My Thoughts

You can't manage what you can't imagine, and too many organisations imagine innovation as a project you bolt on the side of the business like a new coffee machine. It isn't. It's a discipline, a culture and, frankly, a bit of organised chaos that needs a firm hand and a light touch at the same time.

Innovation project management is where the rubber meets the moonshot. It's not just about building a prototype; it's about steering teams through uncertainty, making choices with incomplete information, and keeping sponsors comfortable without letting them strangle creativity. Over the years I've seen the good, the bad and the painfully bureaucratic. There's a better way, and it starts with treating innovation projects differently from your average IT upgrade.

Why innovation projects are different

Traditional projects are recipes: inputs, process, outputs. Innovation projects are experimental cookbooks, you try things, taste them, burn a batch and learn. That unpredictability means the usual Gantt charts, waterfall milestones and command and control governance will often slow you down or kill the idea outright.

A few features that set innovation projects apart:

• High uncertainty: market, technology, and customer responses are often unclear
• Learning over delivery: early stages prioritise validated learning, not polished deliverables
• Cross functional complexity: you'll need people who don't usually sit together, product, research, legal, sales, ops, all in the same room
• Rapid iteration: prototypes, tests, fast feedback loops
• Distinct risks: reputational, regulatory and market timing, rather than just cost and schedule

Two things I'll say that many won't like: big corporates can scale innovation better than startups if they stop pretending they can copy a lean startup model verbatim; and governance, when designed cleverly, actually helps, it directs scarce resources rather than choking creativity. Both statements raise eyebrows, but they're true in practice.

Principles that actually work

First principle: embrace a dual operating model. Run the Business; and run the future of the business. Treat innovation units as semi autonomous entities with a clear set of metrics aligned to strategic outcomes, not just vanity metrics. Give them permission to fail fast, but ask them to show how learning translates into value. This is different from absolving them from accountability.

Second principle: be deliberately cross functional. Put people together who disagree. Mixed teams, engineers, designers, commercial folk, regulatory, bring the tension that surfaces the real trade offs. Conflict, early and constructive, prevents nasty surprises later.

Third principle: define a clear taxonomy for innovation. Is this incremental improvement, adjacent expansion, or radical disruption? Without clarity, your portfolio will mix pilots with bets on the future and confuse sponsors and customers alike.

Methodologies and tools, pick what helps you learn

Agile frameworks, Lean Startup, Design Thinking, they're all useful, but none is a silver bullet. I often see organisations apply them as ideology rather than instruments. Use Agile for speed; use Lean Startup to test market assumptions; use Design Thinking to uncover human led problems. But don't be a zealot. Blend them.

Practical toolkit:

• Rapid prototyping and MVPs: test riskiest assumptions quickly
• Time boxed sprints with clear learning objectives, not just features
• Kanban for flow and visibility
• Customer interviews and ethnography for real insights
• Experiment scorecards: hypotheses, metrics, and decision rules
• Stage gate adapted for innovation, lightweight, with learning gates, not approval gauntlets

Here's an unpopular take: heavy duty product management tools don't guarantee success. They help documentation and scaling, but the core of innovation is the human conversation and the decisions you take after a customer test. So spend less on tools, more on structured human time.

Risk and governance, the tidy bit people skip

Risk management in innovation must focus on three buckets: value risk (will anyone buy this?), execution risk (can we build and scale it?), and external risk (policy, reputational, partner ecosystems). Traditional risk matrices are useful but place too much emphasis on cost and schedule. For innovation, the question is whether an experiment meaningfully reduces the principal unknowns.

Design a governance system that:

• Allocates clear decision rights at each stage
• Uses pre agreed exit and pivot criteria
• Reserves a small "option pool" budget for experiments that might scale
• Keeps senior leaders informed with concise, learning focused dashboards

Two metrics I always insist on being visible: time to validated learning and customer adoption velocity. Time to market is noisy; what you really need to measure early on is how fast you can confirm whether the idea is viable or not.

Balancing exploration and exploitation

This is the classic ambidexterity challenge, invest in the now while funding the next. Many organisations over index on exploitation, squeezing efficiencies out of current business lines, because it feels safer and directly impacts the bottom line. But exploration is the long game.

A practical portfolio approach works: allocate roughly 70% to core activities, 20% to adjacent spaces and 10% to transformational bets, adjust the split based on sector and appetite. These aren't laws; they're starting points that force conversation. Change the allocation as your Organisation proves out or kills ideas. Flexibility is the point.

Culture and people, the soft scaffolding

Culture isn't a poster; it's patterns of behaviour. If you want teams to experiment, reward evidence based decisions. Celebrate when a test fails but teaches you something. Make psychological safety real, not just a training slide. Leaders have to role model curiosity and tolerance for reasonable risk.

Hiring is crucial. Bring in generalists who can translate between functions and specialists who bring deep craft. Rotate people through innovation teams for limited terms so you spread capability. And for the love of good data, train people in basic experiment design and interpretation. You don't need everyone to be a statistician, but you do need them to know what a viable learning pulse looks like.

Common roadblocks, and how to get past them

Resistance to change is a human thing. People worry about redundancy, competence and reputation. Address it by involving stakeholders early, sharing learning, and staging rollouts. Use pilot programs to reduce fear, small, live, measurable. Transparency beats surprise.

Resource constraints often slow innovation. Prioritise ruthlessly. Not every idea deserves funding. Tie resourcing to the risk reduction ladder: the further an idea is from the market, the more prototyping and evidence you should require before scaling investment.

Lastly, a practical tension: speed vs. thoroughness. Speed often wins in markets that reward first mover advantage, but thoroughness protects long term brand and regulatory standing. Pick your battles, move fast where you can, slow down where you must.

Leadership: guiding without suffocating

Leaders should frame the vision, not define every solution. Set clear boundaries, strategic intent, target customers, constraints, and let teams experiment within them. Sponsor involvement matters, but sponsorship doesn't mean micromanagement. Sponsors should be teachers of constraints, not makers of minute decisions.

A short checklist for sponsors:

• Clarify the strategic hypothesis
• Agree the decision milestones
• Define what success looks like numerically and behaviourally
• Commit to a cadence of review that focuses on learning rather than blame

Metrics that matter

Vanity metrics are the enemy. Early stage innovation needs metrics that indicate learning and potential: conversion rates from prototype tests, retention for early adopters, cost per experiment, and importantly, the number of validated hypotheses per quarter.

A reminder: not all innovations will scale. In fact, failure to scale is common. It's not shameful, it's data. Use failed experiments to inform the next bet. That's how cumulative value is created.

Practical example, how this looks in the room

Start with a clear problem statement. Not: "We want to disrupt X." But: "Our Customer is struggling to complete Y because of Z, leading to A impact on retention." Define the riskiest assumption, design a 2 to 6 week experiment to test it, run the test with real users, and decide based on pre agreed criteria. Keep reports tight: one page for executives that answers three questions, What did we test? What did we learn? What's the next step?

We see this model work across industries in Sydney, Melbourne and Brisbane, from finance to logistics. The common thread is deliberate constraint and ruthless learning.

One statistic to keep you honest: many large change programs struggle to meet their goals. For example, McKinsey's work on transformations has often noted high failure rates, a reminder that without the right practices, innovation projects easily derail. Use that as motivation to get structure right from the start.

Final thoughts, a few contrarian pushes

1. Stop worshipping the "fail fast" mantra. Fast failure without reflection is wasteful. Fail fast, learn faster
2. Don't centralise innovation into a black box. Distributed capability with clear hubs and spokes usually wins
3. Invest in "translators", people who can take messy, qualitative insight and turn it into testable hypotheses. They are worth their weight in gold
4. And yes: scale matters more than novelty. A brilliant idea that lives in a lab has zero impact. Focus on pathways to scale from day one

We help organisations structure these conversations, not by dictating one size fits all frameworks, but by designing governance, coaching teams on experiments and helping leaders make trade offs without panic. Practical frameworks, honest debates, and a willingness to learn are the real currency of innovation.

Innovation projects are messy. They should be. They're testing the edges of what your Organisation can become. The job of a good project manager is to keep the experiment honest, the team safe to try risky things, and the sponsors informed but not controlling. That's the balance. Achieve it and you won't just deliver projects, you'll build future businesses.

Sources & Notes

McKinsey & Company, referenced for general findings on transformation success rates and the challenges of organisational change (McKinsey has published multiple articles and research notes on transformation failure rates and organisational agility; consult relevant McKinsey transformation literature for detailed figures).