This was originally published on Fast Company.

Companies don’t need to act like startups—they need to know which problems to build internally and which to solve externally.

In business, it sometimes feels like the grass is greener on the other side. Corporations envy the agility, learning capabilities, and speed of startups. Startups pine for the capital-efficient execution and established resources, assets, and capabilities that corporations possess.

Recently, in an attempt to have the best of both worlds, some corporations have warmed to the concept of venture building as a tool. In fact, according to recent research, more than 20% of Forbes 500 companies are running venture building programs and internally launching what they call “startups.”

But here’s the thing—an internal “startup” is not actually a startup. It’s just an extension of the existing business. This is not to say corporations can’t start new businesses internally, because they can—that’s how they grow—but creating an internal new business is an aspect of regular operations.

While organizations continue to pursue the illusion of “corporate startups,” true innovation requires acknowledging that internal ventures operate under fundamentally different constraints than independent startups, demanding distinct approaches to incentives, governance, and strategic timing rather than superficially mimicking startup methodologies. Corporations that designate a team to build a “startup” within the corporation, thinking that venture will operate like an actual startup, are going to face frustration and, likely, failure.

The Corporate Conundrum of Building a New Business

For corporations, the instruction to operate like a startup is not the best path forward, since most startups are unsuccessful. Corporations don’t want to fail. A crucial but simple fact that is often overlooked when pursuing innovation: Corporations and startups are good at fundamentally different things. Corporations must play to their strengths.

The incentive systems in place within corporations prevent them from learning as startups do, and the type of talent one finds in corporations is often very different from that in startups. Decision-making also differs. Corporations (rightfully) decide via a group of stakeholders because they’re big and need to avoid risk and be efficient with capital, but that means that one “no” vote can veto a project and torpedo innovation plans.

None of that works when you’re building something that pushes the boundaries of what the corporation has historically done. Ventures that are radically new and different need to be built far from the constraints of the corporate structure.

So, here’s the question: When do you build within the corporate structure, and when do you build externally? It really comes down to determining the type of problem that you want to solve: Is it an execution problem, or is it a learning problem?

Corporations excel at solving execution problems, problems that you look at and say, “We know what the issue is and how to solve it. The only thing left to do is act.” When it comes to execution problems, a corporation will always have a leg up over a startup.

However, if the problem at hand requires learning to solve—as in, you understand the issue but don’t have a clear solution—an external startup is the way to go. Learning is decidedly within the startup wheelhouse. They’re not efficient at execution, but if you’re not sure how your ideas are going to play out, a startup is the perfect vehicle to run the experiments necessary to solve the problem.

The Path Forward: Stop Pretending

It’s time to abandon the misleading language of a “corporate startup” and start calling internal work what it is: new business development. This honesty will enable clearer thinking about what corporations can realistically achieve internally versus what requires a different approach.

The real opportunity lies in partnering with external venture builders who can handle the learning problems corporations can’t solve, while the corporation contributes with its strengths: industry insights, capital, customers, and exit pathways. A partnership like this leverages the unique strengths of both parties, resulting in an outcome that truly marries the best of both worlds.

Pursuing a “corporate startup” sets teams up for failure. True success comes from matching problems to capabilities—building internally when you know the solution, and externally when you need to discover it.