Your people are fine. It's the system that's broken.
That's a pretty bold claim, but the evidence keeps pointing the same way. Every year organisations invest heavily in training, leadership curricula, change frameworks and compliance systems. Every year execution stays patchy, the same performance gaps reappear, and managers re-run the same corrective conversations they ran last quarter. And every year the diagnosis is the same: the people must need more training.
Except they don't.
Recent research across organisational psychology, performance science and workforce attitudes suggests the problem is rarely a lack of capability, and almost never a lack of intent. People mostly know what to do. Many have the skills to do it. What they lack is a working environment that lets them deploy any of it. Rummler and Brache called it in 1990: "Pit a good performer against a bad system and the system wins almost every time." Thirty-five years on, we're still betting on the performer and blaming them when the system wins.
Some numbers: Ipsos Karian and Box surveyed 5,000 UK workers and found one in three saying their organisation actively makes it harder for them to do a good job. Asked what gets in the way, they don't cite motivation or mindset. They cite slow processes and approval loops (28%), unrealistic workloads (28%), outdated technology (28%) and too many meetings (27%).
One respondent: "A decent IT system would be a good start." And where organisations make work harder, motivation, pride and advocacy collapse by between 37 and 54 percentage points compared with organisations that make work easier. No engagement campaign ever built moves the dial 54 points. (Though that won't stop anyone commissioning the posters and the launch video!)
So if capability isn't the problem, why do our interventions keep missing? There are three reasons:
The Corporate Research Forum's work on HR's contribution to performance (Briner et al, 2023) contains a phrase I fully intend to steal. One Chief People Officer describes "flower-arranging HR" – policies, frameworks and processes that are beautifully put together and of no use to anyone. Its close cousin is HR theatre: furious activity dressed up in vanity metrics. Attendance figures. E-learning completions. Happy sheets. A Net Promoter Score for a compliance module, as if anyone has ever recommended one to a friend.
The signature move of HR theatre is the tweak. Performance disappoints, so we redesign the appraisal: change the form, move from a three-point scale to a five-point scale, go from annual reviews to quarterly ones. In a large enterprise that's £/$millions in employee and manager time. The CRF researchers note that when leaders proposing such a redesign are asked for one case study where it produced a demonstrable commercial outcome, the conversation goes quiet. It always does.
BTW: beware anything described as cutting-edge. Fads are, by definition, novel, which means they carry the least evidence of anything on offer. Hyped by vendors as the answer to everything, adopted with enthusiasm, abandoned at the first disillusionment in favour of something newer and shinier. Sound familiar? Thought so.
Performance management rests on a comforting fiction: strategic objectives cascade neatly from the boardroom down to individual goals. The CRF's practitioners are blunt – the ladder does not exist. Goals at the top translate into nothing intelligible four levels down. Instead, we run "set and forget" annual objectives while middle managers measure what's easy to count rather than counting what matters. People work hard, hit their targets, and change nothing the business cares about. That isn't underperformance. That's the system performing exactly as designed.
A systematic review of three decades of performance feedback research (Heine, Stouten & Liden, 2025) found 75 different labels for feedback, defined in contradictory ways – and more than half the published studies didn't say whether the feedback studied was positive or negative. Thirty years of research and half of it can't tell you if the message was "well done" or "must try harder". This is what passes for an evidence base.
One useful finding does emerge: the feedback sandwich – taught on management courses since roughly the Bronze Age – doesn't work. Positive and negative feedback travel on different emotional pathways. Praise reliably improves performance; criticism is volatile, triggering overload, defensiveness and shame.
Wrap the criticism in praise and the praise evaporates. The recipient remembers only the filling, and not fondly.
What works better in a time-limited conversation is the reverse: corrective message first, time to absorb it, reinforcement after. And the real buffer isn't sequencing at all, it's the relationship. Where trust exists, correction lands as development. Where it doesn't, no choreography will save you.
There's a deeper principle underneath all three. Behaviour is a state, not a trait. Work by Unlocking Human Potential (2026) on "behavioural readiness" describes performance as the real-time output of five interdependent conditions – cognitive, emotional, motivational, physiological and interpersonal – any of which can be degraded by overload, threat, fatigue or a hostile climate.
This explains something every trainer has witnessed: the delegate who performed beautifully on the course and reverted to previous practice the moment they got back to the day job. The knowledge didn't disappear. The conditions needed to access it did.
Teams? Same story. Research suggests it takes six to eight months for a group to become genuinely high-performing, only about 15% ever get there, and 80–90% hit serious trouble on the way (Bakke & Johansen, 2024). What separates the 15% isn't talent, it's structure, role clarity and the unfashionable habit of facilitated reflection.
Rolfe's three questions will do: What (happened/did we do)? So, what? What next? Simple, evidence-based, and nobody's selling it at a conference. Draw your own conclusions.
So, what would working on the system look like?
None of this is glamorous. There's no launch video. But the evidence is unanswerable: our people are mostly fine. It's the system that needs the development plan – and as performance improvement professionals this is a very large elephant in a very small room. We need to talk about it or be trampled underfoot.