dependency. There are many steps that have to be done before you can really focus on that. Those include: Nail the foundational architecture: Data architecture and metric definitions—have to be rock solid before you can even think about AI. Define your metrics clearly: When you say, “How many searches did we have?” what does that mean? Are they open, closed, or on hold? When a search closes, is it when the offer is signed, or when the candidate starts? Clean your processes: Everything flows from clean data capture. If your outreach is double-logged or you have contact duplicates, your insights will be wrong. There’s a funny meme going around right now about skipping steps. It’s important that people realize that you cannot jump right into AI without tackling data first. P.S. If you need help with your data foundation, reach out. We help Thrive TRM customers think through the analytics that matter to them and how to get insights from their executive talent data via Thrive IQ dashboards. Metrics That Matter (And The Surgeon Paradox) Q: What specific metrics should search leaders be looking at to gauge success? Nemo: Focus on metrics where you have a locus of control. For instance, Time to Place a candidate involves external variables out of your control. Instead, look at Time to Introduce. But even then, define it: Does introduce mean you put them on a slate, or that a candidate call actually happened? Also, you must understand the nuance behind your numbers. I love telling this anecdote: Some of the best surgeons have the highest mortality rates. Why? Because the best surgeons take on the hardest cases. If a search firm boasts that they place candidates the fastest, that’s a number without context. CEO searches take longer than VP of Sales searches on average. But a complex VP of Sales search might take 130 days, while a CEO search for a highly attractive company might take 90. Context is everything. Q: What are some non-obvious analytics areas that could be interesting for executive search leaders to examine? Three come to mind: Network Freshness: How is your talent pool growing? Are you bringing net-new executives to the table, or are you just recycling the same candidates? Engagement: Within your talent network, how many people have you contacted in the past 60 days? Business Health: Is your search inventory growing or shrinking? This indicates your team’s actual capacity and throughput. Nemo’s Takeaways for Executive Search Leaders: Stop obsessing over the AI model, obsess over your data. The firm with the cleanest, most proprietary data will win the AI arms race. Define your terms. If your team doesn’t agree on what “closed” means, your AI won’t either. Contextualize your wins. Don’t punish your team for a long “Time to Place” if they are taking on the hardest searches in the market. Audit your base layer. Ensure the tools you are using integrate smoothly and capture institutional knowledge systemically. We’re hosting a community roundtable on March 18 where Nemo will be answering your specific data questions live! Register here.