Where do good ideas come from? Perhaps we’re asking the wrong question.
Editor’s note: This is an ongoing Research Money series that started in May 2026, with articles by Dr. Andrew Maxwell, the Bergeron Chair in Technology Entrepreneurship in the Lassonde School of Engineering at York University. Every week – and occasionally every other week – we’ll present a new article by “Innovation Doctor” Maxwell, in a series with wide-ranging topics and incisive themes on myriad aspects of innovation.
I’ve been preparing a presentation for a space incubation workshop, where I’ll be discussing commercialization pathways, from research to market, including incubation, partnerships and strategies for creating and scaling new space technology ventures.
It’s an interesting challenge, particularly given the extraordinary technological capabilities emerging across the space sector. Advances in materials, sensors, robotics, communications, propulsion and Earth observation are opening up remarkable possibilities, supported by sophisticated research infrastructure and a growing entrepreneurial ecosystem.
But as I’ve been preparing, I’ve found myself returning to a rather different question. Where do good entrepreneurial opportunities actually come from? We tend to have a fairly conventional answer. Researchers develop something new, recognize its commercial potential, establish a venture, raise investment and begin the difficult journey of taking the technology to market.
It’s certainly one pathway, and an important one. But is it the only one we should be encouraging?
Our fascination with invention may actually be limiting our understanding of where entrepreneurial opportunities come from. We invest enormous resources in developing technologies and then try to work out what to do with them. Yet opportunities may be sitting in front of us in the form of unresolved problems, changing circumstances, underutilized capabilities or solutions developed in completely different industries.
Perhaps we should be spending more time understanding how to recognize these opportunities, how to create something valuable from them, and how to act quickly enough to learn whether they’re worth pursuing.
In my research and teaching, I’ve increasingly come to think about entrepreneurship as a process that we can understand, experiment with and improve. And for this presentation, I’ve found it useful to distinguish three related activities: recognition, creation and action. The distinction matters because identifying a possibility isn’t the same as creating a viable opportunity, and neither necessarily tells us what to do next. More importantly, these activities are not simply consecutive stages. They influence one another, often changing our understanding of the opportunity itself.
Recognition: Why must we always start with a technology?
One of the recurring challenges I encounter when working with researchers and entrepreneurs is that they converge on an idea far too quickly. Someone has developed an interesting technology and immediately starts looking for its most obvious commercial application. Or they identify a problem and begin designing a solution before properly understanding its causes. In both cases, they may be overlooking more interesting opportunities because they’ve already decided what they’re looking for.
Technology commercialization programs often reinforce this behaviour. They begin with an invention, assess its commercial potential, protect the intellectual property, develop a business plan and support the creation of a venture. There’s nothing inherently wrong with this approach. But it starts with a particular assumption: that the entrepreneurial journey begins with a technology. Why should that be our starting assumption?
What if we started with an unresolved industry problem and looked for existing technologies that could address it? What if we began with a changing market or regulatory environment and explored the new possibilities it creates? What if we examined expensive, underutilized facilities and asked how they might become the foundation of a new business? Or what if we looked at successful solutions in other industries and considered how they might be transferred into space applications?
These are not simply alternative ways of commercializing inventions. They are different pathways to creating new ventures. And if we’re serious about increasing the number and quality of space technology ventures, shouldn’t our incubation programs support all of them?
In my work on Innovation Cartography, I emphasize the importance of divergence before convergence. Before selecting an opportunity, we should deliberately expand the range of possibilities we consider. I find four starting points particularly useful.
Start with what we already have. Research institutions and businesses possess extraordinary resources, including technologies, specialized equipment, expertise, data, facilities and relationships. Yet we tend to define these resources by their original purpose. A wind tunnel is for aerodynamic testing, a satellite imaging system is for Earth observation, and a specialized material was developed for a particular engineering application. But what if we examined their underlying capabilities rather than their intended uses?
A wind tunnel, for example, may serve potential customers beyond its original research community. A material developed for spacecraft may have valuable applications in transportation, manufacturing or health care. An underutilized testing facility might become the basis for a specialized service venture. None of these opportunities necessarily requires a new invention, although each may require considerable innovation to turn the underlying capability into something useful and commercially viable.
Start with a problem.
Not necessarily the spectacular challenges that attract major research programs, but also the persistent frustrations, inefficiencies and workarounds that organizations have learned to tolerate. Space operations involve complex testing, certification, integration, procurement and supply chains. These activities may contain problems that have become so familiar that people no longer question them.
The opportunity may involve applying an existing technology, combining several capabilities or redesigning a process. The important starting point is understanding the problem, its consequences and why current approaches remain inadequate. This is also where techniques such as Jobs to Be Done and customer journey mapping become valuable: they help us understand what organizations are actually trying to accomplish, rather than simply asking what new technology they would like to purchase.
Start with what’s changed.
Many entrepreneurial opportunities emerge because the assumptions underpinning an industry are no longer valid. A technology becomes cheaper, regulations change, new capabilities become accessible, supply chains are disrupted or customer expectations evolve.
Consider the falling cost of accessing space, advances in artificial intelligence or developments in additive manufacturing. The opportunities aren’t necessarily confined to the new technologies themselves. They may emerge because existing activities can now be performed differently, because previously unattractive business models have become viable, or because organizations can address problems that were previously too expensive or complex to solve.
Start by looking somewhere else.
Industries develop specialized knowledge, practices and approaches to solving problems, but this expertise can also create blind spots. A technique that is routine in one industry may be unfamiliar in another. Manufacturing methods developed for automotive applications might be adapted for space systems, while technologies created for space may find attractive applications in agriculture, transportation or environmental monitoring.
Some of the most interesting innovations arise from connecting existing capabilities rather than inventing something entirely new. These four starting points are not mutually exclusive. In fact, they become particularly powerful when used together. A new technological capability may allow us to repurpose an existing resource, address an overlooked problem or transfer a solution between industries.
The important lesson is that opportunity recognition is a capability we can develop systematically, rather than simply a talent possessed by unusually creative individuals. And perhaps our incubators should be teaching this capability before asking entrepreneurs to commit to a particular venture idea.
Creation: Recognizing a possibility doesn’t mean we have an opportunity
Recognizing possibilities is only the beginning. The next challenge is turning them into opportunities that can create meaningful value. This distinction is particularly important in technology entrepreneurship because we often confuse technological capability with commercial potential. A technology may offer impressive performance improvements, but unless those improvements matter to someone, and can be delivered in a way they are motivated and able to adopt, they may have little commercial value.
We also tend to talk about entrepreneurial opportunities as though they already exist, waiting to be discovered. Sometimes they do. But in many cases, opportunities emerge through interaction between researchers, potential users, partners and other stakeholders. The technology may change, the application may evolve, and the business model may be developed collaboratively.
Consider a team working on a new sensor technology. The researchers understand its performance characteristics, but there may be several potential applications. A satellite manufacturer may care about weight and power consumption, an industrial customer about reliability and integration, and an environmental monitoring organization about sensitivity and accuracy.
Rather than selecting an application and simply trying to sell the technology, the team could work with potential users to understand how different capabilities might create value. These interactions may reveal requirements the researchers hadn’t considered, applications the users hadn’t imagined, or complementary technologies and partners needed to deliver a complete solution. What emerges is not simply a better understanding of an existing market. It may be an opportunity that neither the researchers nor the potential users could have identified independently.
This is why I place such importance on co-creation. Innovation is not necessarily a one-way process in which researchers develop technologies and subsequently transfer them to users. It can be a two-way process in which both parties learn, challenge assumptions and shape the eventual solution.
Understanding adoption changes what we create
There’s another important dimension to this discussion. It’s easy to identify benefits that a new technology might deliver. It’s much harder to understand what would motivate an organization to adopt it. A new solution may be technically superior, yet require changes to established processes, employee training, regulatory approval and integration with existing systems or substantial financial commitments. It may also threaten established relationships, responsibilities or organizational interests. And the person who benefits from the technology may not be the person who decides whether to adopt it.
This is where I think our conventional approach to technology commercialization often falls short. We focus on demonstrating technical performance and identifying potential markets without investing sufficient effort in understanding the behavioural and organizational factors that determine adoption. We need to distinguish between the benefits a solution could provide and the reasons a particular organization would actually change its behaviour.
What problem is sufficiently important to motivate action? Who benefits, who pays, and who bears the costs and risks of change? What alternatives are available? What would have to happen before the organization felt confident enough to proceed? These aren’t questions to leave until after the technology has been developed. They can fundamentally shape the opportunity itself.
Sometimes the answer may be to simplify the technology, integrate it with an existing system, partner with an established supplier or develop a service rather than a product. Sometimes a different business model can overcome an adoption barrier.
Imagine a university with expensive specialized space testing facilities. Potential customers might value access but lack the resources or sufficient demand to purchase comparable equipment. A conventional product-focused analysis might suggest a limited market. But providing testing as a service, sharing facilities or developing a partnership-based model could create an entirely different commercial opportunity. Nothing about the underlying technology has changed. What has changed is the way value can be delivered and the barriers potential users must overcome. In other words, understanding adoption doesn’t simply help us commercialize an opportunity. It helps us create a better one.
This also raises an important question for incubators and entrepreneurship programs. If opportunities emerge through interaction, shouldn’t we be doing much more to bring researchers, entrepreneurs, potential users and industry partners together before ventures commit to particular products, markets or business models?
Action: Speed matters, but what are we trying to accelerate?
Recognition and creation are important, but neither tells us whether an opportunity is worth pursuing. For that, we need to act. And I would argue that speed matters enormously. Not simply because being first to market can be valuable, but because slow entrepreneurial processes are expensive, consume scarce resources and delay the feedback that helps us make better decisions.
Space technology ventures face particular challenges. Development costs can be high, testing and qualification demanding, procurement cycles lengthy and potential customers understandably cautious about reliability and risk. These conditions make it even more important to distinguish between the activities that genuinely require time and those that can be accelerated through better processes.
We cannot eliminate necessary engineering or qualification work. But do we really need to spend months developing business plans before speaking to potential customers? Must we complete a sophisticated prototype before understanding whether an organization would be willing to adopt it? Should we commit significant resources to a particular application before exploring alternatives? I suspect that many commercialization delays have less to do with the underlying technology than with how we organize the journey from possibility to adoption.
Speed of learning, not simply speed of development
Our conventional approach to commercialization often emphasizes developing business plans, securing intellectual property, raising capital and achieving technical milestones. These activities may be important, but they can encourage entrepreneurs to behave as though the opportunity is already understood and the main challenge is execution.
Unfortunately, at the beginning of an entrepreneurial journey, we rarely know enough to justify that confidence. We may have limited evidence about user needs, adoption motivations, integration requirements, willingness to pay or the capabilities needed to deliver a viable solution.
This is where I think we should borrow more from experimentation and quality management. Rather than treating our initial assumptions as facts, we should identify what we believe, examine the evidence supporting those beliefs and design activities that allow us to learn as quickly and economically as possible. I frequently encourage entrepreneurs to distinguish between what they know, what they think they know and what they still need to discover. Just as importantly, I ask what evidence would cause them to change their minds.
Suppose a space technology venture has identified three possible applications. One appears to offer a very large market, but potential customers are difficult to access, development requirements are uncertain and adoption could take years. Another has a smaller potential market, but an interested organization is willing to test the technology in a real operating environment.
Which should the venture pursue? There isn’t a universal answer, but I would give considerable weight to the second opportunity because access to an early user can provide something extraordinarily valuable: rapid, credible feedback about whether the technology delivers meaningful value in practice. A real user can help establish which benefits matter, reveal operational constraints, identify barriers to adoption and provide evidence that makes subsequent decisions more informed.
That doesn’t mean we should always choose the smallest or easiest market. But it does suggest that the speed with which we can reach an early user and generate meaningful evidence should be an important factor in selecting opportunities. Indeed, sometimes the fastest pathway to a successful venture is not the one with the shortest technical development schedule, but the one that allows us to learn most rapidly about the conditions required for adoption.
What should we measure?
This also changes how we think about progress. A traditional development plan might identify milestones such as completing a prototype, filing a patent, raising investment or launching a company. These outputs may be useful, but none necessarily demonstrates that we are closer to creating something people will adopt.
I would rather ask whether we’ve reduced uncertainty about the opportunity. Have we established that the problem matters to potential users? Do we understand their motivation to adopt? Have we identified the operational and organizational barriers? Have we tested whether the benefits justify the costs and risks of change? Have we identified a realistic pathway to an early adopter?
An experiment might establish that a technical capability performs as expected, but it might also reveal that the customer is unwilling to change established processes, that the benefits are insufficient to justify integration costs, or that another application is more promising. These findings may be disappointing if we define progress purely through predetermined milestones. But from a learning perspective, they can be extremely valuable.
One of the things that concerns me about conventional entrepreneurship programs is our tendency to celebrate activity and outputs without examining whether they represent meaningful progress. Forming a company, developing a prototype or raising investment can create the appearance of momentum, even when fundamental uncertainties about adoption remain unresolved. A faster, more effective process should help us discover these problems earlier, when we still have the flexibility and resources to respond.
This is also why we need to rethink our attitude towards failure. An experiment that disproves an important assumption may prevent months of wasted development effort. A conversation that reveals an unattractive market may redirect the team towards a better opportunity. And a failed attempt to integrate a technology may uncover a new problem worth solving. The goal isn’t to eliminate failure. It’s to reduce the cost of learning and improve the quality of our decisions.
Recognition, creation and action are not a straight line
I’ve described recognition, creation and action separately because doing so helps us understand the different activities involved. But successful entrepreneurship rarely follows such a neat sequence. An experiment may reveal an entirely new application. A conversation with a potential user may change our understanding of the original problem. A new partnership may provide access to capabilities that make a previously unattractive opportunity feasible. And a change in the business model may alter which customers we should pursue.
Action generates information, which changes what we recognize. Recognition leads us to reshape the opportunity, which suggests further actions and experiments. This is an iterative process, not a linear pipeline from invention to commercialization. And importantly, the earlier we engage potential users and begin learning, the faster these feedback loops can operate.
This is where I believe innovation management as a discipline can make an enormous contribution. We have become very good at supporting the development of technologies and increasingly sophisticated in providing entrepreneurship training, mentoring and investment. But we have been less systematic about understanding and improving the processes through which those capabilities are translated into value. A process perspective encourages us to examine not just the quality of the ideas entering the system, but also the quality, speed and effectiveness of the activities that transform those ideas into outcomes.
For the space sector, this matters because creating new ventures involves much more than moving research discoveries into commercial markets. It also involves identifying overlooked problems, repurposing existing capabilities, connecting different industries, engaging first users and developing business models that make adoption possible. These different pathways may require different partners, different resources and different forms of support.
An incubation program designed primarily around developing research-based spin-offs may overlook opportunities that begin with problems, underutilized infrastructure or new applications for established technologies. And one designed primarily around traditional venture milestones may encourage entrepreneurs to spend time and resources demonstrating progress without sufficiently testing whether their solutions will actually be adopted.
Perhaps we should rethink what an incubator is designed to do
As I prepare for the conference, I’ve become increasingly convinced that the most interesting question isn’t simply how we can create more space technology ventures. It’s how we can create better opportunities, learn faster and improve the likelihood that the ventures we support will deliver meaningful value. That suggests several changes in emphasis.
We should broaden the starting points for entrepreneurship, recognizing that valuable ventures can emerge from problems, resources, changes and connections, not simply new technologies. We should create more opportunities for researchers, entrepreneurs, industry partners and potential users to work together before committing to particular solutions.
We should help ventures understand adoption from the beginning, including the motivations, constraints and organizational realities that determine whether someone will actually use what we’re developing. And we should measure progress not simply through activities and outputs, but through the evidence generated, the uncertainties reduced and the pathways to adoption established. Perhaps most importantly, we should focus on improving the speed and quality of the overall process, rather than simply accelerating individual activities.
There is little benefit in developing a technology more quickly if we discover too late that nobody is willing to adopt it. Equally, there may be enormous value in spending a few weeks exploring alternative applications or engaging potential users if that helps us avoid years pursuing an unattractive opportunity.
Faster innovation isn’t simply about doing things more quickly. It’s about learning sooner what is worth doing, and creating the conditions that allow valuable solutions to be adopted.
Which brings me back to my original question.
Where do good entrepreneurial ideas come from?
Perhaps we’ve been looking in the wrong places, or at least looking through too narrow a lens.
We don’t necessarily need more inventions to create more successful space technology ventures. We also need to become much better at recognizing opportunities in the capabilities we already possess, creating new possibilities through collaboration, and taking action that generates the evidence needed to move forward.
And if we can improve those processes, perhaps we’ll discover that some of the most exciting entrepreneurial opportunities have been sitting in front of us all along.
R$
| Organizations: | |
| People: | |
| Topics: |