When Technology Outruns Judgement - Why leadership design, not more tools, will decide who creates real value from AI.
When Technology Outruns Judgement
Why leadership design, not more tools, will decide who creates real value from AI.
AI has compressed weeks of professional work into days. Yet margins and client value aren’t keeping pace. The constraint isn’t the toolset, it’s how we design leadership and operating models around it. This piece explores how the smartest investments can slow firms down when capability design doesn’t evolve as fast as the tools.
The Efficiency–Value Gap: Why Your Smartest Investment Might Be Slowing You Down
Different sectors, same design deficit: faster work, unchanged value architecture.
There's a conversation happening in professional services firms that nobody quite wants to finish. It starts with excitement about productivity gains - research that used to take days now takes hours, analysis that once required senior people now generated in minutes, processes that consumed weeks now compressed into days. The AI revolution, we're told, is here.
However, someone might ask the awkward question: where's the value?
Not the efficiency gains - those are visible enough. The question is about the value that matters to clients, to shareholders, to the people who decide whether your firm is essential or merely useful. Because whilst your teams are processing work faster than ever, something curious is happening. Revenue growth remains stubbornly disconnected from productivity improvements. The machines are getting faster, but the commercial outcomes aren't following at the same pace.
Across the knowledge economy: four versions of the same pattern
Law: AI materially reduces time on first-pass research/drafting, yet pricing and operating models still default to hourly billing, so client value capture lags. GenAI is now embedded in drafting and research across much of the UK legal sector; 61% of lawyers report daily use, yet most firms haven’t changed pricing or workflows. [2]
Pharma: In life-sciences R&D, machine-learning models now predict toxicity and compound efficacy in hours, but the governance frameworks that determine what can move into clinical trial haven’t kept pace. Regulators themselves are still piloting AI internally, such as the FDA’s Elsa review platform - to accelerate their own processes. [3]
Finance: Risk/compliance automation improves accuracy and cycle time, but legacy controls and decision rights slow innovation unless leadership redesigns how decisions are made and measured. (Synthesis from BCG/McKinsey)
Consulting: Analysis that took weeks is produced in minutes, but fee models and propositions often don’t change - clients see speed without differentiated insight. [6]
Different sectors (amongst others), same design deficit: faster work, unchanged value architecture.
BCG’s 2025 global study (over 1,200 executives) finds that only about 5% of companies achieve measurable value from AI at scale, while roughly 60% report limited or no financial impact despite major investment. The small group that succeed - realising up to 5× more business value - combine technology adoption with redesigned leadership and operating models.[1]
This isn't a technology problem. It's a design problem, and it sits squarely in the domain of leadership.
The Quiet Erosion of Judgement-Building
As technology reshapes how work gets done, the erosion isn’t only operational, it’s developmental. The very systems that once built experience and judgement are thinning out. Take succession planning, for example. The pipelines that once grew future leaders through years of progressively complex work are being quietly hollowed out.
Between 60 - 85% of companies across regions now admit they have no credible successor identified for critical roles, according to recent surveys by Heidrick & Struggles, BCG, and CIPD.[4][8][9] Short CEO tenures - averaging between five and seven years in many professional services contexts - often prove shorter than the time required to identify, develop, and prepare credible internal successors.[5][4]
This isn't a recruitment problem - these firms aren't short of graduates. It's a development problem. When graduates enter professional services firms now, they encounter something subtly different from what previous cohorts experienced. Much of the routine analytical work, the foundational research, the first-draft production that used to occupy the first several years of professional development has been accelerated or automated. What remains are either tasks too simple to warrant human attention or problems too complex for someone still learning the craft.
The traditional pathway from junior execution to senior judgement is quietly being compressed or, in some cases, removed. Most leadership teams are only starting to notice the gap when someone attempts to step up and the capability isn't there.
In pharmaceutical research, for instance, computational tools can now model molecular interactions and predict compound efficacy far faster than traditional lab work. Yet as one recent analysis noted, AI accelerates parts of the research pipeline - particularly in early-stage compound identification and toxicology prediction - but the regulatory architecture and approval processes weren't designed for this pace.[3] Companies spend millions to accelerate research whilst their real bottleneck - human decision-making about complex risk-benefit trade-offs in regulatory and clinical contexts - remains unchanged.[3]
The paradox is this: firms are becoming more efficient at the precise moment they're becoming less capable of developing the judgement that creates lasting value.
When Fast Gets Expensive
In consulting specifically, the pattern has become visible enough to warrant academic attention. A Harvard Business Review analysis notes that AI is fundamentally changing how consulting firms structure their work, with artificial intelligence able to analyse competitive positioning or market trends in minutes rather than weeks.[7] Yet firms proudly demonstrate their new capabilities whilst their fee structures and value propositions struggle to evolve. What clients increasingly notice is that the insights, whilst faster, aren't materially different. The speed is impressive; the strategic value is unclear.[7]
Stanford's AI Index Report 2025 and similar research from McKinsey emphasises that AI's capacity to augment human work depends fundamentally on whether organisations have redesigned leadership capability alongside technological capability.[10][11] Simply adding AI tools to existing workflows creates efficiency but rarely captures new value. The firms seeing genuine transformation are those treating AI adoption as a leadership design challenge, not merely a technology deployment exercise.[9]
This is what might be called the efficiency-value gap. You can measure the former easily: productivity gains, time savings, cost reductions. The latter is more elusive. When clients choose you over competitors, what are they actually buying? When investors value your firm, what capabilities are they backing? And when your best people decide to stay or leave, what opportunities for learning and growth do they see?
Professional services firms specifically face what researchers term the "AI value capture paradox” - the disconnect between growing capability and flat or declining margin realisation.[7] Firms invest heavily in tools that make work faster, yet struggle to translate that speed into premium positioning or client willingness to pay more.[10]
The Leadership Architecture Problem
If you ask most senior leaders whether they have a leadership development programme, the answer is almost always yes. If you ask whether that programme was designed for the world they're actually operating in, the conversation becomes more interesting.
Most leadership development still follows a predictable pattern: identify high potentials, send them to training, give them stretch assignments, promote the successful ones. This worked reasonably well in environments where the basic structure of work - and the pathway from junior to senior - remained relatively stable. It breaks down when the structure itself is being redesigned by technology every eighteen months.
The real question isn't whether your managers can run a performance review or deliver difficult feedback - though both matter. It's whether they can do something considerably harder: make judgement calls about problems that have never been solved before, in contexts that are shifting faster than policy can keep up, with consequences that won't be fully understood for years.
Research from the Center for Creative Leadership and McKinsey emphasises that leadership in AI-augmented environments requires fundamentally different capabilities - the ability to diagnose what's actually happening in complex systems, to experiment intelligently rather than optimising prematurely, to build judgement in contexts where historical precedent is increasingly unreliable.[12][11]
The traditional pilot programme is instructive here. Most large firms now run pilots - small-scale tests of new technology, new processes, new ways of working. What's worth noticing is how few of these pilots actually scale. Research suggests that pilots often succeed locally but fail to become organisational practice because they're treated as proof-of-concept exercises rather than as learning systems.[1] The question isn't just "Did it work?" but "What did we learn about how our people need to think differently, decide differently, lead differently for this to become sustainable?” The difference between pilots that die in the middle and those that scale is leadership design - treating pilots as learning systems that test how people think, decide and collaborate, not just whether the tech works.
The Succession Question Nobody's Asking
Across regions, 60–85% of organisations report they have no ready successor for key leadership roles. [2][8] In the UK mid-market, only 58% of companies say succession even features in strategic planning. [9] Meanwhile, FT/25×25 data show that more than half of FTSE 100 CEO appointments in 2024-25 were external hires - evidence of thin internal benches. [15] Average CEO tenure remains 5–7 years, often shorter than the time needed to grow credible successors [5].
iMocha's succession planning metrics research, along with broader industry analysis, suggests that modern organisations need proactive, design-based approaches to developing future leaders - actual pipeline management focused on capabilities the future requires, not just replacement planning for current roles.[4][3] Yet succession planning remains reactive in many firms. According to the Route to the Top 2025 report, only 15 to 60% of companies have any form of continuous succession planning, with the lowest bench strength precisely where it matters most: PE/VC-backed firms and fast-moving technology companies.[4]
When 60 to 85% of firms across regions admit they have no credible successor for critical roles, the immediate response is often to intensify recruitment efforts or buy in senior talent. Both approaches miss the deeper issue: if you're not systematically building the capability your future leaders need, hiring experience from outside simply imports yesterday's thinking into tomorrow's problems.
The succession challenge isn't actually about succession - it's about whether your organisation is designed to develop the judgement your future needs. Many firms have leadership development programmes that prepare people to do what senior leaders do today. Very few have systems that prepare people to lead organisations that will look fundamentally different from the one they're currently in.
What Actually Needs Redesigning
Closing the Efficiency–Value Gap:
> Entrepreneurial Mindset build intelligent experimentation and commercial ownership inside today’s model.
> Future Leaders develops systemic foresight and judgement to design the next one.
> Leadership Evolution sustains energy and decision hygiene at the top so change holds.
If the problem is structural, the response has to be structural. This isn't about better training programmes or clearer competency frameworks - though both can help. It's about fundamentally rethinking how people in your organisation develop the capabilities you need.
This starts with a different question. Not "How do we train our people to use AI?" but "How do we build organisations where AI accelerates work whilst deepening rather than eroding human judgement?" The firms getting this right - and research suggests they remain rare - have made a deliberate choice to treat leadership development as a design problem, not a deployment problem.[1][9]
What does that look like in practice? It means creating structured opportunities for people to engage with genuine complexity and ambiguity, not just execute well-defined processes. It means building learning systems where experimentation is disciplined rather than chaotic, where people develop judgement through real stakes rather than case studies. Research from London Business School on leadership in 2025 emphasises that effective leaders increasingly need to operate in environments where traditional hierarchies and decision-making frameworks prove inadequate - requiring fundamentally different development approaches.[12]
The Innovators Within model draws on more than a decade of leadership and organisational work across sectors. While this integrated offer is new, the patterns it addresses—efficiency gains without value capture—repeat consistently across companies and industries.
Illustration (anonymised):
A scaling fintech engaged us to review its leadership team and decision-making, after growth stalled. Our audit found senior dominance in every material decision and a reluctance to confront known tensions between commercial and product leads. Marketing data showed attrition among existing clients and weak traction with new ones, yet leadership defaulted to familiar narratives. Within a four-hour diagnostic we surfaced the cost of these defaults - lost market share in a category they had originally led. The lesson wasn’t technological; it was architectural. When leaders mistake confidence for clarity, capability erodes faster than technology advances.
The technology didn’t change - the leadership design did.
At RenOC, through our Innovators Within programmes, we partner with scaling organisations across knowledge-based industries, from professional services and finance to technology, life sciences, and beyond. Most are growth-stage firms or divisions of larger enterprises facing inflection points: where technology is accelerating faster than leadership design. We help them build the capability to think, decide, and adapt at the pace their tools now enable.
Our clients typically encounter us at specific trigger moments: plateauing performance despite effort, growth milestones that expose leadership gaps (Series A/B/C rounds, M&A activity), leadership churn, or big external shocks where existing approaches clearly aren't sufficient.
Our Future Leaders, Entrepreneurial Mindset, and Leadership Evolution programmes address different layers of this same challenge - from preparing emerging leaders for complexity to equipping senior teams to sustain clarity and energy at the top. Rather than teaching people what AI is or how to use specific tools, we build the judgement, commercial reasoning, and experimental discipline that AI can't replicate. It's not training that prepares people for roles as they exist today, but architecture that develops leaders for organisations that don't yet exist - because in eighteen months, your organisation won't look like it does now.
This isn't about us - it's about the question every leadership team eventually has to answer: are you designing the organisation that can create the value your future requires, or are you optimising the one that solved yesterday's problems brilliantly?
The Cost of Inertia
The firms that will struggle most over the next five years aren't those who've been slow to adopt AI - they can catch up. They're the ones who've adopted AI without redesigning how they develop human judgement, commercial thinking, and strategic capability. They'll have all the efficiency gains and none of the value capture.
This isn't speculation. We can already see it in the data. The gap between leaders and laggards in professional services isn't primarily about technology adoption rates - many firms have adopted similar tools on similar timelines. Research consistently shows that the differentiator is leadership architecture: how firms have thought about building the capabilities that create value in an accelerated environment.[1][9][11]
PwC's AI predictions research notes that whilst 73% of executives believe AI will fundamentally transform their business, far fewer have addressed the human capability and leadership design questions that determine whether those transformations succeed.[13] The World Economic Forum's analysis of AI's impact on workforce development emphasises that education and capability building must evolve alongside technology, or firms face a widening gap between what their systems can do and what their people can make of that capability.[14]
If you're reading this as a CEO, CFO, or COO of a professional services firm, the question worth considering isn't whether your organisation needs to change. The question is: what's the cost of assuming that existing leadership development approaches will somehow produce different results?
The real expense isn't in the technology investments or the efficiency programmes. It's in the quiet erosion of capability-building that happens when you optimise for speed without redesigning for judgement. It's in the succession gaps that appear three years before you need to fill them. It's in the strategic opportunities you can't pursue because you lack people who can think about problems nobody's solved before.
How we help (in brief)
Every engagement under Innovators Within starts with a first-principles diagnostic pilot, not to fix pre-determined issues, but to surface what’s truly driving or blocking value. We separate symptoms from causes before acting. Sometimes the outcome is a redesign or coaching; sometimes, it’s recognising that what’s working simply needs reinforcing. Or sometimes, the conclusion is nothing needs to change, and that clarity alone creates value. The greatest risk isn’t doing too little; it’s changing the wrong thing.
We’re framework-agnostic; diagnosis first, prescription second.
Depending on what emerges from that discovery, the next step varies. Sometimes it’s focused coaching or team alignment work; sometimes it’s a deeper leadership or cultural intervention. Or sometimes, it is nothing at all. Over time, the most common patterns we’ve seen have shaped three core Innovators Within programmes - the ones that address recurring challenges we’ve uncovered in client work.
Entrepreneurial Mindset (for managers): builds intelligent experimentation and commercial ownership inside today’s model.
Future Leaders (for VPs/SVPs): develops systemic foresight and judgement to design the next model.
Leadership Evolution (for the top team): sustains energy and decision hygiene so change holds.
The efficacy gains are real - the real question is whether you’re improving what matters or just getting faster at what doesn’t. If you’d like to explore what this looks like in your organisation, we begin with a pilot conversation - a focused discovery designed to clarify what’s really happening before deciding whether change is needed.
References
BCG (2025). Are You Generating Value from AI? The Widening Gap.
https://www.bcg.com/publications/2025/are-you-generating-value-from-ai-the-widening-gap
LexisNexis (2025). Generative AI and the UK Legal Sector Report.
https://www.lexisnexis.co.uk/insights/genai-uk-legal-sector-report-2025
AI in Pharma / Regulators. Overview of AI applications in R&D and FDA internal pilot “Elsa” to accelerate review.
(Use your preferred article for the R&D overview, plus the FDA Elsa press note.)
Heidrick & Struggles (2025). Route to the Top 2025: The Ascent Redefined.
London Business School (2025). What Will Leadership Look Like in 2025?
Harvard Business Review (2025). AI Is Changing the Structure of Consulting Firms.
Clayton Chancey (2025). The AI Value Capture Paradox 2025. (LinkedIn analysis piece).
BCG (2024). European Leadership and Succession Bench Strength Survey.
Azets (2024). Business Succession Barometer UK 2024.
Stanford HAI (2025). AI Index Report 2025.
McKinsey (2025). Superagency in the Workplace: Empowering People to Unlock AI’s Full Potential.
Center for Creative Leadership (2025). Leading in the Age of AI.
PwC (2025). AI Predictions 2025.
World Economic Forum (2025). Education in the Age of Disruptive AI.
Financial Times / 25×25 (2025). External CEO Appointments in FTSE Companies (2024–25).
Karen Kwong is the founder of Ren Organisational Consulting (RenOC) and creator of the Innovators Within programmes. She leads a small network of coaches and organisational specialists working with growth-stage and knowledge-based organisations in sectors such as finance, professional services, technology, and life sciences. Together, they help leaders redesign how capability and culture create value in accelerated environments.
