Evidence-based leadership development means replacing intuition-based promotion and development decisions, who feels ready, who seems like leadership material, with measured, repeatable signals that can be checked, compared, and revisited over time. It's a genuine shift happening across talent and HR functions right now, moving away from gut-feel judgement calls that were never actually as objective as they felt in the room, toward decisions grounded in evidence that would hold up if someone outside the original conversation reviewed it later.
Why Intuition-Based Development Feels Objective and Usually Isn't
The uncomfortable starting point for this whole conversation is that intuition-based leadership decisions rarely feel like guesses to the people making them. A senior leader who's promoted people successfully before genuinely believes their read on the next candidate is sound judgement, earned through experience, not a coin flip dressed up in confident language. Sometimes that belief is justified. Often it isn't, because human judgement about leadership potential is subject to well-documented biases, a tendency to favour people who remind us of ourselves at a similar stage, a tendency to overweight recent, vivid performance over a longer, less memorable track record, a tendency to mistake confident communication style for genuine capability. None of that makes the judgement dishonest. It makes it unreliable in ways that are genuinely hard to notice from inside the decision.
The Uncomfortable Test: If you promoted someone last year based on 'they just seemed ready', could you name, specifically, what evidence led to that conclusion? If the honest answer is a feeling rather than a specific, checkable signal, that's intuition-based development, however confident it felt in the room.
What Counts as Evidence in Evidence-Based Leadership Development
Evidence, in this context, isn't a personality test score treated as gospel, and it isn't a single 360 survey either, though both can be legitimate inputs. It's a combination of signals that converge, repeatedly, over time, on a consistent picture. Decision quality, evaluated against the information available at the time a decision was made, not just the outcome. Development pathway progression, measured by concrete capability gained rather than training hours attended. Peer and direct-report feedback, gathered at more than one point in time so a single bad week or a single unusually strong quarter doesn't dominate the picture. Behaviour under genuine pressure, not just performance in calm, low-stakes conditions where almost anyone can look capable.
The distinguishing feature isn't any single signal, it's convergence. One strong data point is an anecdote. Three or four independent signals that consistently point the same direction, over a meaningful period of time, are evidence. That distinction matters enormously in practice, because a business moving toward evidence-based development can still get this wrong by treating one impressive data point, a single glowing 360, a single strong quarter, as sufficient proof, when the actual discipline requires the same conclusion to hold up across multiple, independent sources before it's trusted as a genuine signal rather than a compelling story.
- Decision Quality: Evaluated against the reasoning and information available at the time, not just whether the outcome happened to work out well.
- Pathway Progression: Measured by concrete capability gained, a decision category now genuinely owned, not by training hours attended or workshops completed.
- Multi-Point Feedback: Gathered at more than one moment in time, so no single unusually strong or weak period dominates the overall picture.
- Pressure-Tested Behaviour: Observed under genuine difficulty, not just calm, low-stakes conditions where almost anyone can appear capable.
Why This Shift Is Happening Now, Specifically
Two forces are converging to push this shift faster than it might otherwise have moved. The first is simply accumulated evidence of intuition-based development's failure rate: enough businesses have now watched confidently-promoted leaders struggle, and enough retrospective analysis has connected those struggles back to promotion decisions made on thin, unexamined evidence, that the pattern has become too visible to keep dismissing as individual bad luck. The second is that AI-assisted analysis has made genuinely evidence-based development far more practical to actually run than it used to be. Tracking decision quality, pathway progression, and multi-point feedback convergence across a large leadership population used to require an analytical effort most talent functions simply couldn't sustain manually. That constraint has substantially loosened, which has made the evidence-based approach newly viable at a scale it wasn't before.
The Risk of Overcorrecting Into False Precision
There's a real trap on the other side of this shift, and it's worth naming directly because I've watched businesses fall into it while genuinely believing they'd solved the intuition problem. Evidence-based doesn't mean reducing a person to a single composite score and treating that score as objectively true simply because it emerged from a system rather than a conversation. A poorly designed evidence framework can produce false precision, a number that looks scientific and is actually built on a handful of weak or biased inputs quietly aggregated together, which is arguably worse than honest intuition, because the number's apparent objectivity makes it harder to challenge than a stated opinion ever was. Genuine evidence-based development stays honest about the quality and convergence of its inputs, and resists the temptation to present a shakier read with more confidence than the underlying evidence actually supports.
- Require convergence, not a single strong signal — One glowing data point is an anecdote. Trust a conclusion only once multiple independent signals consistently point the same direction over time.
- Evaluate decisions against reasoning, not just outcomes — A good decision with a bad outcome and a bad decision with a lucky outcome look identical if you only track results. Evaluate the reasoning that was available at the time.
- Gather feedback at more than one point in time — A single 360 survey captures one moment. Development evidence needs multiple points so an unusual week doesn't distort the overall picture.
- Stay honest about evidence quality, resist false precision — A composite score built on thin inputs is more dangerous than an honest 'we're not sure yet', because it's harder to challenge once it looks objectively measured.
Why Manager Assessments Are Often the Weakest Input, Not the Strongest
One finding that surprises leadership teams the first time they build a genuine evidence framework is how weak a single manager's assessment turns out to be as a standalone signal, compared to how much weight it traditionally carries in promotion decisions. A manager's read is shaped by proximity, recency, and the specific working relationship, all real information, but also all subject to exactly the biases that make intuition-based development unreliable in the first place. That doesn't mean manager input should be discarded. It means treating it as one converging signal among several, rather than the dominant or sole input it's traditionally been, particularly given that a single manager relationship, however well-intentioned, is a sample size of one.
This reframing tends to be uncomfortable for managers who've built genuine trust in their own read of their people, and it's worth acknowledging that discomfort honestly rather than dismissing it. A manager's judgement isn't being declared worthless. It's being asked to earn its weight by converging with other, independent signals, the same standard being applied to every other input in the evidence framework, rather than being granted automatic authority simply because it comes from the person closest to the day-to-day work.
Building the Evidence Framework Without Overbuilding It
There's a real risk of building an evidence framework so elaborate that it collapses under its own weight, requiring more data collection than any talent function can sustain and producing analysis paralysis instead of better decisions. The businesses that implement this well start narrow: one or two genuinely high-stakes decision categories, promotion into the executive layer, selection for a critical successor role, rather than attempting to build evidence-based rigor across every people decision the organisation makes simultaneously. Proving the approach works on a bounded, high-value slice builds both the practical muscle and the organisational trust needed to extend it further, which is a far more sustainable path than an ambitious, organisation-wide rollout that burns out the team responsible for maintaining it within two quarters.
The narrow starting point also makes the false-precision risk easier to manage. It's much harder to quietly aggregate thin, biased inputs into an artificially confident score when the framework only covers a small number of genuinely consequential decisions being watched closely, than when it's been stretched across hundreds of routine promotions where nobody has the bandwidth to interrogate whether each underlying signal is actually as solid as the composite output implies.
How This Changes Actual Promotion Conversations
The practical shift shows up most visibly in how promotion decisions get discussed and defended. An intuition-based conversation tends to sound like a character reference: strong presence, clearly ready, the room just knows. An evidence-based conversation sounds different, and noticeably more specific: it points to the pattern across the last six material decisions this person made, describes how their 360 feedback has moved across three measurement points, and cites a specific example of how they handled a genuinely difficult situation under real pressure. That specificity isn't just more rigorous. It's also more defensible, both to the candidate who wants to understand what genuinely drove the decision, and to anyone reviewing the decision after the fact who wants to know it wasn't primarily a popularity contest dressed up in leadership language.
It's worth being honest that this transition is uncomfortable for leaders who've built real careers on trusting their gut, and who have, in fairness, often been right more often than random chance would predict. The goal isn't to dismiss experienced judgement entirely. It's to treat that judgement as one input worth weighing alongside converging evidence, rather than as the sole and sufficient basis for a decision that will shape someone's career and the organisation's leadership bench for years afterward. Experienced intuition that can point to specific evidence when asked is exactly what evidence-based development is trying to produce more of. Experienced intuition that can't, however confidently delivered, is precisely the pattern it's trying to move away from.
I'd add a final practical note on sequencing. The evidence framework doesn't need to be perfect before it starts changing decisions. Even an imperfect, narrow version, tracking just decision quality and multi-point feedback for one high-stakes role, produces a meaningfully better decision than pure intuition, and the framework itself improves through use, as gaps in what's being tracked become visible in practice rather than remaining theoretical concerns debated in a planning meeting before anything has actually been tried.
The Distinction That Actually Matters
Evidence-based leadership development isn't about removing human judgement from promotion and development decisions. It's about requiring that judgement point to something specific and checkable, converging signals across decision quality, pathway progression, multi-point feedback, and pressure-tested behaviour, rather than resting on a feeling that felt like certainty in the room. Most businesses making this shift discover the same uncomfortable thing early on: a meaningful number of their past confident calls wouldn't survive being asked to show their evidence. That discomfort is the actual signal the shift is working, not a reason to quietly abandon it and go back to what felt easier before.