Why Hiring Innovators Doesn’t Make a Company Innovative

Why the search for innovators disappoints

When an organisation’s growth or transformation pipeline stalls, the reflex is to look for better people: recruit a Chief Innovation Officer, hire a specialist digital squad from a competitor, or bring in external consultants to generate ideas. Innovation capability building is the structural alternative: installing the shared methods, decision rights and measures that allow the people you already employ to diagnose a constraint and deliver a commercial result. Relying on recruitment instead of capability building disappoints for three reasons.

1. The role is defined as a personality rather than a discipline.Job descriptions that ask for someone “creative, entrepreneurial, and comfortable with ambiguity” describe a temperament, not an inspectable skill. Because there is no objective test for a temperament, candidates are hired on charisma in the interview and judged on impression afterward.
2. The new hire inherits the old decision system.A talented outsider recruited into an organisation whose funding gates, procurement rules and functional handoffs have not changed can only produce what that system permits. Within two quarters, their proposals are waiting in the same approval queues that slowed the existing team.
3. Individual star performance does not travel cleanly across organisations.Groysberg, Nanda and Nohria tracked more than 1,000 star equity analysts in the US financial sector who moved between investment banks: after changing firms, the stars’ performance declined sharply and stayed depressed for years, and the performance of the teams they joined fell too, because their prior results depended on firm-specific colleagues, data and routines that did not move with them. In a separate study of a US investment-bank division from 2003 to 2009, Matthew Bidwell found external hires were paid roughly 18–20% more than internal promotees into equivalent roles, received lower performance ratings during their first two years, and were significantly more likely to leave. Both studies examine financial-services populations, yet the mechanism they isolate — that much of what looks like personal brilliance is embedded in the surrounding operating system — holds wherever complex work crosses functions.

What capability rests on instead: three institutional conditions

An organisation has a capability when a result does not depend on which individual happens to be in the room. Before training individuals, leadership has to establish three institutional conditions around the work:

1. An inspectable method.A shared, repeatable discipline for framing a constraint, challenging the assumptions behind it, generating options from existing resources, and testing the riskiest assumption before capital is committed. Method is what makes the output inspectable by someone who was not in the room. When a team uses an explicit method, an investment committee can examine how the conclusion was reached rather than voting on how persuasively the presenter speaks.
2. Clear decision rights.Explicit agreement on who funds a small test, who stops a line of inquiry when an assumption fails, and which operational owner absorbs the design once it passes validation. Without clear decision rights, newly trained teams produce better-structured proposals that still pile up in an unowned queue.
3. Pre-agreed measurement.A definition of the operational or financial result the work is meant to move — and the baseline it will be judged against — agreed with finance and the receiving business unit before the work begins.

Three leadership competencies inside three architectural pillars

Within those institutional conditions, our work with leadership teams focuses on building three learnable competencies. None of them is an innate personality trait, and all three can be practised on live commercial problems:

1. Seeing the system.Reading the organisation as an interconnected working system rather than an organisation chart: tracing where decisions actually happen, which downstream step depends on an upstream rule, and which local incentive is producing today’s company-wide bottleneck.
2. Changing the level of the question.Moving deliberately between the immediate symptom, the structural category of problem it belongs to, and the unexamined assumption that makes it a problem at all. Most stalled initiatives fail because the team is working diligently at one level while the real constraint sits at another.
3. Organising creative work.Staging, evidencing and stopping non-routine work with the same rigour applied to operations — without pretending that an unfamiliar problem can be scheduled like a factory line.

These three competencies work inside three architectural pillars, and fail in familiar ways when any one is missing:

The international standards community has arrived at the same conclusion. ISO is drafting a competency framework for innovation management, ISO/DIS 56011, which entered its draft international standard ballot in August 2026: the field is treating innovation competence as a structured capability to define and build, not a persona to find.

What capability looks like in practice: two cases

When an intact team learns to see the system and change the level of the question on a live operational problem, the answer rarely requires hiring new specialists or buying new technology.

1. Seeing the system across functional boundaries (Case 07 · Client decision).A manufacturing client faced a growing order backlog and overflowing inventory (Case 07). Seen within functional silos, production appeared too slow and the warehouse appeared too small, so management planned a capital expansion to build a larger warehouse. When the team traced the system across functions, the backlog actually began in procurement’s performance measure: buyers rewarded on lowest unit price purchased volume-discount lots far exceeding what the production plan required, filling the warehouse and blocking needed inputs for live orders. Linking storage cost to procurement’s metric removed the cause of the backlog, and the CEO cancelled the planned warehouse expansion.
2. Changing the level of the question inside strict constraints (Case 03 · Designed).At a public hospital outpatient clinic treating people living with Alzheimer’s disease (Case 03), patients arrived dressed normally alongside family caretakers. When a caretaker stepped away briefly to register or pay, the patient sometimes walked out of the building because door staff could not distinguish a patient from a visitor. The initial proposal was GPS wristbands with mobile SIMs — a cost neither the public hospital nor low-income families could carry. Changing the level of the question from “How do we track a wandering patient across the hospital?” to “Where does the patient wait when the caretaker steps away, and who holds the handover?” produced a design built from resources already in the ward: a supervised waiting area where the caretaker hands the patient to a nurse in exchange for an identification card and presents the card at pickup. Inside the ward, anyone not in a hospital uniform is a patient, solving the identification problem at essentially no ongoing cost.

What this looks like in Thailand

The cost of trying to buy capability on the external market is particularly steep here. In Adecco Thailand’s survey of 2,456 workers, almost 60% expected a raise of at least 20% to change jobs, and 90% said they were open to new opportunities in the next year. A hire bought from outside at a 20% premium can be bought away by another group at the next 20% premium eighteen months later.

We see a recurring pattern across Thai conglomerates and listed groups: an organisation hires an external transformation team or retains an outside agency to run workshops; the advisers leave behind slide decks and prototypes; and within a year of their departure, the internal team reverts to its old routines because the method stayed in the consultants’ heads and the committee approval rules never changed. Capability that sits in how your own managers frame problems, test assumptions and make funding decisions does not leave when one person resigns.

Does a structured system suppress creativity?

It can, if it asks the wrong question. A system that demands certainty before funding selects against anything novel, because novel work cannot supply five-year certainty before it is tested. The difference lies in the gate question. “Guarantee the return before we approve a small test” suppresses innovation and rewards safe line extensions. “What would have to be true for this option to work, and what would it cost to find out in four weeks?” makes trying safe, because the cost of trying is explicit and bounded.

Where to start on Monday

Before approving a new innovation recruitment requisition or booking a generic training course, run three steps with your leadership team:

1. Trace the path of the last three proposals that stalled.Follow each stalled project from origin to the point where momentum died. Record where it waited, which decision was unclear, and which function could not absorb the handoff. That trace is a more accurate capability assessment than a skills survey.
2. Train intact work groups on a live P&L constraint, not cohorts on hypothetical ideas.Capability sticks when the people who share a real workflow — operations, finance, commercial and technical leads — learn the method together by solving a bottleneck their business unit is already measured on.
3. Require external advisers to transfer the method and leave an inspectable trail.Make it a condition of any consulting engagement that your own managers co-design the solution, own the assumption model, and can run the next cycle without the consultant in the room.

Where this does not apply

When you lack a specific technical, scientific or regulatory capability you cannot practically develop internally — such as a specialist engineering discipline, a clinical domain or a technology you have never operated — hire or partner for it. The mistake is using recruitment to solve a governance and method problem.