Last night I published a research study on what actually motivates corporate leaders. Real data. 467 leaders. 32 motivational dynamics scored for each one. Correlation tables, variance analysis, statistical significance across dozens of pairings.
I don't have a statistics degree. I didn't hire a research team. I didn't spend six months in SPSS or R.
I loaded 450+ MCode assessments into Claude, each with 32 data points of motivational scoring, and I started asking questions.
Here's what I know: this is the part where AI actually changes things for people like me. Not generating blog posts. Not writing emails. Taking a massive dataset that would have sat in a spreadsheet for years and turning it into findings that nobody expected, including me.
How This Actually Worked
Let me be specific about what I mean, because “I used AI” is vague enough to be meaningless.
Each MCode assessment produces scores across 32 motivational dynamics. Things like Finish, Meet the Challenge, Realize the Vision, Take Charge, Collaborate, Be Unique, and 26 others. Every leader gets a score on each one. That's over 14,000 individual data points sitting in my dataset.
I knew the data was in there. I've been doing this work for years. But there's a difference between having a sense that something is true and being able to prove it across 467 people. I could tell you from coaching experience that most leaders aren't motivated by being in charge. But I couldn't tell you that Take Charge ranks 27th out of 32, that only 5.6% of leaders have it in their top five, and that nearly one in five scores below a 5. I needed the data to say that. And I needed a way to get the data to talk.
So I started simple. I asked Claude to rank all 32 motivations by average score across the full population. That took about ten seconds. What would have taken me hours in a spreadsheet, sorting and cross-referencing, just appeared.
Then I asked which motivations had the lowest variance, meaning which ones were most consistent across the entire population. That's the kind of question that matters enormously for understanding leadership but that most people wouldn't think to ask, because variance isn't intuitive. You have to already know that a high average with high variance means something completely different than a high average with low variance. I didn't need to calculate it. I just needed to ask for it.
Then I started asking about correlations. Which motivations travel together? Which ones are independent? Are there any that have effectively zero relationship with each other?
That last question is where things got wild.
The Finding I Wouldn't Have Found on My Own
When you look at correlations, you expect to find relationships. Two things go up together, or one goes up while the other goes down. What you don't expect is nothing. A flat line. Zero correlation.
But that's exactly what showed up with Meet Needs, the motivation to identify and fulfill the needs of others.
Meet Needs ranks 8th overall. Nearly 60% of leaders score 8 or above. Service orientation is genuinely strong in corporate leadership. But when you check its relationship with every achievement-oriented motivation in the dataset, you get numbers like -0.024, -0.018, -0.015, +0.007, +0.011. Those aren't low correlations. Those are absent correlations.
What that means in plain language: knowing that a leader is wired to serve tells you absolutely nothing about whether they're also wired to achieve. And knowing they're wired to achieve tells you nothing about whether they care about serving. The two engines run independently.
I would never have found that staring at a spreadsheet. I might have had a hunch. But the ability to say “show me every correlation between Meet Needs and the achievement cluster, and tell me which ones are statistically meaningful” and get an answer in seconds? That's the difference between intuition and proof.
What the Data Actually Said About Leaders
Once I had the tools to ask real questions, the findings came fast. Let me walk you through the ones that matter most.
The Leadership Signature. Five motivations rose decisively above the rest, with remarkably low variance. Finish (8.38 average), Meet the Challenge (8.32), Realize the Vision (8.31), Make an Impact (8.26), and Be Key (8.17). When we set the bar at 7 or above, more than 82% of leaders clear it on all five. That's not a trend. That's a shared foundation.
Finish is number one. Not ambition. Not influence. The motivation to look at a completed product and know you've accomplished what you set out to do. Seven out of ten leaders score 8 or above. Nearly half score 9 or above. The foundation of corporate leadership is completion.
The Grit Circuit. Meet the Challenge and Overcome have a correlation of 0.67, the strongest pairing in the entire dataset. Over half of all leaders score 8 or above on both. I asked Claude to identify every motivational pairing where more than 40% of leaders scored high on both, and this one jumped off the page. These are leaders who don't just face hard things. They need hard things.
The Quiet Achiever Effect. I asked Claude to create composite scores, one averaging “Visible” motivations (Evoke Recognition, Be Unique, Take Charge, Persuade) and one averaging “Quiet” motivations (Finish, Meet the Challenge, Meet Needs, Do It Right). The result: 81.6% of corporate leaders lean toward the quiet side. The average gap is 1.28 points, which is substantial on a scale where differences are usually measured in tenths.
The leaders in this study are not motivated by commanding a room. They're motivated by finishing what they started, meeting the challenge in front of them, and doing it right.
The Take Charge Paradox. Take Charge ranks 27th out of 32. Fewer than 6% of leaders have it in their top five. This is the finding I kept coming back to, because it cuts against nearly everything popular culture tells us about leadership. These leaders can take charge. It just isn't what fuels them.
The AI Part That Matters
I want to be clear about something. Claude didn't write the study. Claude didn't decide what questions to ask. Claude didn't know that the Meet Needs finding was surprising, or that the Take Charge result would matter to organizations trying to build leadership pipelines.
I knew what questions to ask because I've spent years in this work. I know what motivations people assume drive leaders. I know where the conventional wisdom is wrong. I know what would make an HR executive sit up in their chair.
What I didn't have was the ability to process 14,000+ data points, run correlation matrices, calculate variance, identify statistical outliers, and surface unexpected patterns, all in a conversation. I'm not going to learn R. I'm not going to hire a data scientist for a project like this. But I can sit with Claude for a few hours and ask increasingly specific questions, following threads as they emerge, and end up with findings that hold up to scrutiny.
That's the part people miss about AI. It's not about replacing expertise. It's about giving experts access to capabilities they wouldn't otherwise have. I brought 25 years of coaching and a deep understanding of motivational dynamics. The AI brought the computational horsepower to prove what I suspected and surface what I didn't.
The combination produced a study I couldn't have done alone. And honestly? It produced findings that surprised me, which is the whole point of research.
What This Means If You're Sitting on Data
Here's why I'm writing this on chrislema.com instead of just pointing you to the study.
If you have a dataset, any dataset, with enough depth and enough records, you are sitting on findings you don't know about yet. You don't need a data science team to find them. You need domain expertise and the right questions.
The AI handles the computation. You handle the curiosity.
I started with “rank these by average score.” I ended up discovering that corporate leadership runs on dual fuel, that service and achievement are independent engines, and that 82% of leaders share a motivational signature that looks nothing like the stereotype.
All of that was sitting in a spreadsheet. It just needed someone to ask.
If you want the full study with all the charts, correlations, and detailed analysis, you can read it on MotivationCode.com.