Issue #9: The Portfolio Illusion
Why Most Innovation Portfolios Are Not Real Portfolios
In finance, a portfolio is not a list of things you own. It is a deliberately constructed set of positions, selected against an investment thesis, measured against expected returns, and actively managed as an interdependent whole. Positions get sized, re-balanced, hedged, and killed. The portfolio exists as a system. Every position is understood not just on its own terms but in relation to every other position in the set.
Now consider how most corporate innovation functions use the word “portfolio.”
They mean: the things we are currently doing.
That is not a portfolio. That is a collection. And the distance between those two words is where most innovation value gets destroyed.
The Collection Problem
Most organisations arrive at their “innovation portfolio” the same way. Individual initiatives get proposed, evaluated, and approved through separate business cases. Each one clears its own hurdle. Each one gets its own budget line. Each one reports on its own progress against its own milestones.
At some point, someone assembles a slide that shows all of these experiments, options, ventures or initiatives on a single page. Maybe they get sorted into horizons. Someone may draw a 2x2 with risk on one axis and strategic fit on the other. The slide gets presented at a quarterly review. Everyone nods. The word “portfolio” is used.
But nothing about that process constitutes portfolio management. No one asked whether these ventures, taken together, represent a coherent set of strategic bets. No one assessed whether the risk is concentrated or distributed. No one has the authority to move resources from one initiative to another based on relative performance. No one is measuring return at the portfolio level.
What exists is a collection of individually approved projects that are visible on the same slide. The slide creates the appearance of portfolio thinking without any of the underlying mechanics.
Why the Gap Exists
The collection-to-portfolio gap is not an oversight. It is a structural consequence of how many organisations set up their innovation functions. The gap has at least four root causes, and they compound.
The mandate is undefined.
Before you can construct a portfolio, you need to answer a question that most organisations skip entirely: what is the role of innovation in this organisation’s strategy, and what methods are in scope?
This is a scoping decision that sits upstream of everything else. It determines which instruments are available to the innovation function or functions and which belong elsewhere. An organisation might decide that its innovation mandate covers internal venture building and spinouts, but that partnerships and strategic investments sit with corporate development. Another may include invest-to-partner models within the innovation remit. A third might limit the scope to incremental improvements within existing business lines.
None of these choices is inherently wrong. An organisation that says “we build from within and occasionally spin out” has made a legitimate strategic decision. It narrows the option set but creates clarity about what the portfolio can contain, what capabilities are needed, and what a return looks like.
The problem is when this scoping decision never gets made. Innovation becomes whatever anyone decides it is. Different business units pursue different types of activity under the same label. Some run small incremental projects. Some attempt genuine ventures. Some call process improvements “innovation” because the word carries budget protection.
Without a mandate, there is no boundary. Without a boundary, there is no portfolio. There is just accumulation.
The instruments are treated as interchangeable.
Even where a mandate exists, most organisations fail to recognise that different types of innovation activity are fundamentally different instruments.
Building a venture from within the organisation is not the same as investing in an external startup. A spinout operates under different constraints than a partnership. An invest-to-partner model, where you take a position in an external company with the intent of integrating capability over a three to five-year horizon, looks nothing like a twelve-month internal build.
In finance, no one would manage equities, fixed income, and options as if they were the same thing. The portfolio exists precisely because these instruments are different, and the value comes from how they interact as a set.
Most innovation functions treat every initiative as the same type of bet. The same stage-gate process. The same milestone structure. The same quarterly reporting cadence. This flattening erases the very differences that portfolio construction is designed to exploit.
The structural design prevents portfolio-level decisions.
How the innovation function is organised determines whether portfolio management is even possible.
In a centralised model, a single team owns the innovation mandate. This creates the possibility of portfolio-level authority: someone who can see the full set of bets and make trade-offs across them. But centralised models often lack proximity to the business context where ventures need to operate.
In a decentralised model, each business unit runs its own innovation activity. This provides business context but eliminates portfolio-level visibility. No one can see all the bets. No one can reallocate resources across the set. Each unit optimises locally, and the organisation’s total innovation spend is just the sum of uncoordinated parts.
The hub-and-spoke model is the most common attempt to resolve this tension. A central hub holds processes, standards, budget, governance, and capability. Spokes in business units run execution closer to the operating context. Larger or cross-cutting bets escalate to the hub. Day-to-day innovation stays in the spokes.
In theory, this combines centralised coherence with decentralised proximity. In practice, it drifts. The hub becomes administrative rather than strategic, running reporting cycles and stage-gate reviews rather than making portfolio decisions. The spokes diverge in their methods and definitions, each developing its own view of what counts as innovation. Decision rights between hub and spoke become contested. The quarterly review becomes a status update rather than a governance forum. And no one is confident about where portfolio-level authority actually sits, because in practice, it sits nowhere.
This is also where Issue 8’s stop authority problem becomes a portfolio-level problem. If no one in the structure has the authority to kill an individual initiative, no one can rebalance the set. Stop authority at the project level is a prerequisite for portfolio management. Without it, the portfolio can only grow. It can be difficult to reshape.
The structural design choice matters less than whether the chosen structure enables someone to manage the portfolio as a whole. If no one has authority over the set, the set cannot function as a portfolio regardless of how it is organised.
There is no connection to strategy.
A portfolio without an investment thesis is just a scatter plot.
In finance, the thesis determines the construction: which sectors to overweight, which risks to accept, and which time horizon to optimise for. The portfolio is the execution layer of the thesis.
In corporate innovation, the equivalent is the connection between the innovation portfolio and the organisation’s strategic plan. The portfolio should be the set of strategic options the organisation is actively pursuing to create future value. Each position should trace back to a strategic question the organisation needs to answer or a capability it needs to develop.
Not every bet needs to map neatly to an existing strategic priority. A well-constructed portfolio includes exploratory positions: deliberate, bounded investments in areas where the organisation has no current thesis but wants to develop one. Adjacency plays. Emerging technology watches. Genuine moonshots where the organisation is spending to learn whether a space is relevant to its future.
The critical distinction is that exploratory positions are still deliberate. They are in the portfolio because someone decided to allocate resources to that space, not because they drifted in through an enthusiastic business unit. They still carry a return expectation, even if that return is defined as “develop enough understanding to make a strategic decision within eighteen months.” Exploratory does not mean exempt from rigour.
When innovation operates as a standalone function with its own strategy, disconnected from the organisation’s strategic planning process, the portfolio becomes self-referential. It serves the innovation function’s goals rather than the organisation’s strategic needs. This is how you end up with innovation teams that can demonstrate impressive activity metrics while the organisation’s strategic gaps remain unaddressed.
The connection needs to run in both directions. Strategy should inform what the portfolio contains. And the portfolio’s performance should inform strategic planning, updating the organisation’s view of which options are maturing, which are failing, and where new bets are needed.
To put it plainly, many innovation functions have no defined boundary around what they are responsible for, treat fundamentally different types of bets as if they were interchangeable, are structured in ways that prevent anyone from managing the set as a whole, and operate with little or no connection to the organisation’s strategic plan. Any one of these gaps would prevent real portfolio management. Most organisations have all four. That is not a portfolio with room for improvement. It is a collection with a label.
Before You Diagnose: The State of Play
Before applying any diagnostic, there is a prior step that most organisations skip. And it is broader than most expect.
You need an inventory. Not of what the innovation function is currently running, but of everything the organisation is doing that could constitute innovation or venturing activity, anywhere, under any label.
This means looking beyond the innovation team’s pipeline. Corporate development might be running invest-to-partner plays that are functionally innovation bets but governed as transactions. A business unit might have a skunkworks project that nobody reports centrally. The strategy team might be funding exploratory research that never touches the innovation function’s pipeline. M&A might be doing acqui-hires that are capability bets dressed as deals.
The broad state of play needs to capture each activity’s type, stage, committed resources, expected return, and the strategic question it is intended to answer. It also needs to note where the activity currently sits organisationally and who governs it.
Most organisations cannot produce this on demand. When they try, the exercise itself becomes revealing. Initiatives surface that no one at the centre knew were running. Committed resources turn out to be higher than anyone estimated because costs are spread across business unit budgets that are never consolidated. Several initiatives turn out to be pursuing overlapping objectives with no coordination between them. Activities that one function calls “innovation”, another calls “business development” or “continuous improvement,” and the definitional inconsistency is itself a finding.
The broad inventory matters because it inverts the usual sequence. Most organisations define the mandate first and then inventory what falls within it. This is circular; you end up scoping around what is already visible rather than what should be managed. The stronger sequence is to see the full landscape first, then draw the boundary. The mandate becomes a decision informed by evidence rather than a top-down decree that inherits whatever definitional problems already exist.
Once the mandate is defined, the state of play narrows to the portfolio boundary. But it does not become a one-time exercise. It is the foundation of portfolio governance: updated regularly, reviewed against strategic priorities, and used as the basis for rebalancing decisions. If you cannot produce an accurate state of play, you cannot manage a portfolio. You are navigating without an inventory.
The Portfolio Reality Diagnostic
Six questions to help determine whether what the organisation has is a portfolio or a collection.
Can you define the mandate?
Write down, in one paragraph, what your innovation function is authorised to pursue and what methods are in scope. If you cannot do this, or if the answer is “it depends on the business unit,” you do not have a portfolio boundary. You have an accumulation.
Can you produce a state of play?
List every activity across the organisation that could constitute innovation or venturing, not just what the innovation function manages. Include its type, stage, committed resources, and expected return. If you cannot produce this list within a week, if the exercise reveals activities no one at the centre knew were running, or if different functions cannot agree on what counts, you do not have portfolio visibility. You have fog.
Can you name the instruments?
Take that list and categorise each initiative by type: internal build, spinout, partnership, investment, invest-to-partner, exploratory position, or other. If every initiative is the same type, your portfolio has no diversification. If you cannot categorise them because the types have never been distinguished, your portfolio has no composition logic.
Can you state the expected return (desired outcomes) for each position?
For each initiative, articulate the return the organisation expects and the time horizon. The return does not have to be financial. It can be strategic optionality, capability development, or market learning. But it must be defined. If most initiatives have no stated return expectation, you have no basis for portfolio-level measurement.
Can one person or body rebalance the portfolio?
Identify who has the authority to move resources between initiatives, accelerate one at the expense of another, or kill a position based on the portfolio’s overall performance. If the answer is “no one” or “it would require separate approvals from each business unit,” you have a collection governed by its parts, not a portfolio governed as a whole.
Can you trace each position to a strategic question?
For each initiative, identify the strategic question it answers or the strategic capability it develops. This includes exploratory positions, which should trace to a question the organisation is actively trying to answer, even if the answer is not yet clear. If many initiatives cannot be connected to the organisation’s strategic plan or to a deliberate exploration thesis, the portfolio is self-referential. It serves the innovation function, not the organisation.
Reading the pattern: If you answered “no” to one or two of these, you have specific gaps to close. If you answered “no” to three or more, what you have is not a portfolio in any meaningful sense. It is a collection of individually approved activities that share a label but lack the connective tissue that enables portfolio management.
But We Are Different
Two versions of this argument appear reliably when portfolio discipline is proposed.
“Innovation is inherently uncertain. You cannot apply portfolio rigour to something this unpredictable.”
This inverts the logic. Uncertainty is precisely why portfolio discipline exists. Financial portfolios are not built for environments of certainty. They are built because individual positions are unpredictable, and the portfolio is the mechanism for managing that unpredictability across a set of bets. Diversification, rebalancing, and position sizing all exist because no one knows which individual bet will pay off.
The same principle applies to innovation. Individual ventures are uncertain. That is not a reason to abandon portfolio logic. It is the reason you need it.
A 2022 study published in Production and Operations Management by Si, Kavadias, and Loch at Cambridge Judge Business School found that the results of innovation portfolio management processes remain widely perceived as unsatisfactory, but not because the problem is too uncertain for structure. The root cause they identified was an overemphasis on generic individual project selection, without connection to the organisation’s specific strategy. The problem is not too much rigour. It is the wrong kind of rigour, applied at the wrong level.
McKinsey has consistently found that while more than 80% of executives say innovation is among their top three priorities, fewer than 10% are satisfied with their organisation’s innovation performance. That gap has persisted for nearly a decade. BCG’s 2024 Most Innovative Companies report tells the same story at scale. The report surveyed more than 1,000 senior innovation executives globally and found that the readiness gap is widening. In 2022, 20% of companies qualified as “innovation ready” on BCG’s maturity benchmark. By 2024, that figure had fallen to 3%, even as 83% of companies ranked innovation as a top-three priority. Only 12% of executives reported a strong link between their business and innovation strategies. BCG described the emerging pattern as “zombie” innovation systems: organisations going through the motions without a clear strategy to focus their efforts. The parallel with Issue 8’s zombie pilots is not coincidental. The same structural absence, no one with the authority or mandate to make hard decisions, produces the same outcome at every level: activity without accountability.
McKinsey’s late-2024 survey of over 1,000 C-level executives across 99 countries reinforced this. Most organisations struggle to make trade-offs between short- and longer-term initiatives or among projects with different risk profiles. That is not a description of organisations with too much structure. It is a description of organisations with no portfolio-level decision-making at all.
“Our organisation is too complex for a single portfolio view.”
Complexity is not an exemption from portfolio logic. It is the reason you need it.
Large organisations with multiple business units, geographies, and innovation models will never have a simple portfolio. But the alternative to a single portfolio view is not to have no view; it is multiple uncoordinated views that no one can reconcile. Without a portfolio-level view, the organisation cannot see concentration risk, identify duplication, or make trade-offs across the set.
The operational challenge is real. Constructing a portfolio view across a complex organisation requires shared definitions, consistent categorisation, and a governance layer with authority over the whole. That is difficult. But the difficulty of construction is not an argument against the necessity. It is a description of the work that needs to be done.
The Real Cost of the Illusion
The danger of the portfolio illusion is not just that it is inaccurate. It creates false confidence.
An organisation that knows it has a collection of ad hoc innovation projects at least understands what it is working with. The constraints are visible. The lack of coordination is acknowledged.
An organisation that believes it has a portfolio may assume it is diversified, even though it may be concentrated. It assumes it has strategic coverage when it may have gaps. It assumes someone is managing the set, even though no one has that authority.
The portfolio’s language creates a governance assumption that the structure does not support.
This is how organisations find themselves with twenty innovation initiatives and no strategic options. Activity is high. The slide looks full. But nothing in the collection is constructed to deliver portfolio-level returns, because no one is managing it at that level.
What a Portfolio Actually Means
A portfolio is a deliberately constructed set of strategic options, scoped by mandate, composed across instruments, governed as a whole, measured against return, and connected to strategy.
A portfolio you never rebalance is just a list you approved. And collections do not compound.
References & Further Reading
Si, H., Kavadias, S. and Loch, C. (2022) ‘Managing Innovation Portfolios: From Project Selection to Portfolio Design’, Production and Operations Management.
Available at: https://onlinelibrary.wiley.com/doi/10.1111/poms.13860
Boston Consulting Group (2024). Innovation Systems Need a Reboot: The 2024 Most Innovative Companies Report.
Available at: https://www.bcg.com/publications/2024/innovation-systems-need-a-reboot
McKinsey & Company (2025) ‘Investing in Innovation: Three Ways to Do More with Less’, McKinsey Global Survey on Innovation.
Available at: https://www.mckinsey.com/capabilities/strategy-and-corporate-finance/our-insights/investing-in-innovation-three-ways-to-do-more-with-less
This article is based on the author's professional experience and interpretation of publicly available information. It is provided for general informational purposes only and does not constitute advice. Any views expressed are the author's own and do not refer to any specific organisation, program, or individual.
This is Issue 9 of Put Ideas To Work, a biweekly newsletter currently discussing corporate venturing under constraint. If this issue was useful, consider sharing it with someone who manages an innovation function. The diagnostic works best when it prompts a conversation, not just a self-assessment.
Issue 8 explored stop authority and why corporate pilots rarely get killed. Issue 9 extends that logic to the portfolio level: stop authority is necessary but insufficient if no one has authority over the set.